Clear and high-quality underwater images are essential for marine applications, including autonomous navigation, ecological monitoring, and infrastructure inspection. However, underwater images typically suffer from severe colour distortion, low contrast, and diminished structural visibility due to wavelength-dependent attenuation, scattering, and uneven illumination conditions. Recent deep learning-based underwater image enhancement (UIE) methods primarily adopt end-to-end frameworks, directly regressing enhanced images from degraded inputs. While these approaches have achieved significant progress, they often lack explicit modeling of the degradation process, leading to limited interpretability and suboptimal recovery of fine-grained details. To address these limitations, we propose DRM-Net, an explicit residual learning framework for UIE. Rather than estimating the enhanced image directly, DRM-Net first predicts a pixel-wise Degradation Residual Map (DRM) in the perceptually uniform CIELab colour space. This map explicitly quantifies local colour, contrast, and structural degradations, thereby enabling the network to precisely reconstruct missing visual information. Furthermore, we design a lightweight Subaquatic Multi-Scale Context Fusion module, which utilizes parallel atrous convolutions with softmax-weighted feature aggregation, significantly enhancing robustness against spatially heterogeneous scattering. Trained jointly with pixel-wise DRM and VGG-based perceptual losses, DRM-Net achieves superior colour fidelity, perceptual realism, and structural detail recovery. Comprehensive experiments conducted on multiple benchmarks demonstrate that our proposed approach attains competitive quantitative results and superior qualitative visual performance compared to state-of-the-art UIE methods, while maintaining low computational overhead, making it particularly suitable for resource-constrained underwater robotic systems.

Differentially Private Stochastic Gradient Descent (DPSGD) is widely used to protect sensitive data during the training of machine learning models, but its privacy guarantee often comes at a large cost of model performance due to the lack of tight theoretical bounds quantifying privacy loss. While recent efforts have achieved more accurate privacy guarantees, they still impose some assumptions prohibited from practical applications, such as convexity and complex parameter requirements, and rarely investigate in-depth the impact of privacy mechanisms on the model's utility. In this paper, we provide a rigorous privacy characterization for DPSGD with general L-smooth and non-convex loss functions, revealing converged privacy loss with iteration in bounded-domain cases. Specifically, we track the privacy loss over multiple iterations, leveraging the noisy smooth-reduction property, and further establish comprehensive convergence analysis in different scenarios. In particular, we show that for DPSGD with a bounded domain, (i) the privacy loss can still converge without the convexity assumption, (ii) a smaller bounded diameter can improve both privacy and utility simultaneously under certain conditions, and (iii) the attainable big-O order of the privacy utility trade-off for DPSGD with gradient clipping (DPSGD-GC) and for DPSGD-GC with bounded domain (DPSGD-DC) and strongly convex population risk function, respectively. Experiments via membership inference attack (MIA) in a practical setting validate insights gained from the theoretical results.

Continuous cardiac monitoring during sleep is vital for detecting silent arrhythmia and other nocturnal cardiac events. While electrocardiogram (ECG) is the clinical gold standard, its reliance on electrodes and physical contact makes it intrusive for daily long-term use. Millimeter-wave (mmWave) radar offers a compelling non-contact alternative by capturing cardiac-induced chest-wall micro-vibrations. Existing radar-to-ECG methods often rely on direct waveform regression, assuming posture-stable mappings that break under natural sleep movements and obscure true cardiac rhythms. Inspired by the modality-invariant perception observed in speech and vision, we introduce mmJEPA-ECG, a physiology-guided framework for reconstructing clinical ECGs by anchoring radar sensing to invariant cardiac dynamics. It addresses two fundamental challenges: (i) disentangling robust cardiac representations from posture-induced artifacts, and (ii) generalizing ECG reconstruction across individuals under signal ambiguity. To address these challenges, Physiology-Oriented Self-Supervised Pretraining builds on a Joint Embedding Predictive Architecture (JEPA) with domain-informed masking and heart rate consistency to extract posture-robust cardiac embeddings. Conditional Diffusion-based ECG Reconstruction then generates personalized ECG waveforms through a hierarchical conditional diffusion process by spectral fidelity and denoising constraints. Extensive experiments on both public and self-collected multi-subject datasets demonstrate that our method outperforms state-of-the-art across waveform and rhythm metrics, halving R-R peak errors even under posture shifts and arrhythmic conditions.

Severe acoustic degradation is often caused by overlapping noise, disfluencies, and environmental distortions.This phenomenon results in the dissolution of linguistic structures and the generation of unreliable ASR outputs.Inspired by human speech comprehension, we propose Speech-MLM, a novel multimodal framework that reframes ASR as semantics-guided speech reconstruction.This perspective introduces three core challenges: (C1) collapse of linguistic structure under acoustic degradation, (C2) semantic ambiguity under noise, and (C3) misalignment across modalities.To address these issues, we propose Speech-MLM, a multimodal ASR framework that integrates speech, spectrogram-derived visual cues, and textual variants to enhance robustness.It consists of: (i) Cognitive Structure Extractor that recovers prosodic structure from visualized acoustic features, (ii) Semantic Weaver that learns semantic equivalence across varied textual forms, and (iii) Retrieval-Guided Fusion Learner that unifies modalities within a shared semantic space.Experiments on multiple real-world noisy datasets demonstrate that Speech-MLM achieves an average 38.85% reduction in WER, while also attaining 98.71% BERTScore and 96.7% USE, over advanced baselines, demonstrating substantial gains in semantic robustness and generalization across domains.

Human–robot creative collaboration is often constrained by command–response paradigms that position robots as tools rather than partners. While expressive robotics has shown social values, its role and behaviors in shaping creative partnerships with humans remains underexplored. Therefore, we investigate how robots’ expressive behaviors influence co-creative engagement. In a formative study with 5 participants, we identified design insights for users to perceive a robot arm’s expressive behaviors. We then implemented these expressive behaviors and conducted a within-subject study with 18 participants, comparing functional-only and expressive conditions in figure drawing tasks. Results showed that expressive behaviors significantly enhanced human-robot collaboration where they shifted from viewing the robot as a tool to a partner with a stronger emotional connection and collaborative satisfaction. Our contributions include empirical evidence of partnership transformation and design insights for facilitating human-robot creative collaboration.

Timely nowcasting is critical for public safety during fast-evolving storms, where even short delays can trigger cascading failures—as in the October 2024 Spain flash flood that claimed over 90 lives within minutes. While radar offers reliable real-time sensing of atmospheric structure, models that collapse 3D volumes into 2D slices inevitably discard vertical information essential for capturing storm growth, phase transitions, and collapse. We introduce StormMind, a physically grounded framework that forecasts convective evolution by modeling causal interactions across stratified atmospheric layers. StormMind addresses two fundamental challenges:(1) the nonlinear, asynchronous coupling between low-, mid-, and high-level processes; and (2) reflectivity uncertainty, where storms with distinct vertical structures may appear deceptively similar on radar, masking their true phase and intensity. To tackle these issues, StormMind designs: i) a Convection Dynamics Extractor that models storm evolution from two complementary perspectives—horizontal morphology, capturing the spatial organization of physical processes within individual atmospheric layers, and vertical coupling, modeling energy exchanges across layers; and ii) a Convection Manifestation Reconstructor that adaptively fuses intra- and inter-layer signals, conditioned on the evolving storm state, to infer phase transitions (e.g., initiation, intensification, dissipation). Evaluated on the large-scale 3D-NEXRAD dataset (2020–2022, U.S.), StormMind outperforms strong baselines, achieving a 14.71% gain in CSI40. In real-world deployment with the Guangzhou Meteorological Bureau (Mar–May 2025), it improves CSI40 by 9.39% and boosts early-warning accuracy (98.33%)

Continuous and precise bladder volume monitoring is essential for patients with lower urinary tract dysfunction (LUTD) to support timely voiding and effective rehabilitation management. However, existing devices lack skin conformity and cannot support reliable "stick-once, long-term daily use" in real-world scenarios. We present BladderSense, a skin-conforming wireless wearable system featuring an X-shaped flexible phased-array ultrasound probe. Its development faces three key challenges: preserving beam focusing under skin deformation, maintaining accurate volume estimation despite bladder position shifts, and enabling low-power wireless transmission despite large raw data volume. To address these obstacles, we employ a suitable-frequency deep-focus design to stabilize beam quality, introduce a dual-orthogonal array with a shared geometric anchor and develop a coordinate-encoded deep learning (DL) model, together enabling bladder tracking and end-to-end volume estimation. An envelope-extraction-based compression scheme further enables Bluetooth Low Energy (BLE) transmission, supporting continuous monitoring with intermittent (1-minute) sensing. Experiments with 10 participants show that BladderSense provides accurate, robust bladder volume estimation across bladder changes, posture transitions, and dynamic daily activities, realizing dependable "stick-once, long-term monitoring" for LUTD patients.

In always-on HAR deployments, model accuracy erodes silently as domain shift accumulates over time. Addressing this challenge requires moving beyond one-off updates toward instance-driven adaptation from streaming data. However, continuous adaptation exposes a fundamental tension: systems must selectively learn from informative instances while actively forgetting obsolete ones under long-term, non-stationary drift. To address them, we propose CODA, a continuous online adaptation framework for mobile sensing. CODA introduces two synergistic components: (i) Cache-based Selective Assimilation, which prioritizes informative instances likely to enhance system performance under sparse supervision, and (ii) an Adaptive Temporal Retention Strategy, which enables the system to gradually forget obsolete instances as sensing conditions evolve. By treating adaptation as a principled cache evolution rather than parameter-heavy retraining, CODA maintains high accuracy without model reconfiguration. We conduct extensive evaluations on four heterogeneous datasets spanning phone, watch, and multi-sensor configurations. Results demonstrate that CODA consistently outperforms one-off adaptation under non-stationary drift, remains robust against imperfect feedback, and incurs negligible on-device latency.

Continuous ECG monitoring is vital for early detection of arrhythmias and other cardiac abnormalities—especially during sleep, when symptoms often go unnoticed—yet existing solutions remain expensive, obtrusive, and impractical for long-term daily use. Ballistocardiography (BCG)—a passive, contactless modality that captures cardiac-induced body motion—offers a compelling alternative. However, prior efforts treat ECG reconstruction as waveform regression, leading to overfitting to individual-specific or posture-dependent artifacts. Inspired by the fact that ECG and BCG reflect parallel structures in cardiac event sequences (e.g., P/QRS/T-waves vs. I/J-waves), we design Med2ECG: a structure-aligned system that reconstructs ECG from high-fidelity BCG by explicitly aligning their latent physiological events. Med2ECG incorporates three key designs: (i) Multi-Scale Feature Extractor captures hierarchical temporal dynamics, preserving clinically relevant fine-grained features; (ii) Shared-Personalized Experts employs a Mixture-of-Experts (MoE) to adaptively disentangle signal variations due to individual and environmental factors; (iii) Medical-Informed Strategies introduces a diagnostic-driven multi-objective loss, integrating structural alignment, morphological fidelity, and landmark-aware supervision to preserve clinically critical intervals. Experiments across public and self-collected in-hospital datasets (20 healthy individuals and 10 patients with diverse cardiovascular conditions), Med2ECG achieves > 0.92 Pearson correlation, < 15% amplitude error, and precise PR/QRS/QT/RR interval estimation within 5–20 ms—demonstrating strong generalization across subjects, postures, and environments.

Abstract Multi-source synsemantic domain generalization (MSSDG) for multi-task remote physiological measurement seeks to enhance the generalizability of these metrics and attracts increasing attention. However, challenges like partial labeling and environmental noise may disrupt task-specific accuracy. Meanwhile, given that real-time adaptation is necessary for personalized products, the test-time personalized adaptation (TTPA) after MSSDG is also worth exploring, while the gap between previous generalization and personalization methods is significant and hard to fuse. Thus, we proposed a unified framework for MSSD G and TTP A employing P riors ( GAP ) in biometrics and remote photoplethysmography (rPPG). We first disentangled information from face videos into invariant semantics, individual bias, and noise. Then, multiple modules incorporating priors and our observations were applied in different stages and for different facial information. Then, based on the different principles of achieving generalization and personalization, our framework could simultaneously address MSSDG and TTPA under multi-task remote physiological estimation with minimal adjustments. We expanded the MSSDG benchmark to the TTPA protocol on six publicly available datasets and introduced a new real-world driving dataset with complete labeling. Extensive experiments that validated our approach, and the codes along with the new dataset are in https://github.com/WJULYW/GAP .

In extended reality (XR) virtual interaction environments, both flexible user interaction and reliable authentication are essential. However, current wearable systems relying on single-sensor data suffer from insufficient data dimensionality, often failing to simultaneously achieve fine-grained gesture recognition and continuous authentication. Multi-sensor systems, while more capable, tend to be bulky, comfort-compromising, and computationally demanding for edge deployment. Furthermore, inefficient parallel processing and the decoupled pipeline for gesture and authentication tasks degrade computational efficiency, ultimately hindering practical adoption. To overcome these limitations, we propose M-Ring, a lightweight ring-based system equipped with a compact, optimized sensor layout that synchronously captures impedance and IMU signals. By fusing complementary physiological and behavioral cues, M-Ring enables accurate hand-state perception. We design an adaptive-scale convolutional network with a parameter-separated and progressively routed architecture (PLE), which effectively captures intrinsic correlations within multimodal data and enhances synergy between gesture recognition and user authentication, all while minimizing computational overhead. For fine-grained authentication, we propose a dual-factor framework that integrates Siamese networks with contrastive learning to enhance inter-user discriminability. By jointly modeling physiological activations and kinematic patterns, our approach enables both seamless gesture-based interaction and non-intrusive continuous authentication. Experiments demonstrate that M-Ring achieves 97.8% gesture recognition accuracy and 97.5% identity authentication accuracy under complex conditions, while maintaining low enrollment cost, power efficiency, low latency, and strong edge deployment capability.

Autonomous Underwater Vehicles (AUVs) have been becoming excellent platforms in marine environmental monitoring and data collection missions. However, it is challenging to obtain real-time data (e.g., video and high definition images) from AUV swarms, as traditional methods have limitations, such as low-data-rate communication and restricted mobility. To tackle these challenges, in this paper, we exploit complementary strengths of acoustic-optical technology, as well as the mobility of unmanned aerial vehicles (UAVs), and propose a cooperative cross-boundary multi-modal (CCM) communication scheme, in which multiple UAVs carrying acoustic-optical communication nodes serve as mobile sinks for swarms of AUVs. In this scheme, to deal with the imbalanced coverage problem, we design a cooperative movement strategy (CMS) algorithm for multiple UAVs based on the multi-agent reinforcement learning (MARL) approach, combined with particle filtering and task assignment to UAVs for achieving dynamic coverage of AUVs. Furthermore, to achieve high data-rate transmission dealing with optical beam link misalignment problem caused by interference from currents and movement of AUVs, we design an adaptive pointing adjustment (APA) algorithm to construct a stable optical link by controlling the pointing and divergence angles of optical beam for reliable high-speed data transmission. Through extensive simulations using open-source AUV trajectory datasets and ocean current datasets, we demonstrate the effectiveness of the proposed scheme with robust network performance under dynamic maritime conditions. The code is available at https://github.com/SDUST-smartocean/CCM.

The direct cross-boundary communication between Unmanned Aerial Vehicles (UAVs) and Autonomous Underwater Vehicles (AUVs) is a pivotal component in establishing the 6G integrated air-sea-space network, holding significant importance for applications such as marine data collection and maritime collaborative search and rescue. Nevertheless, existing solutions exhibit pronounced deficiencies in path planning efficiency, the coverage range of wireless optical communication, and edge computing load balancing, which result in a high Age of Information (AoI), severely compromising the performance of time-sensitive maritime missions. To address these challenges, this paper proposes a maritime data collection scheme called MECOS, which includes LMAR2P algorithm for UAVs path planning and MAPBal algorithm for UAVs to deal with the load balancing issue. In LMAR2P, a multi-agent deep reinforcement learning (MARL) architecture is adopted, in which we leverage the global understanding capability of large language models (LLMs) to provide state representation for MARL, in order to improve path planning efficiency. Furthermore, to solve the unbalanced computational load problem, we design a kullback-leibler (KL) divergence-based reward correction mechanism and propose a distributed adaptive offloading balancing algorithm MAPBal, which enables resource-aware task allocation to ensure load balancing and reduce data processing latency. The simulation results indicate that the MECOS scheme reduces the AoI by 23.6% and 26.4% during the data collection and data processing phases of maritime missions, respectively. This research provides a viable technical solution for practical applications such as maritime monitoring, demonstrating significant scientific value and promising application prospects. The code is available at https://github.com/SDUST-smartocean/MECOS.

In the digital era, the increasing demand for network traffic necessitates strategic network infrastructure planning. Accurate modeling of traffic demand through cellular traffic generation is crucial for optimizing base station deployment, enhancing network efficiency, and fostering technological innovation. In this paper, we introduce STOUTER, a spatio-temporal diffusion model for cellular traffic generation. STOUTER incorporates noise into traffic data through a forward diffusion process, followed by a reverse reconstruction process to generate realistic cellular traffic. To effectively capture the spatio-temporal patterns inherent in cellular traffic, we pre-train a temporal graph and a base station graph, and design the Spatio-Temporal Feature Fusion Module (STFFM). Leveraging STFFM, we develop STUnet, which estimates noise levels during the reverse denoising process, successfully simulating the spatio-temporal patterns and uncertainty variations in cellular traffic. Extensive experiments conducted on five cellular traffic datasets across two regions demonstrate that STOUTER improves cellular traffic generation by 52.77% in terms of the Jensen-Shannon Divergence (JSD) metric compared to existing models. These results indicate that STOUTER can generate cellular traffic distributions that closely resemble real-world data, providing valuable support for downstream applications.

Radar-based gesture recognition and person identification offer a natural, convenient, and privacy-preserving approach to human-computer interaction. However, most existing research focuses predominantly on learning for a single task, which requires separate models for each task. This separation increases the complexity of the deployment and the computational overhead. To address these challenges, this study introduces a multi-task learning framework that simultaneously performs gesture recognition and person identification using a miniature radar sensor. By leveraging radar's capacity to capture finegrained spectral and spatial motion patterns, the framework incorporates micro-Doppler and range-Doppler processing, alongside a multi-branch architecture to enhance modality-specific feature representation. It enables unified learning of shared and task-specific features within a single network architecture. The proposed model, MTxLSTM, integrates CNN and the recent xLSTM to mitigate task interference, improve generalization, and improve gesture recognition through person-specific nuances while enhancing person identification by leveraging contextual gesture information. Experimental results reveal that MTxLSTM outperforms existing multi-task learning frameworks and stateof- the-art models, achieving 99.21% in gesture recognition and 98.59% in person identification with moderate model complexity and inference speed. This study concurrently executes gesture recognition and person identification using a miniature radar sensor, and marking the first application of xLSTM in radar sensing technology.

Existing AI-based meeting summarization tools have enabled rapid generation of meeting notes, yet their reliability and user controllability remain limited. This paper explores human-AI collaboration for mobile meeting summarization and presents MeetSumAid, a multifunctional system that integrates summarization algorithms with an interactive user interface. The system is designed to support users in understanding, validating, and refining AI-generated summaries through natural interactions and flexible control mechanisms. By enabling real-time inspection, editing, and feedback, MeetSumAid facilitates reliable collaboration between humans and AI in dynamic meeting scenarios. A user study with 20 participants shows that MeetSumAid significantly improves summary quality, generation efficiency, and user-perceived reliability compared with baseline AI summarizers, while reducing cognitive load. Further analysis reveals how different interface components enhance users' engagement and confidence during collaboration. This work provides a practical step toward reliable and user-centered human-AI collaboration in mobile meeting summarization and offers actionable design implications for future intelligent collaborative systems.

Estimating the soluble solids content (SSC) in fruits is essential for meeting consumer expectations, ensuring the quality of processed products, and minimizing food waste. Current methods often rely on destructive sampling, expensive equipment, and complex procedures, which limit their practicality. This paper presents VNiScan-Fruit, a non-invasive, low-cost, and easy-to-use optical sensing system that utilizes visible-near-infrared (Vis-NIR) spectroscopy to estimate SSC by analyzing the interaction of light with fruit tissues to generate characteristic absorption spectra. Our approach diverges from traditional high-precision spectrometers, employing commercial LEDs and photodetectors (PD) to construct the optical sensing unit. We design and implement innovative ring-shaped rubber enclosure, interference elimination algorithms, and reconstruction strategy to extract low-dimensional reflectance spectral features from various fruits. These features are then processed using a specially designed nonlinear regression model, incorporating sucrose values obtained from a commercial refractometer to estimate SSC accurately. We tested VNiScan-Fruit on a range of fruits with diverse peel characteristics, demonstrating its ability to penetrate exocarps and effectively analyze fruits with rough surfaces and delicate tissues. The system achieved normalized mean absolute errors (NMAE) of 8%, imperceptible to consumers in sweetness difference, and remained stable under varying lighting and temperature conditions. Our findings highlight the potential of VNiScan-Fruit as a practical tool for non-destructive fruit quality assessment.

With the rapid expansion of the Internet of Medical Things (IoMT) and cloud computing, ensuring secure communication in e-health systems has become increasingly critical. However, many existing authentication solutions suffer from excessive overhead and security vulnerabilities. To address these challenges, we present a lightweight, high-speed authentication protocol that relies on secure hash functions and XOR operations, facilitating efficient mutual authentication among users, trusted servers, and medical servers while establishing session keys for data exchange. We then rigorously assess our protocol's security against a comprehensive threat model, employing both informal methods and formal analyses, including Real-Or-Random (ROR) model, BAN logic, and automated verification via ProVerif. The results demonstrate that our protocol remains resilient against known attacks and satisfies e-health security standards. Furthermore, a detailed performance comparison reveals that our approach significantly reduces some costs compared to existing schemes, while reinforcing security and privacy protections.

With the rapid development of new smart devices, such as smart home appliances and VR/AR equipment, there is an increasing demand for novel interaction methods. However, many existing interaction methods require external devices, are unintuitive, and demand substantial user learning effort. To fill this gap, we propose ArmPad, a system that leverages the smartwatch's built-in IMU to enablemultidimensional inputon the user's forearm. Methodologically, ArmPad is explicitly designed to address three core research challenges in forearm-based interaction. First, to resolve the inherent feature conflicts between discrete gesture recognition and continuous distance estimation, we propose a multi-task learning framework with a dynamic gating mechanism for cross-task synergy. Second, to tackle the physical limitation of rapid vibration attenuation across the forearm, we introduce a cross-device guidance strategy that incorporates high-fidelity fingertip knowledge during the training phase. Finally, to ensure robust generalization across diverse populations, we develop task-specific data augmentation and a lightweight user registration mechanism to effectively mitigate physiological variances. Experiments on 20 subjects demonstrate that ArmPad achieves an accuracy of 92.51% on nine gestures and a Mean Absolute Error of 1.53$cm$for sliding distance in cross-user settings. Extensive robustness evaluations and case studies further confirm the system's stability and usability under diverse real-world conditions.

LoRa technology plays a vital role in IoT applications by enabling low-power and long-range communication. Since the LoRa signal may transmit hundreds of meters to tens of kilometers under different environments, estimating the path loss of LoRa links can bring great benefits to the network, such as device localization and network planning. Recent years have seen some mechanisms to improve the LoRa link estimation by analyzing the environments based on satellite images. However, they may be inaccurate when we consider the actual network in the three-dimensional (3D) scene. This paper proposes 3DLoRa, the first framework to refine LoRa link estimation through exploiting 3D point clouds. We generate a 3D map by conducting semantic segmentation and voxelization based on the point clouds. We then propose both model-based and learning-based methods to estimate the path loss of a link according to the environments obtained from the 3D map. We test the 3D map generation process based on 3D point clouds in a campus area and show an accuracy of about 97% for environment categorization. We also deploy a LoRa network on the campus and collect data to evaluate the performance of link estimation. Experimental results indicate that 3DLoRa can significantly improve the accuracy of link estimation in various environments compared to previous works. Specifically, the model-based method can reduce the estimation error to about$7dBm$with low computational overhead, while the learning-based method can reduce the estimation error to about$2.5dBm$with the cost of higher computational overhead. These results can serve as a reference for how to deploy link estimation methods in different network situations.

Depression significantly impacts mental health, severely disrupting patients' daily lives. During depressive episodes, individuals may experience symptoms such as excessive guilt, self-harm, and suicidal ideation. Compared to proprietary devices like brain electrode caps, wearable technologies for depression detection have gained attention due to their affordability and portability—enabling real-time monitoring of depressive states. However, challenges such as low-quality data from ubiquitous devices, individual variability, and the complexity of multimodal physiological signal analysis limit model generalizability. To address these issues, we present DepGuard, a novel ubiquitous wearable system for depression assessment based on multimodal physiological signals. DepGuard performs a two-stage detection process: depression recognition and real-time episode monitoring. For depression recognition, we propose an unsupervised domain adaptation method to reduce the domain gap between source and target subjects. For episode monitoring, we employ a few-shot learning strategy to enable personalized modeling. Both approaches enhance cross-subject generalization. Our system achieves 90.75% accuracy in cross-subject depression recognition using 30 unlabeled samples per target subject, and 93.52% accuracy in episode monitoring using 15 labeled samples per class.

The rapid development of the low-altitude economy (LAE) has significantly enhanced the service diversity of the Internet of Things (IoT) networks, necessitating efficient coordination among the unmanned aerial vehicles (UAVs). In this work, we utilize multiple UAVs to provide both wireless data collection (WDC) service and wireless energy transfer (WET) service, and divide the IoT devices into the I-devices that only need the WDC service and the E-devices that only need the WET service, respectively, from the multiple UAVs. Due to their conflicting service demands on the UAVs with limited resources, we formulate the bi-objective optimization problem (BOOP) to minimize the age of information (AoI) for the I-devices and the hungry-level of energy (HoE) for the E-devices at the same time, by jointly optimizing all the UAVs' trajectories and WET decisions over time slots and their WDC decisions over sub-slots. To efficiently solve the complex BOOP, we innovatively transform it into a single-objective optimization problem (SOOP), in which the two conflicting objectives are scalarized via a self-adaptive objective weight. Unlike the conventional approach reliant on fixed and pre-defined objective weights, we optimize the objective weight jointly with other decision variables, enabling automatic adaptation to various network environments without human intervention. However, the proposed SOOP is NP-hard with a large number of decision variables. Accordingly, we propose a new Multi-Agent Adaptive and Hierarchical Deep Reinforcement Learning (MA$^{2}$HDRL) framework, which leverages a central controller (CC) to guide the local training of multiple individual UAV agents. In this framework, each UAV agent employs a two-tier hierarchical DRL model: tier-1 optimizes the trajectory and WET policies over the time slots, while tier-2 optimizes the WDC policy across the sub-slots. Meanwhile, the CC trains the global reward preference for all the UAV agents over the training episodes, to adaptively balance the WDC and WET service demands. Finally, extensive simulation results are conducted to demonstrate the outstanding performance of the proposed MA$^{2}$HDRL approach as compared to state-of-the-art benchmarks.

Logistics platforms increasingly offer real-time door-to-door order pickup services to enhance customer convenience. However, a high volume of unexpected order cancellations negatively impacts both customer experience and logistics profitability. Accurately identifying whether these cancellations stem from customers' decisions or couriers' non-compliant behaviors is essential for implementing targeted operational improvements. Simply interpreting customer-courier dialogues, however, risks being one-sided. The primary challenge lies in adaptively correlating dialogue content with the varying behaviors of couriers. To address this, we developCOCO, a cause identification framework for order cancellation in logistics that consists of: i)Multi-modal features exploration, which analyzes dialogues and couriers' behaviors (both historical and current); ii)Multi-modal features aggregation, which uses a hierarchical attention mechanism to adaptively capture the dynamic correlations within dialogues and behaviors; iii)LLM-enhanced refinement, which leverages Large Language Models to accurately process a large number of unlabeled dialogues, significantly enhancingCOCO's generalization and performance. To further enhanceCOCO's effectiveness, we introduceCOCO+, which incorporates customers' behaviors to better identify customer-related causes. Our extensive evaluation with data collected from JD Logistics demonstratesCOCO+'s exceptional performance, achieving an 12.2% increase in precision and a 9.1% improvement in recall over existing methods. Furthermore, after deployingCOCO+in real-world scenarios, we observed an adoption rate of 89.5% and improved on-time pickups even under high-demand conditions, as confirmed through A/B testing.

Pedestrian trajectory prediction is a critical task in various mobile computing applications, such as video surveillance, robot navigation, autonomous driving, and human mobility analysis. Although significant progress has been made by current methods, the challenge of observation deficiency in pedestrian trajectory prediction remains largely unaddressed. Since most existing methods focus on optimizing prediction accuracy under the assumption of complete observations, while ignoring the potential for observation deficiency caused by failures in detection or tracking algorithms. To overcome this challenge, we propose a collaborative observation imputation and trajectory prediction framework, which employs consistency evaluation to jointly perform the imputation and prediction tasks. Specifically, we first build a consistency evaluation module to align features between observed and future trajectory pairs using contrastive learning. Then, we design a trajectory imputation and prediction baseline, which adopts a parallel paradigm, to mitigate the impact of coarse imputations on trajectory prediction when performing initial imputation and prediction. Next, we introduce a consistency-guided Skip-Diffusion module, which leverages consistency evaluation between initial imputations and ground truth future trajectories to refine the initial imputations. Finally, we propose a consistency-driven Cross-Mamba module, which uses consistency evaluation between ground-truth observations and initial predictions to refine the initial predictions. Extensive experiments demonstrate the effectiveness of the proposed framework in both imputation and prediction tasks.

Heart rate (HR) and respiratory rate (RR) are essential physiological indicators of people's physical function and exercise performance. Advancement in sensor technology has rendered earable devices with in-ear microphones feasible for vital sign monitoring. However, it is rather challenging to monitor heart rate and respiration simultaneously with a single earable device especially when a person is doing exercises. This is because intense physical activities can lead to significant noise interference which can easily obscure physiological signals. To address this challenge, this paper presents VitalEar, an exercise physiological monitoring system based on in-ear microphones, designed to estimate HR and RR while addressing complex motion interference and variability in users and activities. VitalEar employs Empirical Wavelet Transform (EWT) to decompose heartbeats into periodic and harmonic coefficients, enhancing noise reduction in the ECG spectrogram reconstruction model. Additionally, VitalEar incorporates a DCN-LSTM-based breathing curve reconstruction model to mitigate background noise and variability in user and activity. The experiments show that VitalEar achieves an average MAE of 5.61 BPM and 2.31 RPM, MAPE of 4.16% and 10.58% for HR and RR estimation, respectively. Compared to related work, our approach offers significant advantages in robustness against intense physical activities

Radar-based human activity recognition (HAR) has emerged as a critical component in various applications, ranging from smart homes to healthcare monitoring. Radar has several advantages: a wide detection range, a certain penetration ability, non-contact and non-perception detection ability, not being affected by light, and privacy-preserving. However, achieving high accuracy with efficiency remains a significant challenge due to the complexity of radar signals and the variability in human activities. Currently, the majority of research efforts are centered on enhancing performance, often at the expense of computational efficiency. In this paper, we propose RadarAttn, a novel approach that integrates visual attention mechanisms with self-attention to enhance the performance and efficiency of HAR systems. The architecture of RadarAttn can reduce floating-point operations (FLOPs) and parameter counts while improving accuracy. Our method leverages the visual attention mechanism to focus on the most relevant regions of radar spectrograms. Simultaneously, the self-attention mechanism is used to capture long-range dependencies within the radar signal, enabling the model to learn complex patterns associated with different activities. Experimental results on benchmark radar-based HAR datasets demonstrate that RadarAttn significantly outperforms state-of-the-art methods in both accuracy and computational efficiency. Our approach offers a promising direction for developing robust and scalable radar-based HAR systems for real-world applications.

Psychiatric disorders represent a growing global public health concern. Current screening methods primarily rely on subjective self-reports, which prone to bias and often impede timely intervention. In this paper, we propose G-PsychSW (Generalizable Psychiatric Screening from Wearable), a continuous and objective mental health monitoring based on multi-channel biosignals, including photoplethysmography (PPG), galvanic skin response (GSR), skin temperature (SKT), and blood oxygen (OXY). To address challenges such as individual variability and limited interpretability, we introduce causal disentanglement to learn subject-invariant representations to enable domain generalization (DG). We further incorporate multi-scale attention convolution (MAConv) and functional graph construction (FGC) to model temporal dynamic and inter-channel interaction, uncovering meaningful interpretability analysis corresponding to psychial biomarkers. Experiments on a customized dataset under leave-one-subject-out cross-validation show that G-PsychSW achieves 8 0. 1 6% accuracy and 8 6. 5 2% F1-score across four-class psychiatric screening, including healthy, depression, anxiety and bipolar disorders, outperforming state-of-the-art DG methods and validating its effectiveness.

Mental disorders are increasingly prevalent, requiring timely and accurate recognition. Existing methods often depend on specialized devices and single-modality public datasets, limiting their ability to identify diverse mental disorders. Multiclass mental disorder recognition (MDR) faces challenges such as limited labeled data, overlapping symptoms among disorders, and significant individual variability. This paper introduces CoMe, a system that leverages low-cost, ubiquitous wearable devices to collect multimodal physiological signals, enabling real-time and daily monitoring to support the early detection of mental disorders. CoMe employs a multimodal residual fusion contrastive learning framework to extract both modality-specific and cross-modality knowledge, addressing challenges of sparse labeled data and symptom overlap. Through self-supervised and supervised fine-tuning, CoMe personalizes models for new users, facilitating accurate, real-time mental status monitoring in daily life. Validation with 74 participants demonstrates CoMe's superior performance under low-data conditions, outperforming both self-supervised and supervised baselines, and surpassing a state-of-the-art method. CoMe achieves average recognition accuracies of 84.5%, 85.36%, 83.25%, and 83.54% for depression, anxiety, bipolar disorders, and control group, respectively.

With the rise of virtual communication and smart devices, speech has become the most natural medium of interaction. Yet it remains intrinsically difficult: speech is fleeting, unstructured, and disfluent, making key information prone to loss. Conventional Speech-to-Text (STT) systems attempt to acoustically reconstruct what was said. However, their frame-level alignment and rigid token-by-token decoding break down under noise, interruptions, or fragmentation. Humans, in contrast, readily grasp what was meant by exploiting syntax, discourse, pragmatics, and prosody. We argue for a paradigm shift from acoustic reconstruction to semantic transduction: inferring meaning directly from speech, abstracted from surface distortions. This shift raises two challenges: (C1) the lack of anchors between audio and meaning, and (C2) the need to maintain compositional semantics. To address these, we introduce CogTrans, a cognitively inspired speech-to-meaning framework. CogTrans tackles C1 through a Semantic Anchor Explorer, built on I-JEPA to capture higher-order regularities, prosodic rhythms, cross-frequency coarticulation, discourse continuity-providing resilient semantic scaffolds under noise and fragmentation. For C2, it designs a Lexical-Semantic Harmonizer that dynamically integrates these anchors with lexical embeddings; thereby preserving fine-grained compositional fidelity in roles, order, and entities. Extensive experiments show that CogTrans delivers consistent and substantial gains under challenging conditions. On GigaSpeech, it achieves a 6.58% relative Word Error Rate (WER) reduction, and on the multilingual VoxPopuli benchmark, the gain climbs to 12.97% at 10 dB noise-a regime where conventional models typically collapse. Beyond literal accuracy, CogTrans also boosts semantic fidelity, with a 3.40% increase in ROUGE-L and 3.45% in USE-Sim, ensuring transcripts remain faithful not only in words but also in meaning. Together, these results underscore that CogTrans is robust in noisy, unconstrained environments-precisely the conditions where reliability matters most.

Large language models (LLMs) have received lots of attention for their impressive performance in in-context dialogues and their potential to revolutionize service industries with a new business model, Model-as-a-Service (MaaS). Automated data labeling is a natural and promising service. However, labeling data with LLMs faces two main challenges: 1) the labels from LLMs may contain uncertainty, and 2) using LLMs for data labeling tasks can be prohibitively expensive, as the scales of datasets are usually tremendous. In this paper, we propose a hierarchical framework named LMCrowd that leverages multiple LLMs for efficient data labeling under budget constraints. The proposed LMCrowd framework first aggregates labels from multiple freely available LLMs, and then employs a large, paid MaaS LLM for relabeling selected instances. Furthermore, we formalize the core process as an optimization problem, aiming to select the optimal set of instances for relabeling by the MaaS LLM, given the current belief state. Extensive experimental evaluations across various real-world datasets demonstrate that our framework outperforms human labelers and GPT-4 in terms of both accuracy and efficiency.

Periodic or quasi-periodic phenomena reveal intrinsic characteristics in various natural processes, such as weather patterns, movement behaviors, traffic flows, and biological signals. Given that these phenomena span multiple modalities, the capabilities of Multimodal Large Language Models (MLLMs) offer promising potential to effectively capture and understand their complex nature. However, current MLLMs struggle with periodic tasks due to limitations in: 1) lack of temporal modelling and 2) conflict between short and long periods. This paper introduces Period-LLM, a multimodal large language model designed to enhance the performance of periodic tasks across various modalities, and constructs a benchmark of various difficulty for evaluating the cross-modal periodic capabilities of large models. Specially, We adopt an "Easy to Hard Generalization" paradigm, starting with relatively simple text-based tasks and progressing to more complex visual and multimodal tasks, ensuring that the model gradually builds robust periodic reasoning capabilities. Additionally, we propose a "Resisting Logical Oblivion" optimization strategy to maintain periodic reasoning abilities during semantic alignment. Extensive experiments demonstrate the superiority of the proposed Period-LLM over existing MLLMs in periodic tasks. The code is available at https: //github.com/keke-nice/Period-LLM.

Cross-Document Multi-entity Question Answering (MEQA) demands the integration of scattered information across documents to resolve complex queries involving entities, relationships, and contextual dependencies.Although Large Language Models (LLMs) and Retrieval-augmented Generation (RAG) systems show promise, their performance on crossdocument MEQA remains underexplored due to the absence of tailored benchmarks.To address this gap, we introduce MEBench, a scalable multi-document, multi-entity benchmark designed to systematically evaluate LLMs' capacity to retrieve, consolidate, and reason over scattered and dense information.Our benchmark comprises 4,780 questions which are systematically categorized into three primary categories: Comparative Reasoning, Statistical Reasoning and Relational Reasoning, further divided into eight distinct types, ensuring broad coverage of real-world multi-entity reasoning scenarios.Our experiments on state-of-theart LLMs reveal critical limitations: even advanced models achieve only 59% accuracy on MEBench.Our benchmark emphasizes the importance of completeness and factual precision of information extraction in MEQA tasks, using Entity-Attributed F1 (EA-F1) metric for granular evaluation of entity-level correctness and attribution validity.MEBench not only highlights systemic weaknesses in current LLM frameworks but also provides a foundation for advancing robust, entity-aware QA architectures. 1

Text-to-Vis is an emerging task in the data engineering and mining area that aims to automatically generate data visualizations from natural language questions (NLQs). Despite their progress, existing text-to-vis models often heavily rely on lexical matching between words in the questions and tokens in data schemas. This overreliance on lexical matching may lead to a diminished level of model robustness against input variations. In this study, we thoroughly examine the robustness of current text-to-vis models, an area that has not previously been explored. In particular, we construct the first robustness dataset nvBench-Rob, which contains diverse lexical and phrasal variations based on the original text-to-vis benchmark nvBench. Then, we found that the performance of existing text-to-vis models on this new dataset dramatically drops, implying that these methods exhibit inadequate robustness overall. Finally, we propose a novel framework based on Retrieval-Augmented Generation (RAG) technique, named GRED, specifically designed to address input perturbations in these two variants. The framework consists of three parts: NLQ-Retrieval Generator, Visualization Query-Retrieval Retuner and Annotation-based Debugger, which are used to tackle the challenges posed by natural language variants, programming style differences and data schema variants, respectively. Extensive experimental evaluations show that, compared to the state-of-the-art model Prompt4Vis in the Text-to-Vis field, GRED performs better in terms of model robustness, with a 40% increase in accuracy on the proposed nvBench-Rob dataset.

In recent years, an increasing number of deep reinforcement learning methods have achieved success in domains such as gaming, yet their inherent black-box nature poses significant challenges to the interpretability of the training process. Moreover, there is an urgent need for the ability to intervene directly and simply during training. To address these issues, we present the Interactive Hierarchical Reinforcement Learning Monitoring System (HRLMS). This framework integrates a set of rules derived from both autonomously generated rules and those input through user interaction, showcasing these rules in real-time during the training process. Throughout the system's operation, the input, integration, display, and reuse of rules form a comprehensive chain, enhancing the entirety of the workflow by seamlessly blending the training and display processes.

Federated Learning (FL) enables decentralized model training while preserving data privacy. Despite its benefits, FL faces challenges with non-identically distributed (non-IID) data, especially in long-tailed scenarios with imbalanced class samples. Momentum-based FL methods, often used to accelerate FL convergence, struggle with these distributions, resulting in biased models and making FL hard to converge. To understand this challenge, we conduct extensive investigations into this phenomenon, accompanied by a layer-wise analysis of neural network behavior. Based on these insights, we propose FedWCM, a method that dynamically adjusts momentum using global and per-round data to correct directional biases introduced by long-tailed distributions. Extensive experiments show that FedWCM resolves non-convergence issues and outperforms existing methods, enhancing FL’s efficiency and effectiveness in handling client heterogeneity and data imbalance.

Global aging necessitates tele-monitoring systems to provide real-time tracking and timely assistance for older adults living independently. While pervasive wireless devices (e.g., CSI, IMU, UWB) enable cost-effective, non-intrusive monitoring, existing systems lack flexibility, limiting their adaptability to different environments. In this work, we posit that the motion dynamics of human movement are invariant across sensing modalities, inspiring the design of HARMONY—a privacy-preserving, sensor-agnostic system that supports multi-modal inputs and diverse tele-monitoring tasks. HARMONY incorporates Modality-agnostic Data Processing to uniformly encrypt multi-modal signals and Task-specific Activity Recognition for seamless tasks adaptation. A novel Encrypted-processing Engine then significantly accelerates computations on encrypted data by optimizing matrix and convolution operations. Evaluations across five different sensing modalities show that HARMONY consistently achieves high accuracy while delivering 3.5 × to 130 × speedups over state-of-the-art baselines. Our results demonstrate that HARMONY is a practical, scalable, and privacy-centric prototype for next-generation remote healthcare.

An emerging type of fraud involves malicious senders exploiting the blind shipment and cash-on-delivery (COD) mechanisms by dispatching large volumes of unsolicited, low-cost parcels. If unsuspecting receivers accept these parcels, they pay for both shipping and goods; otherwise, logistics providers bear the round-trip shipping costs. Existing detection techniques, which rely on extensive labeled cases, struggle with this emerging fraud because receivers' unawareness and low transaction values discourage complaints, resulting in few confirmed cases. Therefore, we propose leveraging receivers' complaints, though not initially collected for fraud detection, to uncover subtle indicators of fraud patterns, while addressing three challenges: (C1) noise-rich dialogues(C2) data privacy concerns, and (C3) ever-evolving fraud patterns. To address them, we design BLOFF, a Blind shipment detection Framework for LO gistics Fraud powered by large language models (LLMs). Specifically, BLOFF includes three components: i) Sensitivity Anonymization to protect sensitive user information; ii) Dialogue Profile Distillation to transform informal dialogues into structured representation, addressing C1, and distill knowledge from a teacher LLM (GPT-4o) to a lightweight student LLM (ChatGLM4-9B), addressing C2; ii) Multi-faceted Context Augmentation to enhance the interpretation of fraud signatures and adaptation of evolving patterns, addressing C3. We evaluate BLOFF on about 56,000 complaints records collected from JD Logistics between January and November 2024. Results show that BLOFF outperforms state-of-the-art methods, achieving a 10.19% improvement in precision. Furthermore, during its real-world deployment in December 2024, BLOFF identified over 90 fraudulent parcels with a 91.4% precision.

Underwater images typically suffer from severe colour distortions, low visibility, and reduced structural clarity due to complex optical effects such as scattering and absorption, which greatly degrade their visual quality and limit the performance of downstream visual perception tasks. Existing enhancement methods often struggle to adaptively handle diverse degradation conditions and fail to leverage underwater-specific physical priors effectively. In this paper, we propose a degradation-aware conditional diffusion model to enhance underwater images adaptively and robustly. Given a degraded underwater image as input, we first predict its degradation level using a lightweight dual-stream convolutional network, generating a continuous degradation score as semantic guidance. Based on this score, we introduce a novel conditional diffusion-based restoration network with a Swin UNet backbone, enabling adaptive noise scheduling and hierarchical feature refinement. To incorporate underwater-specific physical priors, we further propose a degradation-guided adaptive feature fusion module and a hybrid loss function that combines perceptual consistency, histogram matching, and feature-level contrast. Comprehensive experiments on benchmark datasets demonstrate that our method effectively restores underwater images with superior colour fidelity, perceptual quality, and structural details. Compared with SOTA approaches, our framework achieves significant improvements in both quantitative metrics and qualitative visual assessments.

Hypertension affects 1.28 billion adults, but blood pressure alone is insufficient for accurate hypertension detection. In this paper, we identified a range of multi-dimensional hemodynamics as novel biomarkers that can offer more comprehensive cardiovascular insights for detecting hypertension. We design a novel system that collects ballistocardiogram (BCG) signals from an optic fiber sensor mat. It features a multi-task, multi-branch, unsupervised domain adaptation learning framework, enabling simultaneous prediction of five hemodynamic biomarkers for hypertension detection. In a 4-month trial with 85 subjects, Hyde achieved 97.65% average accuracy for hypertension detection.

Accurate modeling of traffic demand through cellular traffic generation is crucial for optimizing base station deployment. We thus present STOUTER, a Spatio-Temporal diffusiOn model for cellUlar Traffic genERation. To effectively capture spatial and temporal dynamics, we pretrain both a temporal graph and a base-station graph, and introduce a Spatio-Temporal Feature Fusion Module (STFFM). On five datasets from two regions, STOUTER reduces Jensen-Shannon Divergence by 52.8% over prior methods, generating distributions that closely match real traffic and aiding downstream planning tasks.

The dental plaque disclosing agent helps guide plaque removal by revealing plaque on the teeth. However, existing toothbrushes cannot provide visualization of the stained plaque beneath the toothpaste foam during brushing. To fill this gap, this paper proposes PIT and uses smartphone to provide real-time plaque visualization during brushing. PIT introduces novel hardware, including a micro camera, four green LEDs, and a mechanical structure, offering a stable camera view of bristle position and bristle rotation. To address the challenge of toothpaste foam obstruction, we derived an optical channel model that guided the design of the light source to enhance plaque visibility. Furthermore, we designed a deep neural network specifically for plaque segmentation under thick foam. Finally, the trained distilled student model was run on a smartphone and evaluated on both denture models and human subjects. The results show that PIT achieved an IoU (Intersection over Union) of 75.22% and a latency of 29 ms, with robust performance across various conditions. Evaluation by 10 participants revealed that PIT helped reduce plaque coverage to 5.6% within two minutes of brushing, significantly outperforming existing advanced toothbrushes.

In federated learning (FL), models must \emph{converge quickly} under tight communication budgets while \emph{generalizing} across non-IID client distributions. These twin requirements have naturally led to two widely used techniques: client/server \emph{momentum} to accelerate progress, and \emph{sharpness-aware minimization} (SAM) to prefer flat solutions. However, simply combining momentum and SAM leaves two structural issues unresolved in non-IID FL. We identify and formalize two failure modes: \emph{local–global curvature misalignment} (local SAM directions need not reflect the global loss geometry) and \emph{momentum-echo oscillation} (late-stage instability caused by accumulated momentum). To our knowledge, these failure modes have not been jointly articulated and addressed in the FL literature. We propose \textbf{FedWMSAM} to address both failure modes. First, we construct a momentum-guided global perturbation from server-aggregated momentum to align clients' SAM directions with the global descent geometry, enabling a \emph{single-backprop} SAM approximation that preserves efficiency. Second, we couple momentum and SAM via a cosine-similarity adaptive rule, yielding an early-momentum, late-SAM two-phase training schedule. We provide a non-IID convergence bound that \emph{explicitly models the perturbation-induced variance} $\sigma_\rho^2=\sigma^2+(L\rho)^2$ and its dependence on $(S,K,R,N)$ on the theory side. We conduct extensive experiments on multiple datasets and model architectures, and the results validate the effectiveness, adaptability, and robustness of our method, demonstrating its superiority in addressing the optimization challenges of Federated Learning. Our code is available at \url{https://github.com/Li-Tian-Le/NeurlPS_FedWMSAM}.

Robust and unobtrusive in-vehicle physiological monitoring is crucial for ensuring driving safety and user experience. While remote physiological measurement (RPM) offers a promising non-invasive solution, its translation to real-world driving scenarios is critically constrained by the scarcity of comprehensive datasets. Existing resources are often limited in scale, modality diversity, the breadth of biometric annotations, and the range of captured conditions, thereby omitting inherent real-world challenges in driving. Here, we present PhysDrive, the first large-scale multimodal dataset for contactless in-vehicle physiological sensing with dedicated consideration of various modality settings and driving factors. PhysDrive collects data from 48 drivers, including synchronized RGB, near-infrared camera, and raw mmWave radar data, accompanied by six synchronized ground truths (ECG, BVP, Respiration, HR, RR, and SpO2). It covers a wide spectrum of naturalistic driving conditions, including driver motions, dynamic natural light, vehicle types, and road conditions. We extensively evaluate both signal‑processing and deep‑learning methods on PhysDrive, establishing a comprehensive benchmark across all modalities, and release full open‑source code with compatibility for mainstream public toolboxes. We envision PhysDrive will serve as a foundational resource and accelerate research on multimodal driver monitoring and smart‑cockpit systems.

The rise of spy cameras has become a global concern regarding personal privacy. Unfortunately, affordable and easily accessible detection methods are lacking, leaving individuals vulnerable to these intrusive devices. Currently, most research focuses on detecting cameras based on their wireless transmission. Some of the latest research can detect other types of emissions from cameras but require costly equipment like universal software radio peripherals (USRP) or thermal cameras. However, this paper introduces a ubiquitous, cost-effective (less than $2) system, CamFirm, to identify and locate spy cameras regardless of their transmission mode. CamFirm only requires wired headphones and a Frequency Modulation (FM) module. Despite limitations of the FM module, such as narrow bandwidth, distortion, and vulnerability to interference, CamFirm remains capable of accurately identifying camera signals and pinpointing their locations. By utilizing the harmonics of digital circuits for signal screening, CamFirm can detect the photosensitive sensor signature of cameras and ascertain its direction through the attenuation of electromagnetic waves by the human body. The camera’s position can be further confirmed by activating a phone’s flashlight. Through extensive testing on 18 different cameras, CamFirm can achieve a median detection distance of 2.6 meters, effectively guiding users in locating the hidden camera.

In this paper, we propose Argus, a wearable add-on system based on stripped-down (i.e., compact, lightweight, low-power, limited-capability) mmWave radars. It is the first to achieve egocentric human mesh reconstruction in a multi-view manner. Compared with conventional frontal-view mmWave sensing solutions, it addresses several pain points, such as restricted sensing range, occlusion, and the multipath effect caused by surroundings. To overcome the limited capabilities of the stripped-down mmWave radars (with only one transmit antenna and three receive antennas), we tackle three main challenges and propose a holistic solution, including tailored hardware design, sophisticated signal processing, and a deep neural network optimized for high-dimensional complex point clouds. Extensive evaluation shows that Argus achieves performance comparable to traditional solutions based on high-capability mmWave radars, with an average vertex error of 6.5 cm, solely using stripped-down radars deployed in a multi-view configuration. It presents robustness and practicality across conditions, such as with unseen users and different host devices.

This study unveils a cutting-edge camera-based system for the automated monitoring of hand hygiene practices within healthcare settings. Utilizing advanced computer vision and machine learning technologies, our system employs three strategically placed synchronized cameras around a wash basin. These cameras capture the handwashing process from multiple perspectives, allowing for detailed analysis of hand movements including finger and wrist dynamics. The extracted skeletal coordinate data are processed by a Gesture Category Model (GCM), which automatically identifies handwashing gestures. The model is rigorously trained on a dataset comprising video recordings from 55 healthcare professionals, focusing on the World Health Organizations seven-step hand-washing protocol. Furthermore, we introduce a Counting Algorithm to quantify the frequency and duration of each gesture, coupled with a Quality Assessment Model (QAM) that evaluates compliance with hand hygiene standards. The systems precision and its strong correlation with expert annotations highlight its potential to significantly enhance hand hygiene compliance and reduce healthcare-associated infections.

Remote photoplethysmography (rPPG) has been widely applied to measure heart rate from face videos. To increase the generalizability of the algorithms, domain generalization (DG) attracted increasing attention in rPPG. However, when rPPG is extended to simultaneously measure more vital signs (e.g., respiration and blood oxygen saturation), achieving generalizability brings new challenges. Although partial features shared among different physiological signals can benefit multi-task learning, the sparse and imbalanced target label space brings the seesaw effect over task-specific feature learning. To resolve this problem, we designed an end-to-end Mixture of Low-rank Experts for multi-task remote Physiological measurement (PhysMLE), which is based on multiple low-rank experts with a novel router mechanism, thereby enabling the model to adeptly handle both specifications and correlations within tasks. Additionally, we introduced prior knowledge from physiology among tasks to overcome the imbalance of label space under real-world multi-task physiological measurement. For fair and comprehensive evaluations, this paper proposed a large-scale multi-task generalization benchmark, named Multi-Source Synsemantic Domain Generalization (MSSDG) protocol. Extensive experiments with MSSDG and intra-dataset have shown the effectiveness and efficiency of PhysMLE. In addition, a new dataset was collected and made publicly available to meet the needs of the MSSDG. The code and data are available at https://github.com/WJULYW/PhysMLE.

Radar sensing is gaining increasing attention due to its unique advantages, including being device-free, privacy-preserving, and capable of penetrating obstacles. It has been extensively studied in various applications such as human activity recognition, vital sign monitoring, and person identification. However, most existing research focuses on a single specific application, and there remains a lack of studies or datasets dedicated to multi-task radar sensing. In this paper, we collected a dataset for two sensing tasks, including gesture recognition and person identification, via a miniature mm-wave radar. The raw radar signals were processed using micro-Doppler and range-Doppler techniques to extract spectral and spatial representations. We propose an improved multi-task radar sensing framework (MT-DualFormer) that incorporates attention-based cross-task feature distillation and contrastive learning to maximize task performance. MT-DualFormer consists of dual branches with CNN and Transformer modules, capturing both spatial and temporal dependencies in radar data. Attention-based cross-task feature distillation enables knowledge transfer between gesture recognition and person identification tasks. Meanwhile, contrastive learning ensures embedding space separability, facilitating robust task-specific classification. In the evaluation, MT-DualFormer achieves accuracy rates of 98.87% for gesture recognition and 97.96% for person identification, surpassing five representative multi-task approaches and ten state-of-the-art models. This study underscores the importance of leveraging task correlations to enhance the performance of radar-based sensing systems.

Owing to flexible management and low overhead, wireless physical layer security (PLS) has been applied to support many critical applications (e.g., data dissemination) of unmanned aerial vehicles (UAVs)-based mobile communication networks in emergency scenarios. Although the impact of static network scenarios or receiver mobility on PLS has been well studied, there is no much work that studies the impact of UAV-based transmitter mobility on PLS. To fill this gap, in this paper, we investigate PLS of a scenario, where a random mobile UAV-based transmitter transmits information to a static ground entity under Rayleigh fading channel. More specifically, we consider a communication system, in which a mobile UAV hovers over a region to collect information and then disseminates this information to a static ground network entity in a confidential manner under the presence of an eavesdropper. Because of popularity and practicality, the UAV is assumed to hover following the random way point (RWP) mobility model. We investigate the secrecy characteristics of the UAV under steady running state in terms of ergodic secrecy capacity (ESC), positive secrecy capacity probability (PSCP) and secrecy outage probability (SOP) for the communication between the UAV and the receiver. We then investigate the secrecy performance of the proposed system while considering the pause time of the RWP mobility model adopted by the UAV. We further extend our proposed theoretical model to other realistic scenarios, including the presence of multiple cooperative and non-cooperative eavesdroppers, and study the PSCP and SOP metrics of the corresponding system. Furthermore, we propose three types of secrecy improvement strategies for the considered communication model. We strike a good trade-off between the secrecy improvement and transmit outage probability. Extensive simulations have been conducted to validate our theoretical analysis as well as the effectiveness of the proposed secrecy improvement strategies.

Remote photoplethysmography (rPPG) is a promising technology for capturing physiological signals from facial videos, with potential applications in medical health, affective computing, and biometric recognition. The demand for rPPG tasks has evolved from achieving high performance in intra-dataset testing to excelling in cross-dataset testing (i.e., domain generalization). However, most existing methods have overlooked the incorporation of prior knowledge specific to rPPG, leading to limited generalization capabilities. In this paper, we propose a novel framework that effectively integrates both explicit and implicit prior knowledge into the rPPG task. Specifically, we conduct a systematic analysis of noise sources (e.g., variations in cameras, lighting conditions, skin types, and motion) across different domains and embed this prior knowledge into the network design. Furthermore, we employ a two-branch network to disentangle physiological feature distributions from noise through implicit label correlation. Extensive experiments demonstrate that the proposed method not only surpasses state-of-the-art approaches in RGB cross-dataset evaluation but also exhibits strong generalization from RGB datasets to NIR datasets. The code is publicly available at https://github.com/keke-nice/Greip.

The rapid digitization of healthcare systems has led to a vast accumulation of electronic medical records (EMRs), offering an invaluable source of patient data that can significantly advance medical research and improve patient care. However, sharing EMRs for research purposes presents challenges, particularly concerning data privacy, security, and the limitations of traditional centralized data-sharing models. This paper introduces a novel approach that leverages blockchain technology to facilitate federated learning with EMRs, thereby addressing these challenges. Federated learning enables multiple institutions to collaboratively train a robust machine learning model without sharing raw data, preserving privacy and security. By integrating blockchain, this framework enhances data integrity, immutability, and trust, all in a decentralized environment. Blockchain serves as a transparent and secure ledger, recording model updates and aggregating them through a consensus-based mechanism. Smart contracts further enforce data usage policies, allowing only authorized access and maintaining control over data ownership and sharing. This approach empowers medical researchers and institutions to collaborate more effectively, accelerating the discovery of treatments, advancements in personalized medicine, and insights into rare diseases. It also enables patients to contribute to medical research while retaining control over their personal data, fostering a patient-centered approach to healthcare innovation. Experimental results confirm the efficacy and efficiency of this blockchain-enabled federated learning framework, highlighting its potential to transform medical research and adhere to stringent privacy and security standards. This study emphasizes the pivotal role of blockchain in enhancing Big Data analytics within healthcare, paving the way for improved collaboration, innovation, and patient outcomes.

Map matching aims to align GPS trajectories to their actual travel routes on a road network, which is an essential pre-processing task for most of trajectory-based applications. Many map matching approaches utilize Hidden Markov Model (HMM) as their backbones. Typically, HMM treats GPS samples of a trajectory as observations and nearby road segments as hidden states. During map matching, HMM determines candidate states for each observation with a fixed searching range, and computes the most likely travel route using theViterbialgorithm. Although HMM-based approaches can derive high matching accuracy, they still suffer from high computation overheads. By inspecting the HMM process, we find that the computation bottleneck mainly comes from improper candidate sets, which contain many irrelevant candidates and incur unnecessary computations. In this paper, we present$\mathtt {LiMM}$– a learned road network index structure for efficient map matching.$\mathtt {LiMM}$improves existing HMM-based approaches from two aspects. First, we propose a novel learned index for road networks, which considers the characteristics of road data. Second, we devise an adaptive searching range mechanism to dynamically adjust the searching range for GPS samples based on their locations. As a result,$\mathtt {LiMM}$can provide refined candidate sets for GPS samples and thus accelerate the map matching process. Extensive experiments are conducted with three large real-world GPS trajectory datasets. The results demonstrate that$\mathtt {LiMM}$significantly reduces computation overheads by achieving an average speedup of$11.7\times$than baseline methods, merely with a subtle accuracy loss of 1.8%.

Incorrect brushing methods normally lead to poor oral hygiene, and result in severe oral diseases and complications. While effective brushing can address this issue, individuals often struggle with incorrect brushing, like aggressive brushing, insufficient brushing, and missing brushing. To break this stalemate, in this paper, we proposed LiT, a toothbrushing monitoring system to assess the brushing status on 16 surfaces using the Bass technique. LiT utilizes commercial LED toothbrushes’ blue LEDs as transmitters, and incorporates only two low-cost photodetectors as receivers on the toothbrush head. It is challenging to determine optimal deployment positions and minimize photodetectors number to establish the light transmission channel in oral cavity. To address these challenges, we established mathematical models within the oral cavity based on the two photodetectors’ deployment to theoretically validate the feasibility and prove robustness. Furthermore, we designed a comprehensive framework to fight against the implementation challenges including brushing action separation, light interference on the outer surfaces of front teeth, toothpaste diversity, user variations, brushing hand variability, and incorrect brushings. Experimental results demonstrate that LiT achieves a highly accurate surface recognition rate of 95.3%, an estimated error for brushing duration of 6.1%, and incorrect brushing detection accuracy of 96.9%. Furthermore, LiT retains stable capability under a variety of circumstances, such as various lighting conditions, user movement, toothpaste diversity, and left and right-handed users.

Dynamic ridesharing has gained significant attention in recent years. However, existing ridesharing studies often focus on optimizing order dispatching and vehicle repositioning separately, leading to short-sighted decisions and underutilization of the ridesharing potential. In this paper, we propose a novel joint optimization framework called$\mathtt {JODR}$. By coordinating order dispatching and vehicle repositioning,$\mathtt {JODR}$enhances ridesharing efficiency while ensuring high-quality service. The core idea of$\mathtt {JODR}$is to dispatch ride orders with high demand in specific mobility directions to vehicles with sufficient available capacity, effectively balancing future supply and demand in those directions. To achieve this, we introduce a novel mobility value function that can predict the long-term mobility value of matching an order with its travel direction. By considering orders’ directional mobility values, service quality assessments, and available vehicle capacities,$\mathtt {JODR}$formulates the order dispatching as a minimum-cost maximum-flow problem to derive the optimal order-vehicle assignments. Furthermore, the value function helps the intelligent repositioning of idle vehicles. Extensive experiments conducted on a large real-world dataset demonstrate the superiority of$\mathtt {JODR}$over state-of-the-art methods across various performance metrics. These experimental results validate the effectiveness of$\mathtt {JODR}$in improving the ridesharing efficiency and experience.

Current research in human activity recognition primarily emphasizes enhancing accuracy, with limited exploration into computational efficiency and hardware compatibility. Recently, Mamba has sparked substantial interest within the realm of deep learning. Mamba is a hardware-aware algorithm enabling very efficient training and inference. Researchers are applying Mamba to various tasks, demonstrating significant promise in both language and vision tasks. It is worthwhile to investigate the use of Mamba for efficient human activity recognition. In this paper, we proposed a hybrid neural network that integrates CNN and visual Mamba, called ActivityMamba. The SE-Mamba block in ActivityMamba utilizes both CNN’s local and Mamba’s global context modeling while keeping computation and memory efficiency. We evaluated the ActivityMamba on five public benchmark datasets collected by using three different sensing techniques. ActivityMamba achieved higher performance than vision transformers, vision Mamba, and CNNs with fewer FLOPs and parameters. It sets a new SOTA on all five datasets, which are 91.78% OA and 89.13% F1 on the USC-HAD dataset, 99.19% OA and 98.64% F1 on the UT-HAR dataset, 99.82% OA and F1 on the DIAT dataset, 98.59% OA and 98.65% F1 on the UCI-HAR dataset, and 95.41% OA and 93.14% F1 on the UniMib dataset. Our work is the first to investigate the CNN-Mamba hybrid network for efficient human activity recognition.

Transformer architectures are popularized in both vision and natural language processing tasks, and they have achieved new performance benchmarks because of their long-term dependencies modeling, efficient parallel processing, and increased model capacity. While transformers offer powerful capabilities, their demanding computational requirements clash with the real-time and energy-efficient needs of edge-oriented human activity recognition. It is necessary to compress the transformer to reduce its memory consumption and accelerate the inference. In this paper, we investigated the binarization of a transformer-DeepViT for efficient human activity recognition. For feeding sensor signals into DeepViT, we first processed sensor signals to spectrograms by using wavelet transform. Then we applied three methods to binarize DeepViT and evaluated it on three public benchmark datasets for sensor-based human activity recognition. Compared to the full-precision DeepViT, the fully binarized one (Bi-DeepViT) reduced about 96.7% model size and 99% BOPs (Bit Operations) with only a little accuracy compromised. Furthermore, we explored the effects of binarizing various components and latent binarization of DeepViT to understand their impact on the model. We also validated the performance of Bi-DeepViTs on two wireless sensing datasets. The result shows that a certain partial binarization can improve the performance of DeepViT. Our work is the first to apply a binarized transformer in HAR.

Intelligent transportation systems demand efficient resource allocation and task offloading to ensure low-latency, high-bandwidth vehicular services. The dynamic nature of vehicular environments, characterized by high mobility and extensive interactions among vehicles, necessitates considering time-varying statistical regularities, especially in scenarios with sharp variations. Despite the widespread use of traditional reinforcement learning for resource allocation, its limitations in generalization and interpretability are evident. To overcome these challenges, we propose an Intelligence-based Reinforcement Learning (IRL) algorithm. This algorithm utilizes active inference to infer the real world and maintain an internal model by minimizing free energy. Enhancing the efficiency of active inference, we incorporate prior knowledge as macro guidance, ensuring more accurate and efficient training. By constructing an intelligence-based model, we eliminate the need for designing reward functions, aligning better with human thinking, and providing a method to reflect the learning, information transmission and intelligence accumulation processes. This approach also allows for quantifying intelligence to a certain extent. Considering the dynamic and uncertain nature of vehicular scenarios, we apply the IRL algorithm to environments with constantly changing parameters. Extensive simulations confirm the effectiveness of IRL, significantly improving the generalization and interpretability of intelligent models in vehicular networks.

The increasing integration of Internet of Things (IoT) devices in Wireless Local Area Networks (WLANs) necessitates robust and efficient authentication mechanisms. While existing IoT authentication protocols address certain security concerns, they often fail to provide comprehensive protection against threats such as perfect forward secrecy violations, insider attacks, and key compromise impersonation, or impose significant computational and communication overhead on resource-constrained IoT systems. This paper presents a novel Extensible Authentication Protocol (EAP) based scheme for IoT-WLAN environments that addresses these security challenges while maintaining cost-effectiveness. Our approach utilizes elliptic curve cryptography and incorporates advanced features including perfect forward secrecy, strong identity protection, and explicit key confirmation. We provide a thorough security analysis using informal heuristics, formal methods (Random Oracle Model and BAN Logic), and automated verification with ProVerif. Performance evaluations demonstrate that our protocol achieves lower communication, storage, and computational costs compared to state-of-the-art solutions, with an average 79.6% reduction in computation time. A detailed comparison with existing schemes highlights the efficiency and enhanced security features of our proposed authentication mechanism for IoT-WLAN deployments.

Owing to the cost of collecting labeled sensor data, self-supervised learning (SSL) methods for human activity recognition (HAR) that effectively use unlabeled data for pretraining have attracted attention. However, applying prior SSL to COMPLEX activities in real industrial settings poses challenges. Despite the consistency of work procedures, varying circumstances, such as different sizes of packages and contents in a packing process, introduce significant variability within the same activity class. In this study, we focus on sensor data corresponding to characteristic and necessary actions (sensor data motifs) in a specific activity such as a stretching packing tape action in an assembling a box activity, and propose to train a neural network in self-supervised learning so that it identifies occurrences of the characteristic actions, i.e., Motif Identification Learning (MoIL). The feature extractor in the network is subsequently employed in the downstream activity recognition task, enabling accurate recognition of activities containing these characteristic actions, even with limited labeled training data. The MoIL approach was evaluated on real-world industrial activity data, encompassing the state-of-the-art SSL tasks with an improvement of up to 23.85% under limited training labels.

WiFi has been widely used for local area networking of devices and Internet access in the past two decades. Many researchers exploit WiFi signals for target sensing through analyzing the Channel State Information (CSI) of signals affected by the target movement. With the development of 6 G Integrated Sensing and Communication (ISAC), some researchers further consider using communication data packets for WiFi sensing. However, all the current works do not analyze the impact of ubiquitous interference on WiFi sensing performance. In this paper, we propose IRSensing, an interference-resistance design to improve the CSI quality under interference in the ISAC scenario, aiming to improve the WiFi sensing performance. IRSensing exploits the overall WiFi packet for CSI optimization. It first measures the interference level of each subcarrier based on variance analysis, then proposes a CSI optimization method based on maximal ratio combining to improve the CSI quality. It finally proposes a practical CSI enhancement process to adapt to complex interference situations in actual networks. We implement IRSensing on a hardware testbed and evaluate its performance under different settings. Experiment results show that it can significantly decrease the activity detection error rate by up to 80% and improve the classification accuracy by up to 15$\%$.

The increasing popularity of earable devices stimulates great academic interest to design novel head gesture-based interaction technologies. But existing works simply consider it as a singular activity recognition problem. This is not in line with practice since users may have different body movements such as walking and jogging along with head gestures. It is also beneficial to recognize body movements during human-device interaction since it provides useful context information. As a result, it is significant to recognize such composite activities in which actions of different body parts happen simultaneously. In this paper, we propose a system called CHAR to recognize composite head-body activities with a single IMU sensor. The key idea of our solution is to make use of the inter-correlation of different activities and design a multi-task learning network to extract shared and specific representations. We implement a real-time prototype and conduct extensive experiments to evaluate it. The results show that CHAR can recognize 60 kinds of composite activities (12 head gestures and 5 body movements) with high accuracies of 89.7% and 85.1% in sufficient data and insufficient data cases, respectively.

Due to the advantages of device ubiquity, natural interaction and privacy preservation, acoustic-based gesture input has received widespread attention. Researchers have proposed various techniques for different applications. However, the existing work has shortcomings of heavy data-collection overhead, non-continuous input, and performance degradation in crossuser scenarios. To overcome these shortcomings, we propose UltraWrite, an acoustic-based gesture input system that only needs extremely low data-collection overhead, supports continuous input, and achieves high cross-user recognition accuracy. The key idea of our solution is to synthesize training data of continuous gestures from isolated ones, build a lightweight continuous gesture recognition model based on connectionist temporal classification (CTC) mechanism, and design a novel decoupled model training strategy to improve its cross-user recognition capability. We have implemented prototype systems on commercial devices and conducted comprehensive experiments to evaluate their performance. The results show that UltraWrite achieves an average top-1 word accuracy of 99.3% and top-1 word error rate of 0.34%. In addition, we have also evaluated UltraWrite's robustness to the sensing distance, angle, background noise, and device. The results reveal that UltraWrite possesses strong robustness to these factors.

Acoustic-based human gesture recognition (HGR) offers diverse applications due to the ubiquity of sensors and touch-free interaction. However, existing machine learning approaches require substantial training data, making the process time-consuming, costly, and labor-intensive. Recent studies have explored cross-modal methods to reduce the need for large training datasets in behavior recognition, but they typically rely on open-source datasets that closely align with the target domain, limiting flexibility and complicating data collection. In this paper, we propose${\sf Img2Acoustic}$, a novel cross-modal acoustic-based HGR approach that leverages models trained on open-source image datasets (i.e., EMNIST, Omniglot) to effectively recognize custom gestures detected via acoustic signals. Our model incorporates a task-aware attention layer (TAAL) and a task-aware local matching layer (TALML), enabling seamless transfer of knowledge from image datasets to acoustic gesture recognition. We implement${\sf Img2Acoustic}$on commercial devices and conduct comprehensive evaluations, demonstrating that our method not only delivers superior accuracy and robustness compared to existing approaches but also eliminates the need for extensive training data collection.

With the rise of mobile devices, bel canto practitioners increasingly utilize smart devices as auxiliary tools for improving their singing skills. However, they frequently encounter timbre abnormalities during practice, which, if left unaddressed, can potentially harm their vocal organs. Existing singing assessment systems primarily focus on pitch and melody and lack real-time detection of bel canto timbre abnormalities. Moreover, the diverse vocal habits and timbre compositions among individuals present significant challenges in cross-user recognition of such abnormalities. To address these limitations, we propose TimbreSense, a novel bel canto timbre abnormality detection system. TimbreSense enables real-time detection of the five major timbre abnormalities commonly observed in bel canto singing. We introduce an effective feature extraction pipeline that captures the acoustic characteristics of bel canto singing. By applying temporal average pooling to the Short-Time Fourier Transform spectrogram, we reduce redundancy while preserving essential frequency-domain information. Our system leverages a transformer model with self-attention mechanisms to extract correlation and semantic features of overtones in the frequency domain. Additionally, we employ a few-shot learning approach involving pre-training, meta-learning, and fine-tuning to enhance the system’s cross-domain recognition performance while minimizing user usage costs. Experimental results demonstrate the system’s strong cross-user domain recognition performance and real-time capabilities.

Personalized recommendation is of paramount importance in online content platforms like Kuai and Tencent. To ensure accurate recommendations, it is crucial to consider multi-modal information in both items and user-user/item interactions. While existing works on multimedia recommendation have made strides in leveraging multi-modal contents to enrich item representations, many of them overlook the practical scenario of multiple modality missing. As a result, the performance of recommendation systems can be significantly compromised in such cases. In this paper, we introduce a novel multi-modal adversarial method called$MMAM$, which aims to provide reliable personalized recommendation services even in the presence of uncertain missing modalities. The core idea behind$MMAM$is to design a generator that can effectively encode both user-user/item interactions and multi-modal contents, taking into account various missing cases. The generator is trained to learn transferable features from different combinations of missing modalities in order to deceive a discriminative classifier. Additionally, we propose a modal discriminator that can classify the missing cases of multi-modalities, further enhancing the capability of the model. Moreover, a well-equipped predictor utilizes the transferable features to predict potential user interests. To improve the prediction accuracy, we design a type discriminator that enhances the classification of link types. By employing a mini-max game between the generator and the discriminators,$MMAM$successfully obtains transferable features that encompass multi-modal contents, even when facing uncertain missing modalities. We conduct extensive experiments on industrial datasets, including Kuai and Tencent. Comparing with state-of-the-art approaches, MMAM achieves improvements in personalized recommendation tasks under uncertain missing modalities. MMAM holds promise for enhancing multi-modal personalized recommendations in real-world applications.

The rising prevalence of mood disorders, including depression and anxiety, underscores the critical state of physical and mental health issues in contemporary society. Automatic emotion recognition technology emerges as a potential tool for monitoring mental health disorders, and offering valuable guidance through human-computer interfaces. However, existing technologies grapple with limitations, such as emotional concealment, potential privacy leakage, and specific device location restrictions. To overcome these limitations, we propose a mobile wearable emotion recognition system called EmoTracer incorporating ubiquitous multi-source sensors to measure physiology signals, and address challenges such as the intricate signal-emotion relationship, sparse data processing, and notable disparities among different subjects. We implement a real-time prototype and carry out comprehensive experiments to evaluate its performance. For six basic emotions, the results show that EmoTracer can achieve 95.6% accuracy in intra-subject emotion classification and 80.7% accuracy with 5 shots in cross-subject emotion classification.

With the rapid advancement of supervised learning and the rise of large language models, the demand for high-quality labeled datasets has surged. However, crowdsourced datasets often suffer from label noise. To address this challenge, we present a hierarchical crowdsourcing framework with diverse labor costs that aims to mitigate label noise. Our framework models crowdsourcing workers with varying labor costs and leverages hierarchical crowdsourcing under limited budget constraints to enhance data quality. We establish a loop for label selection and checking, strategically selecting checkers from different cost levels, including perfect workers (Oracles) and regular experts, to optimize the checking process. Additionally, we tackle an NP-hard problem in label selection. Experimental evaluation on a real-world dataset for the Recognizing Textual Entailment (RTE) task demonstrates a significant improvement in labeled dataset quality, leading to state-of-the-art performance in downstream tasks.

Electrocardiography (ECG) for objective cognitive load estimation gained increasing attention, and offers a more feasible and non-invasive alternative to traditional methods such as electroencephalography (EEG). Despite the promise of ECG signal, application in real-world scenarios is hampered by the domain shift present in data collected in controlled environments versus real-world settings. We propose a novel plug-in generalizable framework, CogDG-ECG, assessed on a first-introduced multi-source domain generalization (MSDG) protocol for generalized cognitive load estimation. CogDG-ECG bridges the domain gap by extracting domain-invariant features through adversarial learning, and estimating instance-specific unseen features by synthesizing plausible feature statistical variations. A new benchmark based on three datasets and MSDG protocol was introduced, which demonstrates the superiority of our proposed method.

Dynamic ridesharing has garnered significant attention in recent years due to its numerous benefits. Existing ridesharing algorithms often employ a “filter-and-refine” frame-work, where a large set of candidate vehicles is initially selected for each ride order, followed by computationally intensive route planning for each candidate. However, this process can lead to significant response delays and limit system efficiency. To address this challenge, we propose an order-vehicle pre-matching recommendation approach (PreMR) that refines the candidate set before route planning. PreMR leverages spatial-temporal intervals and a self-attention mechanism to encode diverse order and vehicle information into uniform and informative representations, enabling it to accurately identify the most suitable vehicles for each order. Extensive experiments using real-world datasets and four representative ridesharing algorithms demonstrate that PreMR significantly reduces order response time (by 46.78% on average) while maintaining high service quality, with a slight trade-off in the order completion rate.

The integration of 5G/6G networks with intelligent healthcare systems has enabled early disease detection through patient data monitoring. However, the Internet of Medical Things (IoMT) and remote healthcare services introduce significant privacy and security risks. In this paper, we propose LiteCrypt, which addresses these challenges by introducing an optimized Homomorphic Convolutional Neural Networks (HCNN) structure for secure inference and a lightweight Threshold Signature Scheme (TSS) based dual-authorization mechanism. To enhance the practicality of Homomorphic Encryption (HE)-based secure inference in telemedicine applications, LiteCrypt presents an optimized HCNN framework that ensures efficient and adaptable operations across multiple datasets. A high-performance GPU-accelerated HE engine is developed to address the computational demands of HE operations, enabling real-time processing of encrypted patient data. Besides, LiteCrypt introduces a novel TSS-based dual-authorization protocol, requiring consent from both the patient and the hospital to access patient data, thereby mitigating unauthorized access risks. The system adapts to a flexible 2-out-of-3 authorization scheme for emergencies, ensuring timely data retrieval while maintaining security. To overcome the initial challenge of prolonged computation time due to compute-intensive operations, In LiteCrypt, we utilized the lightweight TSS protocol, based on Oblivious Transfer (OT), which is designed for resource-constrained IoMT devices, reducing computation time from 11.9 to 0.11 seconds. Empirical validation demonstrates LiteCrypt’s superior performance, achieving a 233-fold increase in processing speed, a $96 \%$ reduction in encrypted message size, and a 28-fold speed increase using GPUs.

Keystroke dynamics has proven to be highly effective, with its applications expanding significantly over the years in areas such as preventing transaction fraud, account takeovers, and identity theft. Key-positioning and feature-learning methods are commonly used to identify keystroke signals. However, the existing methods face challenges in detecting overlapping keystrokes and environmentally changed signals. We propose a solution called Chameleon to address these limitations. Unlike previous signal separation and deep learning methods that are ineffective in keystroke signals and computationally demanding, Chameleon employs a low-computation Ranking Model to separate overlapping keystroke signals. Moreover, our experiments demonstrate that Chameleon separated signals can be recognized with an average accuracy of 92.69%, surpassing the commonly used FastICA method, which only reaches 25% accuracy. To account for environmental changes, we utilize the Fréchet Inception Distance (FID) as a guiding metric for model migration. Additionally, we introduce the Inductive Vector, which enables our key-identifying model to adapt to altered environmental conditions such as environment, phone location, and user variety. The Inductive Vector adjusts the model parameters based on the shift in FID. In scenarios with various phone locations, the Inductive Vector significantly improves recognition accuracy from 61% to 98%, outperforming the best existing keystroke recognition algorithm. In other dynamic environmental conditions, our approach achieves an average accuracy rate of 81.7%, which is at least 1.6 times better than the current state-of-the-art keystroke recognition algorithm.

Accurate fall detection systems are vital to address the global health concern of elderly falls, which often lead to severe injuries, hospitalizations, and fatalities. Since falls can happen at any time in any location, it is imperative to have a comprehensive system that boasts high applicability across a broad range of scenarios, operating seamlessly 24/7. However, within a range of fall detection systems, most of the existing work is built upon mono-modal sensors, which are inevitably inherited and constrained by mono-modal shortages. To overcome the constraints of mono-modal systems, we introduce VRFfall, a novel multi-modal fall detection system that seamlessly fuses mmWave radar and camera technologies. As a system with high generalization capabilities, VRFfall supports both multi-modal and mono-modal inputs with its independent feature extraction pipeline for each modality. Utilizing a cross-modal knowledge transfer design, VRFfall enhances performance with mono-modal input by leveraging fused knowledge from the other modality. Moreover, to ensure optimal fusion decisions under modal discrepancies, VRFfall incorporates an adaptive Modal Quality Assessment Module (MQAM) that dynamically evaluates and fuses features from both modalities. Extensive evaluations using a dataset collected from 20 volunteers across two environments and three conditions have been conducted on VRFfall. The results demonstrate its high performance and excellent generalization across diverse environments and conditions, promising a 24/7 continuous fall detection system.

With the advancement of the Internet of Medical Things (IoMT) and cryptographic technologies, remote healthcare services have become more widespread, presenting new challenges for patient privacy and data security. Conventional security mechanisms, such as centralized authentication and key distribution systems, are susceptible to single points of failure and significant management burdens, potentially leading to compromised authentication centers and internal security threats. In response, this study presents a threshold signature algorithm, it uses Distributed Key Generation (DKG) that distributes private keys without the need for a trusted key distributor, requiring the cooperative signature of at least two nodes for authentication. This approach not only circumvents the risk of single points of failure but also enhances the system’s robustness and efficiency. The experimental results validate its prospective utility in safeguarding remote healthcare data.

This paper proposes a novel scan-then-Iocalize scheme for the unmanned aerial vehicle (UAV) to localize a wireless sensor (WS). In both of the scan and the localizing phases, the UAV estimates its range and/or radial speed to the WS, by exploiting the orthogonal frequency division multiplexing (OFDM) sensing signal. We first derive the UAV's maximally allowable sensing radius to the WS to assure a sufficiently high correct estimation probability. Next, in the scan phase, we design the UAV's scan trajectory along a series of waypoints, where the number of the waypoints and their locations are found by solving the classic region least circle coverage problem. The scan phase ends when the UAV's estimated horizontal range to the WS becomes no larger than the maximally allowable sensing radius. The UAV then begins the localizing phase, where unlike the traditional three-measurement based localization, we propose that only two measurements of the UAV's range and radial speed to the WS are sufficient to localize the WS, but may generate a falsely-found ghost WS. By further identifying the UAV's improper flight angles that generate the ghost WSs, we newly propose the two-measurement based accurate localization algorithm. Finally, various simulation results are provided to validate the localizing performance of our proposed scheme.

Due to the advantages of acoustic sensing such as device ubiquity, hands-free interaction and privacy security, acoustic-based gesture input techniques have gained extensive at-tention and many excellent works have been proposed. However, these works have the following shortcomings: high cost of system construction, non-continuous input, and degraded performance in cross-user scenarios. To overcome the above shortcomings, we propose UltraWrite, a continuous gesture input system that needs rather low system construction cost, supports continuous input, and achieves high cross-user recognition performance. The key idea of our solution is to synthesize the data of continuous gestures from isolated ones, and build an end-to-end continuous gesture recognition model by introducing the connectionist temporal classification (CTC) mechanism widely used in natural language processing. We have implemented the system on a commercial tablet and conducted comprehensive experiments to evaluate its performance. The results demonstrate that UltraWrite achieves an average word accuracy of 99.3% and word error rate of 0.34% when considering only the first output candidate word. In addition, we have also evaluated the system's robustness to background noise, sensing distance and angle. The results reveal that UltraWrite displays strong robustness to these factors.

The rapid advancement of intelligent transportation systems necessitates efficient resource allocation for low-latency and high-bandwidth vehicular services. While traditional reinforcement learning has been widely utilized for resource allocation, it suffers from limitations such as poor generalization and interpretability. To overcome these challenges, we propose a novel Intelligence-based Reinforcement Learning (IRL) al-gorithm, which uses active inference to infer the real world and maintain an internal model of the world by minimizing free energy. We address the inefficiency of active inference by incorporating prior knowledge as macro guidance, ensuring more accurate and efficient training. By constructing the intelligence-based model, we eliminate the need for designing reward functions, which aligns better with human thinking and provides a method to reflect the learning, information transmission, and intelligence accumulation processes. Considering the dynamic and uncertain nature of vehicular scenarios, we apply the IRL algorithm to continuously evolving environments where environmental parameters are not fixed. Extensive simulations confirm the effectiveness of IRL, significantly enhancing the generalization and interpretability of intelligent models.

Numerous studies have extensively examined the living conditions of older adults who live alone, highlighting falls as the most significant threat to their well-being. Recognizing the critical importance of fall detection for older adults, researchers have conducted numerous investigations over the past decade to develop fall detection systems utilizing various detection methods. In this article, we further explore the potential of acoustic signals in detecting fall events and propose a novel system called FA-Fall, which fully harnesses acoustic signals for fall detection. Our approach revolves around fusing passive and active acoustic sensing through a pair of audio transceivers. We design a multimodal classification framework that integrates an attention mechanism and an anomaly detection mechanism to capitalize on the complementary and redundancy of passive and active sensing. We implemented the FA-Fall system and carried out extensive experiments to evaluate its performance. The results demonstrate that FA-Fall achieves an impressive overall accuracy of 98.97% under typical environmental conditions. Furthermore, it can still detect fall events with over 90% accuracy in challenging environments characterized by considerable background noise or nonline-of-sight conditions.

Fall detection and recognition play a crucial role in enabling timely medical interventions for people who are at risk of falls, especially among vulnerable populations like older adults and those with mobility limitations. In this article, a cost-effective integrated sensing and communication system, namely, FallDR, is presented for fall detection and recognition using ultrawideband communication. First, we collected the time of flight information of falls (four types) and nonfall events by 10 participants using FallDR. We then proposed a convolutional neural network incorporated with squeeze-and-excitation blocks to detect and recognize falls based on fall trajectories. It proves that the proposed model is accurate, energy-efficient, and lightweight to achieve 100% accuracy in fall detection and recognition. Our proposed solution is proven to be highly robust against environmental changes, such as interference, distance, and direction changes. Further tests in an office showed that FallDR could achieve nearly 100% accuracy, even when the environment was changed. FallDR efficiently employs the characteristics of fall trajectory and the advanced modeling ability of the neural network. We have published our archived data sets and code for comparisons and improvements.

Wireless sensing and communication evolved separately in the past. However, Integrated Sensing and Communication (ISAC) unlocks a new era of mobile network capabilities, with WiFi emerging as a prime candidate. By leveraging existing WiFi infrastructure and frequencies, ISAC enables powerful services like accurate localization and human activity recognition (HAR). WiFi-based HAR is a prime example powered by the magic of ISAC. WiFi Channel State Information (CSI) is susceptible to human movement disturbances; the alterations in CSI mirror the dynamic attributes of human activities. Given the intricate relationship between human activities and CSI, numerous deep learning models have been introduced to enhance HAR accuracy. Recently, transformer-based models have achieved excellent performance in various tasks, including speech recognition, natural language processing, and image classification. This has spurred research into incorporating transformer-based models into WiFi sensing applications. However, their application in WiFi-based HAR remains nascent. Vision transformer is well-suited for analyzing WiFi CSI signals in the form of spectra, such as the Doppler frequency spectrum frequently utilized in related studies, owing to its data structure mimicking that of images. In this study, we explored five widely used Vision Transformer architectures (vanilla ViT, SimpleViT, DeepViT, SwinTransformer, and CaiT) for WiFi CSI-based HAR using two publicly available datasets, UT-HAR and NTU-Fi HAR. Our work aims to assess and compare the performance of diverse ViT architectures for WiFi CSI-based HAR and provide guidelines for WiFi-based HAR modeling and ViT selection, considering accuracy, model size, and computational efficiency.

Maritime data transmission with unmanned aerial vehicles (UAVs) in maritime Internet of Things (MIoT) systems has received increasing attention due to its flexibility and low cost. To further improve the efficiency of maritime data transmission between the UAVs and the maritime buoys, optical communication is considered as a promising technique because of its low latency and high bandwidth. However, optical communication encounters the challenge of beam pointing alignment, particularly in maritime data transmission involving wave disturbance and drift of buoy, which deteriorates and even interrupts the line-of-sight (LOS) optical transmission. To tackle the challenge, this paper proposes DERLOC, a reliable data transmission solution based on deep reinforcement learning (DRL). We first provide the optimization analysis of reliable data transmission and formulate the data transmission procedure as a Markov decision process (MDP) aiming at maximizing the received signal intensity. Afterwards, we propose a beam pointing adjustment algorithm based on the soft actor-critic (SAC) approach to alleviate the performance deterioration caused by waves. Then, we analyze the drift characteristic of a buoy and develop a method which enables UAV to predict the position and determine an optimal movement control strategy for ensuring the effectiveness of beam pointing and maintaining stable LOS communication. Through extensive simulations and real-time data validation, the results demonstrate that DERLOC is effective and enables a reliable data transmission via optical links.

Recent research studies have made significant progress in acoustic-based gesture recognition. However, existing methods lack the capability to expand to customized gestures and adapt to different practical environments. We propose a highly scalable gesture recognition system called EchoGest which integrates a well-designed feature-wise transformation layer into prototypical network framework, and accomplishes unseen gesture recognition with a device’s built-in speaker and microphone. Our key insight involves gauging the similarity between query sample representations and class prototypes in the embedding space, and thus enabling the scalability to unseen gestures. Meanwhile, we introduce a feature transformation layer to linearly adjust feature maps and propose an efficient two-stage training strategy to obtain regularized parameters for this layer. Specifically, this layer employs affine transformation to enhance intermediate feature activations and yield more diverse feature distributions for cross-domain recognition, and it improves recognition accuracy by 10% in 1-shot cases. We train the system with a collected a letter gestures (i.e., writing ’A’ to ’Z’) dataset and test it on a digit gestures (i.e., writing ’0’ to ’9’) dataset with 10 volunteers. The results show that EchoGest can recognize unseen digit gestures with an accuracy of 93.7% in 2-shot cases, and 93.2% in the leave-one-user-out testing setting. We also explore a semi-supervised clustering approach in which each user’s data can be used to update his or her prototypes for personalized customization. The comprehensive experiments also verify that EchoGest remain good performance across various environments, age groups, and different devices.

Federated learning (FL) offers distributed machine learning on edge devices. However, the FL model raises privacy concerns. Various techniques, such as homomorphic encryption (HE), differential privacy, and multiparty cooperation, are used to address the privacy issues of the FL model. Among them, HE ensures greater security and privacy since end-to-end encryption maintains data privacy throughout the computation process. Compared with other privacy-preserving techniques, HE does not require the establishment of a trusted environment or protocol among multiple parties and does not involve any artificial noise that can impair system performance. Unfortunately, it suffers from efficiency overhead when applied to privacy-preserving FL (PPFL). Some existing surveys on PPFL discuss the generic construction and organization of PPFL from the perspective of practical HE deployment in PPFL. However, none of them covers the efficiency optimization of HE when applied to PPFL. This article conducts a comprehensive review of the efficiency optimization of HE when applied to PPFL. First, we review general optimization strategies and discuss their limitations when applied directly to HE-based PPFL. Second, an overview of algorithmic, hardware, and hybrid optimizations is provided, along with a discussion of their adaptation. Additionally, we provide a detailed taxonomy of optimizations. Finally, we suggest future HE-based PPFL research directions.

With the increasing pervasiveness of smart earphones, it is appealing to propose more unobtrusive and convenient wearable authentication methods. Researchers have designed earphone-based authentication systems which utilize high-frequency audio signals to scan ear canal structure. Nevertheless, they possess shortcomings of low unobtrusiveness and robustness. In this article, we put forward an earphone-based passive authentication system which makes use of physiological and behavioral acoustic signals caused by a user’s natural actions, including putting on earphones and inner organs’ activities, respectively. By introducing attention mechanism into the network design, our method adaptively weighs two channel signals, and extracts stable fingerprints for different people, which relieves model retraining for unseen users and improves its scalability. We have built a real-time prototype called EarPrint by designing the earphones and a mobile application, and conducted comprehensive experiments under diverse settings. Experimental results demonstrate that EarPrint has low false acceptance rate (FAR) and equal error rate (EER) less than 1% and 5% in most cases, respectively.

Abstract Estimating the population of defects is a key reference for completion evaluation in crowdsourced testing. Current studies commonly use the biological capture–recapture model (CRC model) to estimate the population of defects in crowdsourced tests with good results. However, the lack of consideration of false negatives for defects in existing studies leads to imprecise estimation of completion in crowdsourced testing, which in turn influences decision support during crowdsourced testing. This study analyzes the basic assumptions of the CRC model in biology and maps them to the field of crowdsourced testing, improves the Lincoln–Peterson estimator and sample coverage estimator utilizing false negative assumptions, and compares them with original estimators in 29 real crowdsourced testing projects. The experimental results show that our improved estimators can effectively reduce the estimation relative error of the population of defects caused by false negatives.

In unsupervised domain adaptation (UDA), a classifier is applied to the target domain without or with limited labels, when the target domain has no or few labels. Recently, inspired by their capabilities of long-distance feature dependencies, vision transformer (ViT)-based methods have been used in UDA, however, they ignore the fact that ViT lacks strength in extracting local feature details. To handle the above problems, the purpose of this article is to demonstrate how to take advantage of both convolutional operations and transformer mechanisms for adversarial UDA by using a hybrid network structure called transformer fused convolution (TFC). TFC integrates local features with global features to boost the representation capacity for UDA which can enhance the discrimination between foreground and background. Moreover, to improve the robustness of the TFC, we leverage an uncertainty penalty loss to make incorrect classes have consistently lower scores. Extensive experiments validate the significant performance gains compared to conditional adversarial domain adaptation (CDAN) on all five datasets including DomainNet ($\uparrow ~8.5$%), VisDA-2017 ($\uparrow ~14.9$%), Office-Home ($\uparrow ~18.9$%), Office-31 ($\uparrow ~11.5$%), and ImageCLEF-DA ($\uparrow ~5.5$%).

Biochemical information sensing has always been one of the challenges in ubiquitous sensing research for mobile computing. Microorganisms will cause undetectable deterioration in drink production, such as wine and beverage, and microbial contamination is highly susceptible during storage like some liquors can be bottled for sometimes over ten years. Microbial culture methods are common for quality monitoring but unsuitable for real-time beverage quality monitoring. As far as we know, we are the first to use ubiquitous sensing for real-time microbial contamination detection. We designed a lightweight monitoring system called Microbe-Radar, which uses light signals to monitor real-time beverage quality. Microbe-Radar uses eight LEDs and a photodiode to detect fine-grained surface tension and absorption spectrum changes caused by microbial metabolites and growth during deterioration. Characteristic offset degree measurement and absorption spectrum dimension expansion are two critical technologies. Moreover, we implemented countermeasures against ambient light noise and sloshing interference. Microbe-Radar's surface tension and absorption spectrum measurement errors are only 0.89 mN/m and 2.4%, respectively, making identifying the contamination duration, microorganism content, and microorganism composition worthwhile. Experiments showed Microbe-Radar could determine potential issues with liquor quality when the liquid becomes health-threatening or even just contaminated, with an accuracy of 97.5%. Microbe-Radar can also be extended to beverage deterioration warning, with deterioration prediction accuracy of more than 90.6% for five beverages (milk, apple juice, etc.).

Global Positioning System (GPS) has benefited many novel applications, e.g., navigation, ride-sharing, and location-based services, in our daily life. Although GPS works well in most places, its performance in urban canyons is well-known poor, due to the signal reflections of non-line-of-sight (NLOS) satellites. Tremendous efforts have been made to mitigate the impacts of NLOS signals, while previous works heavily rely on precise proprietary 3D city models or other third-party resources, which are not easily accessible. In this paper, we presentDeepGPS, a deep learning enhanced GPS positioning system that can correct GPS estimations by only considering some simple contextual information.DeepGPSfuses environmental factors, including building heights and road distribution around GPS's initial position, and satellite statuses to describe the positioning context, and exploits an encoder-decoder network model to implicitly learn the complex relationships between positioning contexts and GPS estimations from massive labeled GPS samples. As a result, the well-trained model can accurately predict the correct position for each erroneous GPS estimation given its positioning context. We further improve the model with a novel constraint mask to filter out invalid candidate locations, and enable continuous localization with a simple mobility model. A prototype system is implemented and experimentally evaluated using a large-scale bus trajectory dataset and real-field GPS measurements. Experimental results demonstrate thatDeepGPSsignificantly enhances GPS performance in urban canyons, e.g., on average effectively correcting 90.1% GPS estimations with accuracy improvement by 64.6%.

Ride-hailing service (RHS) has become an important transportation mode in our daily life. Although many works have been proposed to improve RHS from different aspects, only few works focus on the selections of pick-up locations, where rider and driver meet and start a trip. In this paper, we presentMPLRec, a data-driven pick-up location recommendation system that exploits riders' specific mobility demands,e.g.destination, and historical experiences to meet riders' travel requirements.MPLRecgenerates potential pick-up locations over the road network and characterizes them with rich features that describe a location from the riders' perspective. We also build spatio-temporal indexes to organize potential pick-up locations and historical data for facilitating online recommending. When processing an online recommendation request,MPLRecderives candidate pick-up locations and investigates them with materialized features, which are computed from historical order and trajectory data while considering rider's mobility demands. Based on these features, a novel scoring function is used to derive the best pick-up location for each request. Moreover, we implement an RHS simulator to evaluateMPLRecusing large-scale practical ride-hailing datasets. Extensive experiments and simulations demonstrate the effectiveness and efficiency ofMPLRec, which can complete each request within 0.5 s and largely reduce the ride-hailing costs when compared to baseline methods.

Human activity recognition (HAR) has become a research hotspot because of its wide range of application prospects. It has higher requirements for real-time and powerefficient processing. However, a large amount of data transfer between sensors and servers, and computation-intensive recognition models hinder the implementation of real-time HAR systems. Recently, edge computing has been proposed to address this challenge by moving computational and data storage resources to the sensors, rather than depending on a centralized server/cloud. In this paper, we investigated binary neural networks for edge intelligence-enabled HAR using radar point cloud. Point cloud can provide 3-dimensional spatial information, which is helpful to improve recognition accuracy. Time-series point cloud also brings challenges, such as larger data volume, 4-dimensional data processing, and more intensive computation. To tackle these challenges, we adopt the 2-dimensional histograms for point cloud multi-view processing and propose the EdgeActNet, a binary neural network for point cloud-based human activity classification on edge devices. In the evaluation, the EdgeActNet achieved the best results with average accuracies of 97.63% on the MMActivity dataset and 95.03% on the point cloud samples of the DGUHA dataset respectively; and saved 16.9× memory consumption and 11.5× inference time compared to its full-precision version. Our work also is the first to apply 2D histogram-based multi-view representation and BNNs for timeseries point cloud classification.

With the rapid growth of Internet of Things, the number of heterogeneous wireless devices working in the same frequency band increases dramatically, leading to severe cross-technology interference. To enable coexistence, researchers have proposed a large number of mechanisms to manage interference. However, existing mechanisms have severe modifications in either the physical or MAC (medium access control) layers, making them very different from the standard. In this paper, we design and implement SledZig to boost cross-technology coexistence for low-power devices through both enabling more transmission opportunities and avoiding interference. SledZig is fully compatible with the standard in both physical and MAC layers. It decreases the WiFi signal power on the channel of low-power devices while keeps the WiFi transmission power unchanged, through making constellation points on the overlapped subcarriers have the lowest power, which can be achieved by just encoding the WiFi payload. We implement SledZig on hardware testbed and evaluate its performance under different settings. Experiment results show that SledZig can effectively increase ZigBee transmissions and improve its performance over a WiFi channel under various WiFi data traffic, with as low as 6.94$\%$WiFi throughput loss.

This paper proposes a novel design on the wireless powered communication network (WPCN) in dynamic environments under the assistance of multiple unmanned aerial vehicles (UAVs). Unlike the existing studies, where the low-power wireless nodes (WNs) often conform to the coherent harvest-then-transmit protocol, under our newly proposed double-threshold based WN type updating rule, each WN can dynamically and repeatedly update its WN type as an E-node for non-linear energy harvesting over time slots or an I-node for transmitting data over sub-slots. To maximize the total transmission data size of all the WNs over$T$slots, each of the UAVs individually determines its trajectory and binary wireless energy transmission (WET) decisions over times slots and its binary wireless data collection (WDC) decisions over sub-slots, under the constraints of each UAV's limited on-board energy and each WN's node type updating rule. However, due to the UAVs’ tightly-coupled trajectories with their WET and WDC decisions, as well as each WN's time-varying battery energy, this problem is difficult to solve optimally. We then propose a new multi-agent based hierarchical deep reinforcement learning (MAHDRL) framework with two tiers to solve the problem efficiently, where the soft actor critic (SAC) policy is designed in tier-1 to determine each UAV's continuous trajectory and binary WET decision over time slots, and the deep-Q learning (DQN) policy is designed in tier-2 to determine each UAV's binary WDC decisions over sub-slots under the given UAV trajectory from tier-1. Both of the SAC policy and the DQN policy are executed distributively at each UAV. Finally, extensive simulation results are provided to validate the outweighed performance of the proposed MAHDRL approach over various state-of-the-art benchmarks.

Acoustic-based human gesture recognition (HGR) applications have drawn increasing academic attention in order to overcome the shortcomings of conventional interaction methods on tiny devices. Existing techniques following a learning-based routine requires collecting massive application-specific training data. What is worse, the cross-domain problem induces additional retraining overhead to enable the systems recognize unseen gestures in different environments. This obviously decreases their scalability and prevent them from real-world deployment. Although some recent works propose different few-shot learning solutions to deal with the cross-domain problem in HGR, they possess shortcomings of being application-specific, high training overhead, and/or incapability to recognize unseen gestures. In this paper, we propose PreGesNet, a few-shot acoustic gesture recognition framework based on task-adaptive pretrained networks whose novelty lies in three aspects: i) leveraging pretrained feature extractor which captures generic knowledge of our collected and open-source large-scale gesture datasets; ii) designing task-specific parameter adaptation mechanism to efficiently update the feature extractor to adapt the pretrained feature extractor to each target task; iii) discovering suitable distance metric and task generation strategy which fit HGR application. According to the experiments, when the model is trained with 10 digit gestures, its recognition accuracies of 26 kinds of letter gestures and 8 kinds of other hand gestures can be up to 80.5% and 93.4% with only two shots, respectively. In addition, the average recognition latency of PreGesNet is less than 0.4 second.

The use of speakers in electronic devices has become widespread, but the security risks associated with micro-speakers, such as earphones, are often overlooked. Many assume that soundproof barriers can prevent sound leakage and protect privacy. This article presents the prototype MagEar, an eavesdropping system that exploits magnetic side-channel signals leaked by a micro-speaker to restore intelligible human speech. MagEar outperforms some high-precision magnetometers in detecting magnetic fields at the nanotesla level. Even at a distance of 60 cm, it can recover high-quality audio with a 90% similarity to the original audio. Moreover, the MagEar prototype is portable and can be concealed within a headset housing. We have implemented MagEar as a proof-of-concept system and conducted multiple case studies on the eavesdropping of various speaker-embedded devices, including earphones. The recovered speech can be transcribed using automatic speech recognition techniques, even when obstructed by soundproof walls. It is our aspiration that our work can prompt manufacturers to reconsider the security vulnerabilities of speakers.

With the rapid and continuous development of data-driven technologies such as supervised learning, high-quality labeled data sets are commonly required by many applications. Due to the easiness of crowdsourcing small tasks with low cost, a straightforward solution for label quality improvement is to collect multiple labels from a crowd, and then aggregate the answers. The aggregation strategies include majority voting and its many variants, EM-based approaches, Graph Neural Nets and so on. However, due to the uncertainty information loss and commonly existing task correlations, the aggregated labels usually contain errors and may damnify the downstream model training.To address the above problem, we propose a hierarchical crowdsourcing framework1for data labeling with noisy answers about correlated data. We make use of the heterogeneity of the labeling crowd and form an initialization-checking-update loop to improve the quality of labeled data. We formalize and successfully solve the core optimization problem, namely, selecting a proper set of checking tasks for each round. We prove that maximizing the expected quality improvement is equivalent to minimizing the conditional entropy of the observations given the crowdsourced answer families for the selected task set, which is NP-hard to solve. Therefore, we design an efficient approximation algorithm and conduct a series of experiments on real data. The experimental results show that the proposed method effectively improves the quality of the labeled data sets as well as the SOTA performance, yet without extra human labor costs.

Graphs have been widely used for modeling large-scale data generated from real-world applications, while compact representation of such graphs is beneficial for efficient storage and effective graph analysis. As a promising solution, lossless graph summarization can compactly represent a given graph as a summary graph, which consists of supernodes (i.e., sets of nodes) and superedges (edges between supernodes), and the correction edge sets, which together with summary graph can exactly reconstruct the original graph. Although many research efforts have been devoted to develop graph summarization methods, existing works are still inefficient in terms of computation efficiency and representation compactness. To address their limitations, we propose optGS that includes a set of optimization techniques, including computation-oriented supernode re-dividing, degree-aware approximation metric for selecting the best merge, and redundant computation avoidance, to improve current advances. Extensive experiments on a variety of large graph datasets demonstrate the computation efficiency and compression effectiveness of our optGS, e.g., improving the representation compactness by up to 20.28% and achieving 3.53× speedup in running time than the state-of-the-art methods.

To implement high-speed wireless communication from the deep ocean to the sea surface leveraging multiple autonomous underwater vehicles (AUVs) with underwater wireless optical communication (UWOC) technique is an emerging and promising technology that enables real-time data collection for accurate underwater exploration and monitoring, e.g., coordinated moving targets tracking. However, multi-hop UWOC is more susceptible to beam misalignment and positional uncertainty caused by external interference in the harsh marine environments. To address these challenges, we design a cooperative movement scheme for multiple AUVs based on deep reinforcement learning (DRL) approach to realize robust and reliable optical communication under mobile target tracking scenarios. In this scheme, we first model the optical channel with optical noise and then analyze the link performance of multiple AUVs to meet the bit error rate (BER) requirements. Afterwards, we map the cooperative optical communication problem to a Markov decision process (MDP) by incorporating the extended Kalman filter (EKF) technique to enhance effective communication. Finally, we propose a cooperative control strategy for multiple mobile AUVs based on deep deterministic policy gradient (DDPG). Through extensive simulations, it is demonstrated that the proposed algorithm is effective in achieving reliable underwater optical communication under mobile scenarios.

In marine wireless sensor networks, swarms of unmanned aerial vehicles (UAVs) based optical communication system can be leveraged to transmit underwater real-time monitoring data which enables a variety of potential maritime missions, such as video streaming for underwater surveillance and target tracking. However, due to the high directionality of optical beams and the mechanical instability of UAVs caused by the wind, the optical link via UAVs as relays is very fragile. To deal with the challenge, in this study, we model the problem as a partially observed Markov decision process (POMDP), and propose a novel link maintenance scheme based on the cooperative multi-agent deep deterministic policy gradient (MADDPG) approach to control the swarms of UAVs, which integrates the UAV dynamics model and the optical communication model. In this scheme, multiple UAVs act as agents controlling their own states in real-time under the complex wind field, in order to keep mechanical stability through cooperation, and dynamically maintain the reliability of the optical links which maximize communication performance to achieve reliable end-to-end optical communication and reduce energy consumption. Through numerical simulations, it demonstrates that the proposed optimization scheme is effective and achieves robust performance in terms of communication quality and energy consumption.

Neglecting proper oral hygiene has proven to potentially lead to severe oral disease, resulting in complications over time. Careful brushing can mitigate the problem, but it is common for individuals to dedicate insufficient time to the various areas of their teeth. We propose LiT to monitor the brushing situation of 16 Bass technique surfaces in real-time. LiT relies on commercial toothbrushes with blue LEDs as a transmitter and requires only 2 low-cost photosensors as receivers on the toothbrush head. However, the transmission channel of light in the oral cavity is unclear. Finding the optimal deployment positions and minimizing the number of photosensors is challenging. To tackle these obstacles, we design the positioning of the 2 photosensors and create a transmission model within the oral cavity to verify the feasibility theoretically. Additionally, obstacles in implementation include separating brushing action accurately, interference of light on the outer surfaces of front teeth, and individual variability. To overcome these challenges, we develop corresponding technologies and a comprehensive framework. Experiments with 16 users show that LiT achieves a highly accurate recognition rate of 95.3% with an error estimate for brushing duration of 6.1%. Furthermore, LiT also proves resilient under user motion and environmental interference.

With the deployment of Internet of Things, we have seen many wireless devices working in the crowded unlicensed 2.4GHz band, leading to severe cross-technology coexistence problem. The sensing of signal type and channel can assist wireless devices with more efficient transmission strategy to improve the network performance. In this paper, we present DeepSpectrum, a wireless signal identification system based on convolutional neural networks, to identify the type and estimate the channel of four wireless signals, including Wi-Fi, Zigbee, Lora and Bluetooth. DeepSpectrum consists of a master model for the initial identification of the wireless signals, and an auxiliary model to further improve the accuracy of the Lora channel estimation. We thoroughly evaluate DeepSpectrum using various test data at different signal-to-interference ratio (SIR) levels. Our evaluation results show that at $0\mathrm{~dB}$ SIR, the average identification accuracy for the four signals is over $\mathbf{9 9 \%}$ and the average error of central frequency is only 21KHz, which indicates that we can get the channel of Wi-Fi, Zigbee, and Bluetooth correctly. As for Lora with flexible central frequencies, the average bandwidth error is only 76KHz.

The increasing popularity of earable devices stimulates great academic interest to design novel head gesture-based interaction technologies. But existing works simply consider it as a singular activity recognition problem. This is not in line with practice since users may have different body movements such as walking and jogging along with head gestures. It is also beneficial to recognize body movements during human-device interaction since it provides useful context information. As a result, it is significant to recognize such composite activities in which actions of different body parts happen simultaneously. In this paper, we propose a system called CHAR to recognize composite head-body activities with a single IMU sensor. The key idea of our solution is to make use of the inter-correlation of different activities and design a multi-task learning network to extract shared and specific representations. We implement a real-time prototype and conduct extensive experiments to evaluate it. The results show that CHAR can recognize 60 kinds of composite activities (12 head gestures and 5 body movements) with high accuracies of 97.0% and 89.7% in user- dependent and independent cases, respectively.

The rapid development of AI applications powered by deep learning in edge devices boosts the opportunity for real-time health monitoring. To address the potential privacy concern in the inference phase, homomorphic encryption (HE) is an alternative solution that encrypts inference data without exposing raw data and has several distinct advantages, (i.e., single-round communication, lightweight bandwidth consumption, and non-interactive computation). However, the computational overhead on the current HE-based privacy-preserving inference necessitates a substantial amount of time, which is not feasible for some real-time applications on edge devices. To address this issue, we propose CNN-guardian, a unified and compact neural network structure for real-time inference in HE-based inference on edge GPU. CNN-guardian designs a HE-friendly neural network and GPU engine that optimizes HE operations to accelerate the inference in the HE domain.

This paper proposes a new on-demand wireless energy transfer (WET) scheme of multiple unmanned aerial vehicles (UAVs). Unlike the existing studies that simply pursuing the total or the minimum harvested energy maximization at the Internet of Things (IoT) devices, where the IoT devices' own energy requirements are barely considered, we propose a new metric called the hungry-level of energy (HoE), which reflects the time-varying energy demand of each IoT device based on the energy gap between its required energy and the harvested energy from the UAVs. With the purpose to minimize the overall HoE of the IoT devices whose energy requirements are not satisfied, we optimally determine all the UAVs' trajectories and WET decisions over time, under the practical mobility and energy constraints of the UAVs. Although the proposed problem is of high complexity to solve, by excavating the UAVs' self-attentions for their collaborative WET, we propose the multi-agent graph reinforcement learning (MAGRL) based approach. Through the offline training of the MAGRL model, where the global training at the central controller guides the local training at each UAV agent, each UAV then distributively determines its trajectory and WET based on the well-trained local neural networks. Simulation results show that the proposed MAGRL-based approach outperforms various benchmarks for meeting the IoT devices' energy requirements.

Transforming physical surfaces into virtual interfaces can extend the interaction capability of many exciting metaverse applications in the future. Recent advances in vibration-based tap sensing show promise for this vision using passive vibration signals. However, current approaches based on Time-Difference-of-Arrival (TDoA) triangulation suffer the impact of fluctuant wave velocity due to the dispersive and heterogeneous nature of solid mediums, failing to meet the performance requirement for practical use. In this article, we present MM-Tap, a vibration-based tap localization system that can transform ubiquitous surfaces into virtual touch screens with low overhead. A novel localization scheme is proposed based on the finding of spatiotemporal mapping between tap locations and TDoA values, which pushes the accuracy limits of vibration-based tap sensing from unstable cm-level to mm-level. We investigate the geometry of the sensor layout and design a model-based method to synthesize tap data, which enables MM-Tap to adapt to various surface materials and respond to arbitrary sensing scales after a few seconds of calibration. We combine MM-Tap with a COTS projector and facilitate a digitally augmented surface where users can play video games with low latency.

Wearable devices equipped with a variety of sensors facilitate the measurement of physiological and behavioral characteristics. Activity-based person identification is considered an emerging and fast-evolving technology in security and access control fields. Wearables, such as smartphones, Apple Watch, and Google glass can continuously sense and collect activity-related information of users, and activity patterns can be extracted for differentiating different people. Although various human activities have been widely studied, few of them (gaits and keystrokes) have been used for person identification. In this article, we performed person identification using two public benchmark data sets (UCI-HAR and WISDM2019), which are collected from several different activities using multimodal sensors (accelerometer and gyroscope) embedded in wearable devices (smartphone and smartwatch). We implemented eight classifiers, including an multivariate squeeze-and-excitation network (MSENet), time-series transformer (TST), temporal convolutional network (TCN), CNN-LSTM, ConvLSTM, XGBoost, decision tree, and$k$-nearest neighbor. The proposed MSENet can model the relationship between different sensor data. It achieved the best person identification accuracies under different activities of 91.31% and 97.79%, respectively, for the public data sets of UCI-HAR and WISDM2019. We also investigated the effects of sensor modality, human activity, feature fusion, and window size for sensor signal segmentation. Compared to the related work, our approach has achieved the state of the art.

In conventional federated learning (FL), multiple edge devices holding local data jointly train a machine learning model by communicating learning updates with a centralized aggregator without exchanging their data samples. Owing to the communication and computation bottleneck at the centralized aggregator and inaccurate learning model caused by the non-independent and identically distributed (IID) data, we here consider a two-tier FL network, in which Internet of Things (IoT) nodes are the core clients that hold data, the model aggregators at the middle tier are the low altitude aerial platforms (UAVs), and the model aggregator at the top-most layer is the high-altitude aerial platform (UAV with relatively high altitude). Under the assumption that each IoT node has parallel computing ability, we study the energy-efficient computation and communication resource allocation in such a network within some time budget. Upon formulating the problem as an optimization problem, we solve the computation and communication resource allocation problems as the separate subproblems within a time frame, and then propose an iterative algorithm to solve the entire problem jointly. More specifically, we solve both the energy-efficient computation and communication resource allocation subproblems using the dual decomposition technique, and then apply a bisection search-based recursive technique to solve the entire energy efficiency problem jointly. Moreover, we propose offline and online client scheduling schemes that not only select the optimal edge nodes for association but also assign workload to each client based on the data quality and workload constraint. With real data, extensive simulations are conducted to verify the effectiveness of the proposed resource allocation scheme. The results further reveal that the learning performance not only is dependent on the computation and communication energy consumption of the FL process but also the model divergence weight owing to the non-IID data at client IoT nodes.

Clustering a big distributed dataset of hundred gigabytes or more is a challenging task in distributed computing. A popular method to tackle this problem is to use a random sample of the big dataset to compute an approximate result as an estimation of the true result computed from the entire dataset. In this paper, instead of using a single random sample, we use multiple random samples to compute an ensemble result as the estimation of the true result of the big dataset. We propose a distributed computing framework to compute the ensemble result. In this framework, a big dataset is represented in the RSP data model as random sample data blocks managed in a distributed file system. To compute the ensemble clustering result, a set of RSP data blocks is randomly selected as random samples and clustered independently in parallel on the nodes of a cluster to generate the component clustering results. The component results are transferred to the master node, which computes the ensemble result. Since the random samples are disjoint and traditional consensus functions cannot be used, we propose two new methods to integrate the component clustering results into the final ensemble result. The first method uses component cluster centers to build a graph and the METIS algorithm to cut the graph into subgraphs, from which a set of candidate cluster centers is found. A hierarchical clustering method is then used to generate the final set of$k$cluster centers. The second method uses the clustering-by-passing-messages method to generate the final set of$k$cluster centers. Finally, the$k$-means algorithm was used to allocate the entire dataset into$k$clusters. Experiments were conducted on both synthetic and real-world datasets. The results show that the new ensemble clustering methods performed better than the comparison methods and that the distributed computing framework is efficient and scalable in clustering big datasets.

Leveraging emerging mobile edge computing and 5G networks, researchers proposed to offload the 3D rendering of interactive applications (e.g. virtual reality and cloud gaming) onto GPU-based edge servers to reduce the user experienced latency. A task offloading problem arises, that is where to offload rendering tasks such that each user will experience tolerable delay and meanwhile the cost of used servers is minimized. The multi-dimensional resource sharing feature of rendering tasks makes the problem challenging. We formulate the task offloading problem into a boolean linear programming. We propose a sharing-aware offloading algorithm which decomposes the problem into two subproblems (user assignment and server packing) and solves them alternately and iteratively. We compare our algorithm with the one without resource sharing in consideration, and the simulations demonstrate that our method can effectively reduce cost as well as satisfy delay requirement.

With the rapid development of the Internet of Things technology, the concept of smart cities that aims to help residents improve their quality of life has raised much attention in several application areas. In the context of smart cities, the provision of point of interest (POI) recommendations become an important requirement because a wide range of POIs are available for urban dwellers. Location-based social networks (LBSNs) such as Foursquare and Gowalla provide a massive volume of user check-in records that can assist users in choosing new POIs. However, user trajectories are mostly sparse in the real world. For example, users only check in a few POIs, and this makes it difficult to provide recommendations based on limited history trajectories. Though some attempts have adopted auxiliary geographical information to enhance POI recommendation, they still encounter the following problems: 1) the geographical trajectories of users are usually sparse in real-world datasets; 2) users may be more interested in the remote POIs; and 3) the previous models inherently perform transductive learning that cannot handle well the recommendation of unseen users and POIs. To address these problems, we propose an inductive representation learning model (IRLM) for location recommendation. IRLM contains two parts, namely geographic feature extraction and inductive representation learning. IRLM first captures global geographical influences among POIs through a standard Gaussian mixture model (GMM). Then IRLM adopts an attention neural network for the recommendation. Experimental results indicate that our proposed model can achieve superior performance over state-of-the-art models.

Automatic building instance extraction from high-resolution (HR) remote sensing imagery (RSI) is crucial for urban planning and mapping. The dominant approaches are based on the full-supervised learning paradigm that requires a large number of labeled samples to train their models, which is very time-consuming and labor-intensive. To alleviate this problem, this study proposes a semi-supervised building instance extraction method that integrates teacher-student learning and pseudo-labeling to improve the building instance extraction from HR RSI. Specifically, the proposed method consists of three modules, the hybrid data augmentation (HDA) module, the pseudo label generation (PLG) module and the contour refinement (CR) module. The HDA module is designed to enrich the diversity of labeled samples to optimize the teacher model. The PLG module generates pseudo labels from unlabeled data, and to train student model with pseudo-labels. Finally, the CR module is designed to refine the contours of buildings. Experimental results on three challenging public datasets demonstrate that the proposed method achieves superior performance and exhibits great robustness at different proportions of labeled data and different building scenarios. This study provides a new approach for extracting building instances from HR RSI in scenarios with insufficient labeled samples, and a methodological reference for various applications of semi-supervised on RSIs.

Radar-based human activity recognition is attracting a wide range of interest from both industry and academia because of its through-wall ability, privacy-preserving capability, and device-free detection. Currently, most radar-based systems consider signal analysis and feature extraction in the frequency domain or the temporal domain independently without fusing them together. In this article, in order to model both frequency properties and temporal profiles of human activity, we proposed a spectro-temporal network (STnet) that integrates a temporal convolutional network (TCN) and a convolutional neural network (CNN). It can extract temporal patterns and micro-Doppler features from radar signals for human activity recognition. In the experiments, two radar sensors and one base station were used to build a low-power wireless radar sensor network. Fifteen activities were investigated in a real kitchen scenario by using this radar sensor network. Frequency spectrograms were obtained after signal processing using a short-time Fourier transform (STFT). They were further segmented using a short sliding window (2.5 s), which enables a very small latency. The proposed STnet achieved 99.64% overall accuracy (OA) in testing, which is superior to the other three networks that we implemented in this work. Our work also can be used as a generic solution to other sensor-based (wearable sensors, WiFi channel state information (CSI), etc.) activity recognition.

Federated learning (FL) is a new dawn of artificial intelligence (AI), in which machine learning models are constructed in a distributed manner while communicating only model parameters between a centralized aggregator and client internet-of-medical-things (IoMT) nodes. The performance of such a learning technique can be seriously hampered by the activities of a malicious jammer robot. In this paper, we study client selection and channel allocation along with the power control problem of the uplink FL process in IoMT domain under the presence of a jammer from the perspective of long-term learning duration. We map the interaction between the FL network and the jammer in each learning iteration as a Stackelberg game, in which the jammer acts as the leader and the FL network serves as the follower. We consider the client and channel selection as well as the power control jointly as the strategy of this game. Upon formulating the game, we find the joint best response strategy for both types of players by leveraging the difference of convex (DC) programming approach and the dual decomposition technique. Beside the availability of the complete information to both the players, we also study the problem from the perspective that the FL network knows the partial information of the other player. Extensive simulations have been conducted to verify the effectiveness of the proposed algorithms in the jamming game.

Personalizing the analysis for web search potentially improves the search experience. A good analytical model for web search should leverage not only collective wisdom but also individual characteristics. Most of the existing analytical models, however, focus on how to utilize the collective wisdom, from a crowd, for instance. In this paper, we address the problem of user-specific web search analysis by considering the so-called burstiness in web search, which captures the behavior of rare words appearing many times in a single document. We go beyond click graph and propose two probabilistic topic models, Topic Independence Model and Topic Dependence Model. The former adopts the assumption that the generation of query terms and URLs are topically independent, and the latter captures the coupling between search queries and URLs. We also capture the temporal burstiness of topics by utilizing continuous Beta distribution. Through a large-scale analysis of a real-life search query log, we observe that each user's web search trail enjoys multiple kinds of user-based unique characteristics. On a massive search query log, the new models achieve a better held-out likelihood than existing baselines, and they can also effectively reveal the latent evolution of topics on the corpus level and user-based level.

in this paper, we have made the first contribution by designing two right types of questions, namely Routing Query (RQ) to ask the crowd to decide the direction at each road intersection. Secondly, we propose a series of efficient algorithms to dynamically manage the questions in order to reduce the selection hardness within a limited budget. In particular, we show that there are two factors affecting the informativeness of a question: the randomness (entropy) of the question and the structural position of the road intersection. Furthermore, we extend the framework to enable multiple RQs per round. To ease the pain of the sample sensitiveness, we propose a new approach to reduce the selection hardness by reasoning on a so-called Selective Bayesian network. We compare our approach against several baselines, and the effectiveness and efficiency of our proposal are verified by the results in simulations and experiments on real-world datasets. The experimental results show that, even the Selective Bayesian Network provides only partial information of causality, the performance on the reduction of the selection hardness are dramatically improved, especially when the size of samples are relatively small.

Traditional liquid identification instruments are often unavailable to the general public. This paper shows the feasibility of identifying unknown liquids with commercial lightweight devices, such as a smartphone. The wisdom arises from the fact that different liquid molecules have various viscosity coefficients, so they need to overcome dissimilitude energy barriers during relative motion. With this intuition in mind, we introduce a novel model that measures liquids’ viscosity based on active vibration. The idea sounds straightforward, yet, it is challenging to build up a robust system utilizing the built-in accelerometer in smartphones. Practical issues include under-sampling, self-interference, and volume change impact. Instead of using machine learning techniques, we tackle these issues through multiple signal processing stages to reconstruct the original signals and cancel out the interference. Our approach achieved the liquid viscosity estimates with a mean relative error of 2.3% and distinguish 30 kinds of liquid with an average accuracy of 97.33%.

Characterizing human driver's driving behaviors from GPS trajectories is an important yet challenging trajectory mining task. Previous works heavily rely on high-quality GPS data to learn such driving style representations through deep neural networks. However, they have overlooked the driving contexts that greatly govern drivers' driving activities and the data sparsity issue of practical GPS trajectories collected at a low-sampling rate. Besides, existing works omit the cold start problem, where the newly joined drivers usually have insufficient data to learn accurate driving style representations. To address these limitations, we present an adversarial driving style representation learning approach, named$\mathtt {Radar}$. In addition to summarizing statistic features from raw GPS data,$\mathtt {Radar}$also extracts contextual features from three aspects of road condition, geographic semantic, and traffic condition. We exploit the advanced semi-supervised generative adversarial networks to construct our learning model. By jointly considering statistic features and contextual features, the trained model is able to efficiently learn driving style representations from practical GPS trajectory data. Furthermore, we enhance$\mathtt {Radar}$'s representation learning for drivers owning limited training data with some basic data augmentation strategies and a novel auxiliary driver based data augmentation method. Experiments on two benchmark applications,i.e., driver identification and driver number estimation, with a large real-world GPS trajectory dataset demonstrate that$\mathtt {Radar}$can outperform the state-of-the-art approaches by learning more effective and accurate driving style representations.

A wide diversity of sensors has been applied in human activity recognition. These sensors generate enormous amounts of data during human activity monitoring. The long-distance data traveling between sensors and servers increases the costs of bandwidth and latency. However, human activity recognition has a high demand for real-time processing. Recently, edge computing is surging to solve this problem by moving computation and data storage closer to the sensor devices, rather than relying on a central server/cloud. Edge servers are usually designed for low power, low cost, and low computation. They do not support computation-intensive deep learning algorithms or will result in high latency. Fortunately, the development of binarized neural networks enables edge intelligence which supports AI running at the network edge for real-time applications. In this paper, we implement a binarized neural network (BinaryDilatedDenseNet) to enable low-latency and low-memory human activity recognition at the network edge. We applied the BinaryDilatedDenseNet on three sensor-based human activity recognition datasets and evaluated it with four metrics. In comparison, the BinaryDilatedDenseNet outperforms the related work and other three binarized neural networks in accuracy and saves 10 memory and 4.5--8 inference time compared to the FPDilatedDenseNet(the full-precision version of the BinaryDilatedDenseNet).

Mobile Crowdsensing (MCS), which assigns outsourced sensing tasks to volunteer workers, has become an appealing paradigm to collaboratively collect data from surrounding environments. However, during actual task implementation, various unpredictable disruptions are usually inevitable, which might cause a task execution failure and thus impair the benefit of MCS systems. Practically, via reactively shifting the pre-determined assignment scheme in real time, it is usually impossible to develop reassignment schemes without a sacrifice of the system performance. Against this background, we turn to an alternative solution, i.e., proactively creating a robust task assignment scheme offline. In this work, we provide the first attempt to investigate an important and realisticRoBustTaskAssignment (RBTA) problem in MCS systems, and try to strengthen the assignment scheme's robustness while minimizing the workers’ traveling detour cost simultaneously. By leveraging the workers’ spatiotemporal mobility, we propose an assignment-graph-based approach. First, an assignment graph is constructed to locally model the assignment relationship between the released MCS tasks and available workers. And then, under the framework of evolutionary multi-tasking, we devise a population-based optimization algorithm, namelyEMTRA, to effectively achieve adequate Pareto-optimal schemes. Comprehensive experiments on two real-world datasets clearly validate the effectiveness and applicability of our proposed approach.

User authentication and identification on smart devices has great significance in keeping data privacy and recommending personalized services. With the rising popularity of smart earphones recently, they open up a new world for users to enjoy music individually, but also bring about privacy concerns at the same time. Existing few research works propose positive sensing systems that emit and receive inaudible acoustic signals to authenticate users. However, they share shortcomings of intrusiveness to users, high power consumption, and purely focusing on authentication. Instead, in this paper, we propose a passive sensing system called${{\sf EarID}}$with low-cost customized earphones which attains user authentication and identification at once. It makes use of a embedded microphone to sense body sounds spread out through ear canals and extract ‘fingerprints’ as a novel biometric feature. With self-designed earphones, we design a deep learning-based real-time data processing pipeline and cope with external interference. Extensive experiments under different real-world settings show that${{\sf EarID}}$can achieve a rather low false acceptance rate of$3.4\%$for user authentication and a high$F1$score of$95.5\%$for legitimate user identification.

Energy efficiency is an important criterion to judge the quality of mobile apps, but one third of our arbitrarily sampled apps suffer from energy issues that can quickly drain battery power. To understand these issues, we conduct an empirical study on 36 well-maintained apps such as Chrome and Firefox, whose issue tracking systems are publicly accessible. Our study involves issue causes, manifestation, fixing efforts, detection techniques, reasons of no-fixes, and debugging techniques. Inspired by the empirical study, we propose a novel testing framework for detecting energy issues in real-world mobile apps. Our framework examines apps with well-designed input sequences and runtime context. We develop leading edge technologies, e.g., pre-designing input sequences with potential energy overuse and tuning tests on-the-fly, to achieve high efficacy in detecting energy issues. A large-scale evaluation shows that 90.4% of the detected issues in our experiments were previously unknown to developers. On average, these issues can double the energy consumption of the test cases where the issues were detected. And our test achieves a low number of false positives. Finally, we show how our test reports can help developers fix the issues.

The inevitable aging trend of the world’s population brings a lot of challenges to the health care for the elderly. For example, it is difficult to guarantee timely rescue for single-resided elders who fall at home. Under this circumstance, a reliable automatic fall detection machine is in great need for emergent rescue. However, the state-of-the-art fall detection systems are suffering from serious privacy concerns, having a high false alarm, or being cumbersome for users. In this article, we propose a device-free fall detection system, namely G-Fall, based on floor vibration collected by geophone sensors. We first decompose the falling mode and characterize it with time-dependent floor vibration features. By leveraging Hidden Markov Model (HMM), our system is able to detect the fall event precisely and achieve user-independent detection. It requires no training from the elderly but only an HMM template learned in advance through a small number of training samples. To reduce the false alarm rate, we propose a novel reconfirmation mechanism using Energy-of-Arrival (EoA) positioning to assist in detecting the human fall. Extensive experiments have been conducted on 24 human subjects. On average, G-Fall achieves a 95.74% detection precision on the anti-static floor and 97.36% on the concrete floor. Furthermore, with the assistance of EoA, the false alarm rate is reduced to nearly 0%.

Goal-oriented dialog policy learning algorithms aim to learn a dialog policy for selecting language actions based on the current dialog state. Deep reinforcement learning methods have been used for dialog policy learning. This work is motivated by the observation that, although dialog is a domain with rich contextual knowledge, reinforcement learning methods are ill-equipped to incorporate such knowledge into the dialog policy learning process. In this paper, we develop a deep reinforcement learning framework for goal-oriented dialog policy learning that learns user preferences from user goal data, while leveraging commonsense knowledge from people. The developed framework has been evaluated using a realistic dialog simulation platform. Compared with baselines from the literature and the ablations of our approach, we see significant improvements in learning efficiency and the quality of the computed action policies.

Recent studies show that deep learning models perform well in many medical tasks such as medical imaging and automated diagnosis. With qualified training datasets, some models can achieve or even surpass expert-level performance on some tasks. However, as a typical black-box-style approach, deep learning lacks theoretical interpretability, which is especially important for medical tasks. On the other hand, there are many sources of domain knowledge for medical diagnosis from human experts, such as clinical guidelines. How to sufficiently integrate human knowledge in the model is crucial for explainable diagnosis. In this paper, we propose a novel framework for explainable automated diagnosis that leverages explicit medical knowledge. We automate the knowledge extraction from textual clinical guidelines with prompt-based learning, train a set of weighted first-order logical rules with constructed evidence database, and finally infer the diagnosis result with integrated knowledge and multi-sourced data. We instantiate the framework for pulmonary disease diagnosis, and our experiments on a real dataset show that our method outperforms the state-of-the-art baselines in accuracy and interpretability.

Automatic fall detection and recognition are challenging problems. In this paper, a novel solution is proposed based on the trajectories of human falls by using the ultra-wideband (UWB) communication system and machine learning methods for fall detection and recognition. Most previous studies of fall detection based on active UWB sensing used electromagnetic signals directly, which may bring problems like radar clutter, signal coupling, multi-path, fading, and interference. Our proposed method only uses human falls trajectories by passive UWB sensing, which achieved fall recognition performance of 93.26% by using the support vector machine with RBF kernel function (SVM-RBF). Compared with previous research, the superiority of this study is that our solution is robust against interference and environmental changes, which means it is reliable for real-world applications. The archived UWB datasets and code have been already published, which may provide the basis for the comparison of techniques and improvements.

The predominant user activities on mobile architectures (e.g., smartphones) involve entering text in instant messaging apps, short message services, and social networking services. Recent research reveals that the normal use of input methods drains approximately half of the battery capacity due to their above-average power requirements and frequent use. In this paper, we first study the power characteristics of mobile input methods and find that they consistently over-provision resources to satisfy users. For example, the psychophysical evidence available indicates the response of spell-checking features within a time threshold makes users feel they have instant feedback. However, current systems perform it very quickly, which is imperceptible to users and costly in terms of energy use. Given this over- provisioning, the system can be slowed down to save energy while retaining the feeling of instant response. Inspired by this observation, we also exploit several other psychophysical facts to identify the exact criteria to satisfy users. As a result, we present a user experience-oriented technology, utexia, to optimize the energy use of mobile input methods. The evaluation shows that utexia conserves up to 42.9% in energy use while strictly ensuring a good user experience.

In this paper, we consider the anti-jamming problem for the uplink scenario in federated learning (FL) networks under the uncertainty condition of jamming channels. We first consider a single-carrier single-user FL network, in which the channel gain of the jammer is known to the FL user in probabilistic terms. We formulate the interaction between the FL user and the jammer by a Bayesian Stackelberg game. Upon the derivation of the best response strategy for both the players, we prove the existence and uniqueness of the equilibrium state. We further extend the game formulation for the multi-user multi-channel case, in which multiple FL users access the global aggregator using multiple channels. Similar to the single-user case, in this game, the jammer acts as the leader and the FL network works as the follower. Since the derivation of the optimal strategy is intractable due to the non-convexity nature of the problem, we provide a suboptimal algorithm that provides an equilibrium state for both the players. Extensive simulations on real data set are conducted to verify the effectiveness of the proposed games for both the single and multi-user cases. The results reveal that the priori probabilities of jamming channels greatly affect the strategies of both the players as well as the equilibrium state.

With the rapid growth of Internet of Things, the number of heterogeneous wireless devices working in the same frequency band increases dramatically, leading to severe cross-technology interference. To enable coexistence, researchers have proposed a large number of mechanisms to manage interference. However, existing mechanisms have severe modifications in either the physical or MAC (medium access control) layers, making them hard to be deployed on commercial devices. In this paper, we design and implement SledZig to boost cross-technology coexistence for low-power devices through both enabling more transmission opportunities and avoiding interference. SledZig is fully compatible with the standard in both physical and MAC layers. It decreases the WiFi signal power on the channel of low-power devices while keeps the WiFi transmission power unchanged, through making constellation points in the overlapped subcarriers have the lowest power, which can be achieved by just encoding the WiFi payload. We implement SledZig on hardware testbed and evaluate its performance under different settings. Experiment results show that SledZig can effectively increase ZigBee transmissions and improve its performance over a WiFi channel under various WiFi data traffic, with as low as 6.94% WiFi throughput loss.

Knowledge graphs (KGs) have been widely used in recent years and proved to be useful in many applications. However, large scale KGs that are collected by crowdsourcing or constructed by automated algorithms, are usually incomplete. Most existing works focus on how to complete the KG based on embedding learning, but ignore the semantic properties of the relations in KGs. In this paper, we propose a knowledge-enhanced approach for KG embedding learning that leverages the explicit knowledge about the relation properties. The knowledge is used to enhance the training data and guide the model on data selection during each training iteration. We conduct a series of experiments on datasets WN18RR and FB15k-237. The experiment results show that our method outperforms the previous state-of-the-art models.

Knowledge graphs (KGs), either constructed automatically from texts or collected manually from crowdsourcing workers, may contain uncertainty. The uncertainty may propagate into the knowledge graph embedding and downstream tasks, which is potentially harmful, especially for those confidencesensitive applications such as medical diagnostic suggestion. Crowdsourcing workers with domain knowledge can help improve the data quality of knowledge graphs, by knowledge checking. However, due to the large scale of knowledge graphs and the limitation of adequate crowdsourcing workers, it is unrealistic to check all triplets in a knowledge graph to improve the data quality. Therefore, in this paper, we propose a crowdsourcing framework that efficiently improves the confidence of knowledge graphs with limited budget. We instantiate the framework in the medical domain and conduct a series of experiments with realworld medical data. We deploy the framework for knowledge graph embedding UKGE and corresponding downstream tasks. The experimental results show that the proposed method efficiently improves the quality of the knowledge graphs, and hence improves the performance of probabilistic knowledge graph embedding in the downstream tasks.

Optical wireless communication (OWC) is an emerging technology for direct communication through the water-air interface. However, due to the high directionality of optical beams and the harsh oceanic environment, it faces significant challenges to achieve the alignment and preserve the link availability, as the waves cause beam deflection and the mobility of the transceivers makes the link worse. To tackle these challenges and achieve reliable optical communication between autonomous underwater vehicles and unmanned aerial vehicles, we propose a deep reinforcement learning algorithm assisted by an extended Kalman filter to solve the alignment issue. To improve the reliability of communication, we present an algorithm to obtain the optimal beam divergence angle to maximize the link availability. The numerical simulations demonstrate that the proposed scheme achieves better performance in terms of energy consumption and alignment accuracy, and the link availability is increased by 25% compared to that without adjustment.

Speakers have been widely embedded in various electronic devices as a standard configuration. The security vulnerability of microspeakers (such as earphones) is commonly overlooked because it is often assumed that soundproof boundaries, such as walls, can prevent privacy-infringing sound leakage. In this paper, we present the prototype MagEar, an eavesdropping system that leverages magnetic side-channel signals leaked by a microspeaker to recover intelligible human speech. MagEar has sufficiently high sensitivity to detect magnetic fields on the order of nanotesla, exceeding some high-precision magnetometers. It can recover high-quality audio with 90% similarity to the original audio even at a distance of 60 cm. In addition, the MagEar prototype is portable and can be hidden in a headset shell. We have implemented MagEar as a proof-of-concept system and conducted several case studies of eavesdropping on different types of speaker-embedded devices, including earphones, and we have demonstrated the ability to successfully transcribe the recovered speech using automatic speech recognition techniques even when blocked by soundproof walls. We hope that our work can push manufacturers to rethink this security vulnerability of speakers.

This paper presents a breakthrough framework for smartphone user authentication by analyzing the physiology and behavior of articulation, namely the vocal tract, tongue position, and lip movement, to expose individual uniqueness during uttering. The main idea is to leverage a smartphone's speaker and microphone to transmit and receive speech and ultrasonic signals, construct identity-related features, and determine whether the samples are legitimate users or attackers. The proposed system requires a smaller number of samples by utilizing single utterances resulting that the system resisting playback and mimicry attacks with an average accuracy of 99% in three different scenarios.

Abstract This paper applies the wireless powered communication network (WPCN) to an indoor communication system with two energy harvesting (EH) enabled users. Unlike the existing WPCN works designed for outdoor communications, where each user harvests energy only from the signals transmitted by a dedicatedly deployed hybrid access point (H‐AP), due to the short device‐to‐device distance in the indoor scenario, each user additionally harvests a sufficient amount of energy from the information signals transmitted by the H‐AP and the other user. First, the joint downlink and uplink throughput are maximized for the wireless powered indoor communication system. This problem is non‐convex. Thus, the authors manage to transform this problem into a convex one and solve it using convex optimization techniques. The solutions reveal that the total throughput increases largely over a reduced device‐to‐device distance due to the resultant harvested energy from both energy and information signals at each user. However, an unfair downlink versus uplink rate allocation phenomenon is observed. Thus, considering the importance of uplink communication quality for various indoor applications, a new problem is further proposed to maximize the uplink sum‐throughput over both users with an additional constraint to ensure a sufficient downlink rate. This problem is also shown to be non‐convex and is solved optimally by using a method similar to that in the first problem. Numerical results demonstrate the effectiveness of the proposed approach for improving the downlink versus uplink rate allocation fairness in the indoor wireless‐powered communication system.

In this article, a cost-effective ultrawideband (UWB) communication system for gesture recognition in a smart home environment is proposed, which uses gesture trajectories and a deep learning model. Most previous studies of gesture recognition using the UWB technology used electromagnetic signals directly, which may bring problems, such as radar clutter, signal coupling, multipath, fading, and interference. However, instead of using UWB’s high-frequency pulse signals, the proposed method only uses gesture trajectories by data positioning. To this end, first, a data set of four gesture activities was created. Then, this data set was trained using a convolutional neural network (CNN) integrated with a squeeze-and-excitation (SE) block, namely, the SE-Conv1D model. Finally, the system was prototyped to interact with appliances in practical smart homes. The experimental data was used to demonstrate the superiority of the SE-Conv1D model in comparison with four baselines: 1) support vector machines; 2)$K$-nearest neighbor; 3) random forest; and 4) binarized neural networks. Experimental results show that all collected gesture activities are correctly recognized with an overall accuracy of over 95%, among which the proposed SE-Conv1D model achieves the best accuracy of 99.48%. The proposed system is a complete end-to-end sensing system specifically designed for tracking and recognizing human gestures, which is robust against interference and changes in distance or direction. In addition, the proposed system can tackle the device selection problems for smart homes, which means it is reliable for real-world applications.

Due to the wide availability of taxis in a city and the tremendous benefits of ridesharing, taxi ridesharing becomes promising and attractive in recent years. Existing taxi ridesharing schemes simply match ride requests and taxis based on partial trip information and omit the offline passengers, who will hail a taxi at the roadside without submitting the ride requests to the system. Thus, they are still not efficient and practical. In this article, we consider the mobility-aware taxi ridesharing problem and presentmT-Shareto address these limitations.mT-Sharefully exploits the mobility information of taxis and ride requests to achieve efficient indexing of taxis/requests and better passenger–taxi matching, while still satisfying the constraints on passengers’ deadlines and taxis’ capacities. Specifically,mT-Sharemakes use of both geographical information and travel directions to index taxis and ride requests and supports the shortest path-based routing and probabilistic routing to serve both online and offline ride requests. In addition,mT-Shareproposes a novel payment model to share the ridesharing benefits among the taxi driver and passengers. Extensive evaluations using a large real-world taxi data set demonstrate the efficiency and effectiveness ofmT-Share, which can respond each ride request in milliseconds and be with moderate detour costs and passengers’ waiting time. Compared to state-of-the-art schemes,mT-Sharecan serve 42% and 62% more ride requests in peak and nonpeak hours, respectively. Furthermore,mT-Sharecan save 8.6% taxi fare for passengers and meanwhile increase 7.8% incomes for taxi drivers, when compared with the regular taxi services.

With the increasing demand for security in many sectors, such as defense and health systems, developing secure Internet of Things (IoT) networks is a matter of great urgency. Looking at a potential solution for secure IoT systems, we investigate the physical layer security of cooperative nonorthogonal multiple access (NOMA) systems. After decoding information signal, the idea that a strong IoT node can serve as a relay node for other weak IoT nodes in enhancing their signal reception reliability, is known as cooperative NOMA. We consider both single-antenna and multiantenna aided transmission scenarios, where the base station (BS) communicates with two IoT nodes of different strengths. In the multiantenna scenario, artificial noise (AN) is generated at the BS and the strong IoT node for improving the security of the system. In order to characterize the secrecy performance, we derive new exact expressions of the security outage probability for both the IoT nodes under both the single-antenna and multiantenna aided scenarios. For the single-antenna scenario, we show that the power optimization at the BS and the strong IoT node can enhance the secrecy performance to some extent. For this case, we further study the secrecy diversity order of the overall system, which is mainly determined by the IoT node with the worse channel condition. For the multiantenna scenario, we derive the asymptotic secrecy outage probability (SOP) when the number of antennas tends to infinity. Extensive simulations have been conducted to verify the accuracy and effectiveness of the proposed analytical derivations. The presented results verify that the security performance of the cooperative NOMA-based IoT network can be improved through an appropriate power control scheme and by generating AN at the BS and the strong IoT node. The simulation results further illustrate that the asymptotic SOP is close to the exact one.

Unmanned aerial vehicles (UAVs) are one of the effective means to provide emergency communication services in post-disaster areas. In this article, we consider data dissemination in post-disaster areas, where all Internet of Things (IoT) nodes may not have data needs all the time. The energy consumption in data dissemination is one of the key metrics to pay attention to since the charging facilities for UAVs may be limited due to the destruction of existing infrastructure. In addition, UAVs have limited endurance or lifetime, so unnecessarily flying over IoT nodes that may not have data is time consuming. Therefore, given the energy budget and data requirements of the IoT nodes, we formulated a data dissemination problem using multiple UAVs in a post-disaster area while optimizing their trajectory, mission completion time, and energy consumption. After time discretization, the formulated problem is a mixed-integer nonconvex problem and thus difficult to solve in general. For this reason, we jointly use the bisection search technique and the block coordinate descent (BCD) method to solve the entire problem while aiming to optimize the trajectory and mission completion time of the considered UAV as well as the overall energy consumption. In each iteration of the BCD method, we solve the user association, trajectory optimization, and power optimization subproblems one after the other in an alternating fashion. To solve each subproblem, we employ the geometric programming (GP)-based optimization technique that transforms the variables and constraints. Regarding the initial trajectory of the UAV, we utilized dynamic programming techniques based on unsaturated data requirements of IoT nodes. We performed extensive simulations in many realistic environments to verify the effectiveness and efficiency of the proposed data dissemination scheme in post-disaster scenarios.

Node localization is one of the major challenges that exist in underwater communication. Various techniques exist for terrestrial networks, while few of them are applicable in underwater networks due to the dynamic characteristic of the underwater channels, e.g., the lack of global positioning system (GPS) coverage under the water surface. Moreover, assorted environmental properties affect almost all employed communication techniques. In this letter, we propose an environmental-aware positioning system by considering the variations of the underwater speed of sound according to the dynamic changes in the physical properties of the seawater, such as temperature, salinity, and pressure, besides the internal waves’ effects. The proposed system employs the received signal strength (RSS) technique in estimating the distances between the network nodes. Moreover, we examine the application of various dynamic responses neural networks (NNs) in predicting the underwater node position, such as the feedforward, recurrent, time delay, and distributed delay NNs. The results show that the NN-based prediction models enhance the performance of the positioning system and could achieve small prediction errors in the range of 0.002 for both training and testing patterns.

Limited by size, shape, and other factors, it is rather inconvenient to interact with new smart devices by traditional methods. Acoustic-based methods following a machine learning approach have been put forward to resolve this problem in previous works. But they possess limitations of heavy training overhead, low performance for unseen users, and intensive computation cost. Following our previous work in this area, we further overcome shortcomings of existing work and propose a lightweight and zero-shot text-entry system for unseen users based on acoustic sensing. The key novelty of this work is proposing a new model training strategy including dataset construction and augmentation methods to effectively enhance generalization ability of a simple learning model with as few training data as possible, based on our insight into the problem. We design and implement a real-time Android application system called EchoWrite 2.0 to validate our idea with extensive experiments. Results show that EchoWrite 2.0 can recognize digits, English letters, and words with an accuracy of 85.3%, 73.2%, and 96.9%, respectively, for unseen users without providing any data to the learning model. The comparison with related work in different aspects shows overall superiority of EchoWrite 2.0 .

Proactive taxi dispatching is of great importance to balance taxi demand-supply gaps among different locations in a city. Recent advances primarily rely on deep reinforcement learning (DRL) to directly learn the optimal dispatching policy. These works, however, are still not sufficiently efficient because they overlook several pieces of valuable context information. As a result, they may generate quite a few improper actions and introduce unnecessary coordination costs. To improve existing works, we presentCOX– a context-aware taxi dispatching approach that incorporates rich contexts into DRL modeling for more efficient taxi reallocations. Specifically, rather than simply dividing the service area into grids,COXproposes a road connectivity aware clustering algorithm to divide the road network graph into zones for practical taxi dispatching. In addition,COXcomprehensively analyzes zone-level taxi demands and supplies through accurate taxi demand prediction and timely updates of taxi statuses.COXimproves the DRL modeling by integrating these derived contexts,e.g., state representation with complete demand/supply data and sequential action generation with full coordination among idle taxis. In particular, we implement an environment simulator to train and evaluateCOXusing a large real-world taxi dataset. Extensive experiments show thatCOXoutperforms state-of-the-art approaches on various performance metrics,e.g., on average improving the total order values by 6.74%, while reducing the number of unserved taxi orders and passengers’ waiting time by 4.92% and 44.84%, respectively.

Since inaccuracies commonly exist in many applications, data uncertainty has become an important problem in database systems. To deal with data uncertainty, probabilistic databases can be used to store uncertain data, and querying facilities are provided to yield answers with confidence. However, the results from a query or mining process may not be reliable when the uncertainty propagates in the systems. In this paper, we leverage the power of crowdsourcing by designing a set of Human Intelligence Tasks, or HITs in short, to ask a crowd to improve the quality of uncertain data. In particular, we consider crowds consists of workers with diverse accuracy rates when answering the HITs. We design solutions to maximize the data quality with minimal number of HITs. There are two obstacles for this non-trivial optimization, which lead to very high computational cost for selecting the optimal set of HITs. First, members of a crowd may return incorrect answers with different probabilities. Second, the HITs decomposed from uncertain data are often correlated. We have addressed these challenges in this paper by designing an effective approximation algorithm and an efficient heuristic solution, especially for crowds with diverse individual accuracy rates. To further improve the efficiency, we derive tight lower and upper bounds for effective filtering and estimation. Extensive experiments on both a simulated crowd and a real crowdsourcing platform are conducted to evaluate our solutions.

Facial expressions are an essential form of human nonverbal communication. Recognition of this nonverbal sign may enable developers to understand the feedbacks on smart device functionality and advertising. Existing approaches for facial expression recognition are mainly based on cameras or on-body sensors, which are either sensitive to lighting conditions or cumbersome for users to wear devices on their faces. In this paper, we propose a new facial expression recognition system based on Wi-Fi signals, named WiFace. Our fundamental intuition is that facial muscle movements in different expressions will induce distinctive waveform patterns in the time-series of channel state information (CSI) in Wi-Fi signals. We develop a series of algorithms to process the CSI signals and extract the most representative waveform patterns for facial expression classification. We build a fully-functional prototype of WiFace using commercial off-the-shelf devices, which can recognize six typical facial expressions. We conduct extensive experiments to evaluate the performance of WiFace, and the experimental results show that the average recognition accuracy is 94.80 percent.

The coexistence of heterogeneous devices in wireless networks brings a new topic on cross-technology communication (CTC) to improve the coexistence efficiency and boost collaboration among these devices. Current advances on CTC mainly fall into two categories,physical-layer CTCandpacket-level energy modulation(PLEM). Thephysical-layer CTCachieves a high CTC data rate, but with channel incompatible to commercial devices, making it hard to be deployed in current wireless networks. PLEM is channel and physical layer compatible, but with two main drawbacks of the low CTC data rate and MAC incompatibility, which will induce severe interference to the other devices’ normal data transmissions. In this paper, we propose symbol-level energy modulation (SLEM), the first CTC method that is fully compatible with current devices in both channel and the physical/MAC layer processes, having the ability to be deployed in commercial wireless networks smoothly. SLEM inserts extra bits to WiFi data bits to generate the transmitting bits, so as to adjust the energy levels of WiFi symbols to deliver CTC information. We make theoretical analysis to figure out the performance of both CTC and WiFi transmissions. We also conduct experiments to demonstrate the feasibility of SLEM and its performance under different network situations.

Network embedding (NE) aims to encode the relations of vertices into a low-dimensional space. After NE, we can obtain the learned vectors of vertices that preserve the proximity of network structures for subsequent applications, e.g., vertex classification and link prediction. In existing NE models, they usually exploit the skip-gram with a negative sampling method to optimize their objective functions. Generally, this method learns the vertex representation only from the local connectivity of vertices (i.e., neighbors). However, there is a larger scope of vertex connectivity in real-world scenarios: a vertex may have multifaceted aspects and should belong to overlapping communities. Taking a social network as the overlapping example, a user may subscribe to the channels of politics, economy, and sports simultaneously, but the politics share more common attributes with the economy and less with the sports. In this article, we propose an adversarial learning approach (ACNE) for modeling overlapping communities of vertices. Specifically, we map the association between communities and vertices into an embedding space. Moreover, we take further research on enhancing our ACNE with the following two operations. First, in the initialization stage, we adopt a walking strategy with perception to obtain paths containing more possible boundary vertices to improve overlapping community detection. Then, after representation learning with ACNE, we use soft community assignments from a simple classifier as supervision to update the weights of ACNE. This self-training mechanism referred to as ACNE-ST can help ACNE to achieve better performance. Experimental results demonstrate that the proposed methods, including ACNE and ACNE-ST, can outperform the state-of-the-art models on the subsequent tasks of vertex classification and overlapping community detection.

Network representation learning (NRL) has far-reaching effects on data mining research, showing its importance in many real-world applications. NRL, also known as network embedding, aims at preserving graph structures in a low-dimensional space. These learned representations can be used for subsequent machine learning tasks, such as vertex classification, link prediction, and data visualization. Recently, graph convolutional network (GCN)-based models, e.g., GraphSAGE, have drawn a lot of attention for their success in inductive NRL. When conducting unsupervised learning on large-scale graphs, some of these models employ negative sampling (NS) for optimization, which encourages a target vertex to be close to its neighbors while being far from its negative samples. However, NS draws negative vertices through a random pattern or based on the degrees of vertices. Thus, the generated samples could be either highly relevant or completely unrelated to the target vertex. Moreover, as the training goes, the gradient of NS objective calculated with the inner product of the unrelated negative samples and the target vertex may become zero, which will lead to learning inferior representations. To address these problems, we propose an adversarial training method tailored for unsupervised inductive NRL on large networks. For efficiently keeping track of high-quality negative samples, we design a caching scheme with sampling and updating strategies that has a wide exploration of vertex proximity while considering training costs. Besides, the proposed method is adaptive to various existing GCN-based models without significantly complicating their optimization process. Extensive experiments show that our proposed method can achieve better performance compared with the state-of-the-art models.

Supporting a massive amount of Internet of Things applications requires a large pool of spectrum. DSM is a promising ecosystem to improve the spectrum efficiency. In the era of LoRaWAN, the physical hardware constraints, along with the bandwidth-hungry applications pose new challenges. In this article, we investigate a novel deep-reinforcement-learning-based spectrum-sharing paradigm, termed Intelligent Overlapping, that explores partially overlapping channels for concurrent spectrum access in LoRaWAN. Our key insight is to leverage the coding redundancy to expand the available spectrum without complicated data processing algorithms. In particular, we learn the extra coding redundancy from the data on the non-overlapping spectrum via a deep-Q-learning network, and we apply such redundancy to recover the data on the overlapping spectrum. In the Media Access Control layer, we predict the channel condition and strategically learn and assign the appropriate overlapping portion to the concurrent access end devices. In the Physical layer, we harness interleaving to randomize the mutual interference to ensure that all the data remains decodable. Simulation results demonstrate that Intelligent Overlapping greatly improves the spectrum efficiency with a fast convergence rate compared to the conventional DSM mechanisms.

Leveraging emerging edge computing and 5G networks, researchers proposed to offload the 3D rendering of interactive multimedia applications (e.g., virtual reality and cloud gaming) onto edge servers. For high resource utilization, multiple rendering tasks run in the same GPU server and compete against each other for the computation resource. Each task has its requirement for performance, i.e., QoS target. A significant problem is how to schedule tasks so that each preset QoS is met and the performance of all tasks are maximized. We make the following contributions. First, we formulate the problem into a QoS constrained max-min utility problem. Second, we find that using the common natural logarithm as a utility function overly promotes one performance but demotes another. To avoid this phenomenon, we design a special utility function. Third, we propose an efficient scheduling algorithm, consisting of a resolution adjustment algorithm and a frame rate fair scheduling algorithm, both of which interact with each other. The former selects resolutions for tasks and the latter decides which task to process. We evaluate our method with actual rendering data, and the simulations demonstrate that our method can effectively improve task performance as well as satisfy QoS simultaneously.

Non-orthogonal multiple access (NOMA) has been investigated to support massive connectivity for Internet-of-things (IoT) networks. However, since most IoT devices suffer from limited power and decoding capabilities, it is not desirable to pair a large number of devices simultaneously, which encourages two-user NOMA grouping. Additionally, most existing techniques have not considered the diversity in the target QoS of IoT devices, which may lead to spectrum inefficiency. Few investigations have partially considered that issue by using an order-based power allocation (OPA) approach, where the power is allocated according to the order to the user’s target throughput within a priority-based NOMA (PNOMA) group. However, this does not fully capture the effects of diversity in the values of the users’ target throughputs. In this work, we handle both problems by considering a throughput-based power allocation (TPA) approach, that captures the QoS diversity, within a three-users PNOMA group as a compromise between spectral efficiency and complexity. Specifically, we investigate the performance of a time-division PNOMA (TD-PNOMA) scheme, where the transmission time is divided into two-time slots with two-users per PNOMA group. The performance of such TD-PNOMA is compared with a fully PNOMA (F-PNOMA) scheme, where the three users share the whole transmission time, in terms of the ergodic capacity under imperfect successive interference cancellation (SIC). The results reveal the superiority of TPA compared with OPA approach in both schemes, besides that the throughput of both schemes can outperform each other under imperfect SIC based on the transmit signal-to-noise ratio and the deployment scenarios.

Underwater wireless optical communication (UWOC) has received widespread attention recently due to its advantages in terms of high bandwidth, low power consumption and low delay. However, unlike underwater acoustic communication systems, the underwater wireless optical communication system is limited by its short communication range. In order to enlarge its application domain, it requires relaying and routing technology to achieve long-distance high bandwidth reliable transmissions (e.g., real-time video streaming). Although pre-deploying large number of sensor nodes is one of the possible solutions, this is impractical in response to emergency events such as oil pipeline leakage and shipwreck recovery. To this end, in this paper, we leverage swarms of autonomous underwater vehicles (AUVs) to build underwater optical communication links to transfer real-time video streaming. We first model the optical transmission channel considering solar noise, and then calculate both the optical communication angle and the reliable communicating range that satisfy a pre-defined bit error rate (BER). We then formulate the deployment optimization problem while targeting on minimizing both deployment energy consumption and delay. We also take positioning errors of AUVs into account to guarantee reliable communications. We provide an energy efficient algorithm to solve the formulated deployment problem. Through extensive simulations, we show that the algorithm effectively improves the deployment performance and achieves reliable communications.

In industrialized wine production, brewing and aging are two key steps. These two processes require the liquors to be bottled for a long time, sometimes more than ten years. The liquor is vulnerable and highly susceptible to microbial contamination during storage, causing undetectable deterioration. During the production process, wineries control the indoor temperature and carbon dioxide concentration to slow down other microorganisms' reproduction speed. These methods, however, do not prevent pathogenic microorganism growth. Currently, microbial culture methods are not suitable for real-time liquor quality monitoring in wineries. Therefore, we have designed a lightweight monitoring system called Lili, which uses light signals to monitor real-time liquor quality changes. Lili detects the changes in surface tension and absorption spectrum caused by microbial metabolites and growth during deterioration. Lili employs eight LEDs and one photodiode to achieve fine-grained surface tension and absorption spectrum measurements. By analyzing these changes, Lili realizes real-time quality monitoring. In this paper, the characteristic offset degree measurement and the absorption spectrum dimension expansion are two critical technologies. In addition, we implemented countermeasures against ambient light noise and sloshing interference. Lili's surface tension and absorption spectrum measurement errors are only 0.89 mN/m and 2.4%, respectively, making it useful to identify the contamination duration, microorganism content and microorganism composition. These two data points can be used to determine potential issues with liquor quality when the liquor becomes health-threatening or even just contaminated, with an accuracy of 97.5%.

Traditional liquid identification instruments are often unavailable to the general public. This paper shows the feasibility of identifying unknown liquids with commercial lightweight devices, such as a smartphone. The wisdom arises from the fact that different liquid molecules have various viscosity coefficients, so they need to overcome dissimilitude energy barriers during relative motion. With this intuition in mind, we introduce a novel model that measures liquids' viscosity based on active vibration. Yet, it is challenging to build up a robust system utilizing the built-in accelerometer in smartphones. Practical issues include under-sampling, self-interference, and volume change impact. Instead of machine learning, we tackle these issues through multiple signal processing stages to reconstruct the original signals and cancel out the interference. Our approach could achieve the liquid viscosity estimates with a mean relative error of 2.9% and distinguish 30 kinds of liquid with an average accuracy of 95.47%.

Mobility is one of the most fascinating features of unmanned aerial vehicles (UAVs) to support many critical applications (e.g., data dissemination) in emergency scenarios. Communication information in such applications could be confidential, and hence an effective security technique is required to mitigate the presence of an eavesdropper to some extent. On the other hand, blessed by the flexibility in the management and low overhead, physical layer security has received a significant attention in the recent years. To this end, we consider a communication system, in which a sensor-equipped mobile UAV hovers over a region to collect information and then disseminate this information to static a ground network entity in a confidential manner. Because of the popularity and practicality, the UAV is assumed to hover following the random way point (RWP) mobility model. We investigate the secrecy characteristics of the UAV under steady-state running state in terms of positive secrecy capacity probability and secrecy outage probability for the communication between the UAV and the receiver. Because of the intractable singleton analysis owing to the UAV mobility, we derive four separate closed form expressions of these performance metrics in four quadrants of conventional coordinate systems. We then investigate the secrecy performance of the UAV while considering the pause time of its RWP model. Furthermore, we propose two types of secrecy improvement strategies for the considered communication model. We strike a good trade-off between the secrecy improvement and transmit outage probability. Extensive simulations have been conducted to validate our theoretical analysis.

The vowel is deemed the essence of the syllable in which controls the articulation of each word uttered. However, articulation sensing has not been adequately evaluated. The challenging task is that the speech signal contains insufficient information for articulation analysis. We propose a new approach to identify the articulation of monophthongs in multiple languages. We employ simultaneously two ranges of acoustic signals, both speech and ultrasonic signal, to recognize lip shape and tongue position, which is implemented into an off-the-shelf smartphone to be more accessible. The articulation recognition accuracy is 94.74%. The proposed system also applies to an alternative model for a pronunciation training system that gives articulation feedback to a user.

Strength training is essential for both physical and mental well-being. Muscular mass and strength gain can help with weight loss, balance improvement, and fall prevention. The neuromuscular connection, or mind-muscle connection, is critical for improving strength training performance. However, many fitness trackers and applications are missing a feature that allows users to track their neuromuscular workout performance. The goal is to immerse the user experience while keeping the cost and size of the healthcare device to a minimum. A wearable EEG hairband and EMG shirt are outfitted with dry and non-invasive bio-signal detecting that securely attaches to the body's surface during exercise. Participants in our study are exposed to five upper-limb free-weight exercises. The result shows that low-intensity exercise can increase upper-limp muscle contraction by over 30%, and individuals with mental effort have an average precision of 81%.

Being a beneficial service in emergency situations, UAV-aided relay communications have received tremendous attentions in the recent years. In this paper, we consider a three-node UAV-based relay network, in which a mobile UAV establishes communication between two remaining nodes in the system. Despite numerous works available for such a system on the optimization of trajectory and other communication resources, none of these have addressed the delay performance of the system at the granular packet level. Furthermore, it is already established that the equipment of buffer at the relay node provides more flexibility in the delivery of packets through the exploitation of better channel quality. On the other hand, with the continued popularity of multimedia and similar other applications, it is very likely that the source node in the system receives bursty traffic from different external networks. Given that the source and UAV nodes have finite packet-level buffers and the trajectory of the UAV is known, under a bursty traffic model, we aim to study the average end-to-end packet delay and buffer overflow performance of the system. While capturing the predictable channel variation due to the movement of the UAV, we establish a queuing model for the source and UAV nodes based on the stochastic process. Then, from the dynamics of the established queuing model, we derive the average end-to-end packet delay and queue overflow probability of such a system. Through extensive numerical simulation, we justify the accuracy and effectiveness of the proposed analytical model while comparing with three static deployment scenarios of the UAV. Through providing sufficient analytical evidence as well as the numerical results, we exhibit that a mobile relay system outperforms the static relay one in terms of both the delay and buffer overflow metrics.

In this paper, we investigate the uplink communication of a reconfigurable intelligent surface (RIS) assisted system, in which an user equipment (UE) with single radio frequency (RF) chain delivers information to an access point (AP) by adopting the spatial modulation (SM). Specifically, we first investigate the transmit SM (TSM) scheme and jointly optimize the UE’s power allocation matrix and the RIS reflection coefficients to enhance the system reliability. We formulate a non-convex optimization problem to reduce the system symbol-error-rate (SER) and propose a novel penalty-alternative optimizing algorithm to obtain a near-optimal solution. Following this, we show that with the assistant of RIS, receive SM (RSM) scheme can also be performed even if the UE has only one RF chain. Based on this observation, a novel RIS-assisted RSM scheme is proposed, which can provide a low cost and complexity solution for system realization. The reflection coefficients of the RIS are also optimized for the proposed RSM. Numerical results show that the RIS-assisted TSM can achieve a lower SER than the conventional communication scheme (CTS) without SM and the proposed RSM scheme has lower detection complexity than that of CTS. It is also shown that the performance of the TSM and RSM schemes is more sensitive to the quantization accuracy of the phase of the RIS coefficients than that of the amplitude.

Smart wristband has become a dominant device in the wearable ecosystem, providing versatile functions such as fitness tracking, mobile payment, and transport ticketing. However, the small form-factor, low-profile hardware interfaces and computational resources limit their capabilities in security checking. Many wristband devices have recently witnessed alarming vulnerabilities, e.g., personal data leakage and payment fraud, due to the lack of authentication and access control. To fill this gap, we propose a secure text pin input system, namely Taprint, which extends a virtual number pad on the back of a user's hand. Taprint builds on the key observation that the hand “landmarks”, especially finger knuckles, bear unique vibration characteristics when being tapped by the user herself. It thus uses the tapping vibrometry as biometrics to authenticate the user, while distinguishing the tapping locations. Taprint reuses the inertial measurement unit in the wristband, “overclocks” its sampling rate with the cubic spline interpolation to extrapolate fine-grained features, and further refines the features to enhance the uniqueness and reliability. Extensive experiments on 128 users demonstrate that Taprint achieves a high accuracy (96 percent) of keystrokes recognition. It can authenticate users, even through a single-tap, at extremely low error rate (2.2 percent), and under various practical usage disturbances.

Mobile video chatting has emerged as an important Internet multimedia application that greatly enriches interpersonal communications. Mobile power efficiency is crucial to the service quality and time of video chatting on battery-limited smartphones. However, the power characteristics of the video coding and data communication are highly complex due to the time-varying network conditions and dynamic mobile energy features. This incurs crucial challenges to maintaining the low power dissipation of mobile chatting application while streaming satisfactory-quality videos. To address these challenges, this paper presents a joinT cOding-tranSmission Optimization (TOSO) protocol at application layer that performs machine learning based adaptation of the video bit rate and FEC (Forward Error Correction) coding parameters. By taking advantage of analytical and empirical models characterizing the quality-power relationship, TOSO is able to maximize video quality subject to a specified upper bound of power consumption in mobile chat application. This distinguishing feature prevents the video chat from draining battery too quickly. Moreover, it allows the smartphone operating system or the mobile user to define a desired video chat duration given the remaining battery, avoiding unpleasant conversation disruption due to battery depletion. Extensive experiments based on the Linphone platform and Exata network emulator show that TOSO outperforms baseline approaches by 29.3 percent in power conservation while achieving the same video quality level.

Recently, wearable devices have become increasingly popular in our lives because of their neat features and stylish appearance. However, their tiny sizes bring about new challenges to human-device interaction such as texts input. Although some novel methods have been put forward, they possess different defects and are not applicable to deal with the problem. As a result, we propose an acoustic-based texts-entry system, i.e., EchoWrite, by which texts can be entered with a finger writing in the air without wearing any additional device. More importantly, different from many previous works, EchoWrite runs in a training-free style which reduces the training overhead and improves system scalability. We implement EchoWrite with commercial devices and conduct comprehensive experiments to evaluate its texts-entry performance. Experimental results show that EchoWrite enables users to enter texts at a speed of 7.5 WPM without practice, and 16.6 WPM after about 30-minute practice. This speed is better than touch screen-based method on smartwatches, and comparable with previous related works. Moreover, EchoWrite provides favorable user experience of entering texts.

Despite extensive research efforts, underwater sensor networks (UWSNs) still suffer from serious performance issues due to their inefficient and uncoordinated channel access and resource management. For example, due to the lack of holistic knowledge on the network resources, existing decentralized routing protocols fail to provide globally optimal performance. On the other hand, Software Defined Networking (SDN), as a promising paradigm to provide prominent centralized solutions, can be employed to address the aforementioned issues in UWSNs. Indeed, SDN brings unprecedented opportunities to improve the network performance through the development of advanced algorithms at controllers. In this paper, we study the routing problem in such a network with new features including centralized route decision, global network-state awareness, seamless route discovery while considering the optimization of several long-term global performance metrics. We formulate the entire routing problem of a multi-modal UWSN as an optimization problem while considering the interference phenomenon of ad hoc scenarios and some long-term global performance metrics of an ideal routing protocol. Our formulated problem nicely captures all possible flexibilities of a sensor node no matter it has the full-duplex or half-duplex functionality. Upon the formulation, we recognize the NP-hard nature of the problem for all possible scenarios. We adopt a rounding technique based on the convex programming relaxation concept to solve the formulated routing problem that considers full-duplex scenarios, whereas we solve the problem for half-duplex scenarios using a greedy method upon interpreting it as a submodular function maximization problem. Through extensive simulation via our Python-based in-house simulator, we verify that our proposed globally optimal routing scheme always outperforms three existing decentralized routing protocols (each of these protocols are selected from each of three prominent protocol types, i.e., flooding, cross-layer information and adaptive machine learning based, respectively) in terms of reliability, latency, energy efficiency, lifetime and fairness.

To develop sustainable fifth generation (5G) wireless networks and utilize the unused spectrum, this paper focuses on cognitive radio (CR) based wireless information and energy transmissions from an unmanned aerial vehicle (UAV) to multiple low-power ground terminals (GTs). By practically considering the location-dependent air-to-ground (A2G) channel states and the non-linear energy harvesting (EH), we propose a dynamic fly-hover-transmit scheme, where the UAV successively flies between GTs, and hovers close to each GT for efficient wireless energy transfer (WET) or wireless information transfer (WIT) when the primary user (PU) is idle. By causally and optimally determining the UAV's mobility and transmit power for each selected transmission mode (WIT, WET, or being silent), we formulate the UAV's sum-throughput maximization over all GTs as a constrained Markov decision process (MDP) problem with battery energy constraints at all GTs and the UAV. Due to the infinitely large MDP system state space, this problem is difficult to solve. We then decompose this problem into two subproblems, by first deciding the UAV's transmission mode and power above a given GT, and then optimizing the UAV movement policy over multiple GTs. In the first subproblem, we propose an approximate to the complicated MDP value function of low complexity in closed-form, and then analytically derive the threshold-based suboptimal transmission policies. In the second subproblem, we optimally solve a simple-but-fundamental two-GT case, and then extend the general location-dependent GT weight design to an efficient suboptimal UAV movement policy. Simulation results show the significantly improved system performance under the proposed suboptimal policies over various benchmarks in dynamic networks.

Deep reinforcement learning has been successfully applied in many decision making scenarios. However, the slow training process and difficulty in explaining limit its application. In this paper, we attempt to address some of these problems by proposing a framework of Rule-interposing Learning (RIL) that embeds knowledge into deep reinforcement learning. In this framework, the rules dynamically effect the training progress, and accelerate the learning. The embedded knowledge in form of rule not only improves learning efficiency, but also prevents unnecessary or disastrous explorations at early stage of training. Moreover, the modularity of the framework makes it straightforward to transfer high-level knowledge among similar tasks.

Bus bunching, a phenomenon due to the failure of headway or timetable adherence, often causes low level of public transit service with poor bus on-time performance and excessive passenger waiting time. To mitigate bus bunching, an accurate and real-time prediction method plays an important role. In this paper, we propose a supply-demand seq2seq model called SD-seq2seq to predict bus bunching using smart card data. Features from both supply and demand sides of bus service are taken into account, like bus stop type, dwelling time, passenger demand and type, and so on. Extensive experiments on multiple bus routes in real world demonstrate that our method outperforms other baseline methods. The proposed method is expected to provide useful online information of bus operation to both bus operators and passengers.

Taxi ridesharing becomes promising and attractive because of the wide availability of taxis in a city and tremendous benefits of ridesharing, e.g., alleviating traffic congestion and reducing energy consumption. Existing taxi ridesharing schemes, however, are not efficient and practical, due to they simply match ride requests and taxis based on partial trip information and omit the offline passengers, who hail a taxi at roadside with no explicit requests to the system. In this paper, we consider the mobility-aware taxi ridesharing problem, and present mT- Share to address these limitations. mT-Share fully exploits the mobility information of ride requests and taxis to achieve efficient indexing of taxis/requests and better passenger-taxi matching, while still satisfying the constraints on passengers' deadlines and taxis' capacities. Specifically, mT-Share indexes taxis and ride requests with both geographical information and travel directions, and supports the shortest path based routing and probabilistic routing to serve both online and offline ride requests. Extensive experiments with a large real-world taxi dataset demonstrate the efficiency and effectiveness of mT-Share, which can response each ride request in milliseconds and with a moderate detour cost. Compared to state-of-the-art methods, mT-Share serves 42% and 62% more ride requests in peak and non-peak hours, respectively.

PageRank has a wide applications in online social networks and serves as an important benchmark to examine graph processing frameworks. Many efforts have been made to improve the computation efficiency of PageRank in shared-memory platforms, where a single machine can be sufficiently powerful to handle a large-scale graph. Existing methods, however, still suffer from synchronization issues and irregular memory accesses, which will deteriorate their overall performance. In this paper, we present an accelerated parallel PageRank computation approach, named APPR. By investigating the characteristics of parallel PageRank computation and degree distributions of social network graphs, APPR proposes a series of optimization techniques to improve the efficiency of PageRank computation. Specifically, a destination-centric graph partitioning scheme is designed to avoid the synchronization issues when concurrently updating the common vertex data. By exploiting power-law structure of social network graphs, APPR can intelligently schedule the computations of vertices to save computing operations. The vertex messages are adjusted by APPR for transmission to further improve the locality of memory accesses. Empirical evaluations are performed based on a set of large real-world graphs. Experimental results show that APPR significantly outperforms the state-of-the-art methods, with on average 2.4x speedup in execution time and 16.4x reduction in DRAM communication.

With the prosperity of Internet of Things, a large number of heterogeneous wireless devices share the same unlicensed spectrum, leading to severe cross-technology interference (CTI). Especially, the transmission power asymmetry of heterogeneous devices will further deteriorate this problem, making the low-power devices prohibited from data transmission and starved. This paper proposes an enhanced CCA (E-CCA) mechanism to mitigate CTI, so as to improve the performance and fairness among heterogeneous networks. E-CCA contains a signal identification design based on deep learning to identify the signal type within a tolerable time duration, it also contains a CCA adaptive mechanism based on the signal type to avoid CTI. As a result, the ZigBee devices could compete for the channel with WiFi devices more fairly, and the network performance can be improved accordingly. We set up a testbed based on TelosB, a commercial ZigBee platform, and USRP N210, which can be used as the WiFi platform. With the collected signals through USRP N210, over 99.9% signal identification accuracy can be achieved even when the signal duration is tens of microseconds. Simulation results based on NS-3 shows that E-CCA can increase the ZigBee performance dramatically with little throughput degradation for WiFi.

Long Range (LoRa) communication, thanks to its wide network coverage and low energy operation, has attracted extensive attentions from both academia and industry. However, existing LoRa-based Wide Area Network (LoRaWAN) suffers from severe inter-network interference, due to the following two reasons. First, the densely-deployed LoRa ends usually share the same network configurations, such as spreading factor (SF), bandwidth (BW) and carrier frequency (CF), causing interference when operating in the vicinity. Second, LoRa is tailored for low-power devices, which excludes LoRaWAN from using the listen-before-talk (LBT) mechanisms commonly used in wireless communication technologies, such as WiFi and ZigBee -LoRaWAN has to use the duty-cycled medium access policy and thus being incapable of channel sensing or collision avoidance. To mitigate the inter-network interference, we propose a novel solution achieving the online concurrent transmissions at LoRa gateway, called OCT, which recovers collided packets at the gateway and thus improves LoRaWAN's throughput. Moreover, OCT achieves the online concurrent transmission using only LoRa's (de)modulation information, thus can be easily deployed at LoRa gateway. We have implemented and evaluated OCT on USRP platform and commodity LoRa ends, showing OCT achieves: (i) >90% packet reception rate (PRR), (ii) 3 × 10-3bit error rate (BER), (iii) 2x and 3x throughput in the scenarios of two- and three- packet collisions respectively, and (iv) reducing 67% latency compared with state-of-the-art.

Energy efficiency is an important criterion to judge the quality of mobile apps, but one third of our randomly sampled apps suffer from energy issues that can quickly drain battery power. To understand these issues, we conducted an empirical study on 27 well-maintained apps such as Chrome and Firefox, whose issue tracking systems are publicly accessible. Our study revealed that the main root causes of energy issues include unnecessary workload and excessively frequent operations. Surprisingly, these issues are beyond the application of present technology on energy issue detection. We also found that 25.0% of energy issues can only manifest themselves under specific contexts such as poor network performance, but such contexts are again neglected by present technology. In this paper, we propose a novel testing framework for detecting energy issues in real-world mobile apps. Our framework examines apps with well-designed input sequences and runtime contexts. To identify the root causes mentioned above, we employed a machine learning algorithm to cluster the workloads and further evaluate their necessity. For the issues concealed by the specific contexts, we carefully set up several execution contexts to catch them. More importantly, we designed leading edge technology, e.g. pre-designing input sequences with potential energy overuse and tuning tests on-the-fly, to achieve high efficacy in detecting energy issues. A large-scale evaluation shows that 91.6% issues detected in our experiments were previously unknown to developers. On average, these issues double the energy costs of the apps. Our testing technique achieves a low number of false positives.

Emotions show high significance on human health. Automatic emotion recognition is helpful for monitoring psychological disorders, mental problems and exploring behavioral mechanisms. Existing approaches adopt costly and bulky specialized hardware such as EEG/ECG helmet, possess privacy risks, or with low accuracy and user experience. With the increasing popularity of wearables, people tend to equip multiple smart devices, which provides potential opportunity for emotion perception. In this paper, we present a pervasive and portable system called MW-Emotion to recognize common emotional states with multi-modal wearable devices. However, ubiquitous wearable devices perceive shallow information which is not obviously related to human emotions. MW-Emotion excavates intrinsic mapping relationship between emotions and sensing data. Our experiments show that MW-Emotion can recognize different emotion states with a relatively high accuracy of 83.1%.

Signature is one of the most prevailing identity authorization approaches. It is yet inconvenient to use in real life in the sense that a majority of existing signature verification approaches rely on additional digital signing devices. In this paper, we propose a portable device-free signature verification system named SilentSign which makes use of acoustic sensors (i.e., microphone and speaker) embedded in smart devices to enable secure and convenient signature verification service. The basic idea is to leverage acoustic signals to measure the distance variation of the tip of the pen while signing. We carefully design the signal modulation scheme, develop a phase-based distance measurement technique, and train the verification model for high performance and robustness. Compared with conventional digital signing systems, SilentSign allows users to sign more invisibly and conveniently. We conduct extensive experiments involving 35 participants to evaluate SilentSign. Results show that SilentSign can achieve 98.2% AUC and 1.25% EER.

The coexistence of heterogeneous devices in wireless networks brings a new topic on cross-technology communication (CTC) to improve the coexistence efficiency and boost collaboration among these devices. Current advances on CTC mainly fall into two categories, physical-layer CTC and packet-level energy modulation (PLEM). The physical-layer CTC achieves a high CTC data rate, but with channel incompatible to commercial devices, making it hard to be deployed in current wireless networks. PLEM is channel and physical layer compatible, but with two main drawbacks of the low CTC data rate and MAC incompatibility, which will induce severe interference to the other devices’ normal data transmissions. In this paper, we propose symbol-level energy modulation (SLEM), the first CTC method that is fully compatible with current devices in both channel and the physical/MAC layer processes, having the ability to be deployed in commercial wireless networks smoothly. SLEM attaches the CTC bits to one WiFi data packet through adjusting the energy levels of WiFi symbols, achieving concurrent CTC and WiFi transmissions. We make theoretical analysis to figure out the performance tradeoff between CTC and WiFi, and also conduct experiments to demonstrate the feasibility of SLEM and its performance under different network situations.

User authentication and identification on smart devices has great significance in keeping data privacy and recommending personalized services. Existing few research works propose active sensing systems that emit and receive inaudible acoustic signals to authenticate users. But they share shortcomings of intrusiveness to users, high power consumption, and purely focusing on authentication. Instead, in this paper, we propose a passive sensing system called EarID with low-cost customized earpieces which attains user authentication and identification simultaneously. It makes use of a embedded microphone to sense body sounds spread out through ear canals and extract 'fingerprints' as a novel biometric feature. With self-designed earpieces, we design a deep learning-based real-time data processing pipeline. Extensive experiments under different real-world settings show that EarID can achieve a rather low false acceptance rate less than 5% for user authentication and a high F1 score of 96% for user identification.

Surface identification provides contextual services during humancomputer interaction, which is important for target detection and scene understanding. A robust and ubiquitous surface recognition system has a wide range of applications such as context awareness and robot operation. Existing methods have shortcomings of requiring specialized devices and limited usage scenarios. In this paper, we introduce Surtify, a surface identification system based on acoustic dispersion with a smartphone. By combining the intrinsic physical phenomenon (i.e., acoustic dispersion) with a deep learning model, Surtify can identify eleven kinds of surfaces with accuracies up to 96%, even in cross-person and cross-location scenarios.

Strength training is becoming increasingly popular among all age groups, as it helps the participants increase muscle strength, improve body flexibility, reduce health risks, and reshape physical forms. However, strength training imposes strict regulations on gestures and requires professional instruction in real time for the sake of body-building efficiency and safety. For this purpose, in this article, we propose a novel low-cost system named iCoach, to provide real-time monitoring and coaching service for strength training participants. Specifically, we design and implement a smart fitness glove, which can be seamlessly equipped with a pervasive inertial unit. With this customized but low-cost device, we can recognize various training programs, detect nonstandard behaviors while exercising, and assess exercising qualities of a user. Our primary experimental results show that iCoach can recognize 15 sets of training programs, detect three common nonstandard behaviors, and assess the quality of training with high accuracy and reliability.

Mobile edge computing provides users with low response time and avoids unnecessary data transmission. Due to the deployment of 5G, the emerging edge systems can provide gigabit bandwidth. However, network protocols have not evolved together. In TCP, the initial congestion window (IW) is such a low value that most short flows still stay in slow start phase when finishing, and do not fully utilize available bandwidth. Naively increasing IW may result in congestion, which causes long latency. Moreover, since the network environment is dynamic, we have a challenging problem-how to adaptively adjust IW such that flow completion time is optimized, while congestion is minimized. In this paper, we propose an adaptive online decision method to solve the problem, which learns the best policy using deep reinforcement learning stably and fast. In addition, we propose an approach to further improve the performance by supervised learning, using data collected during online learning. We also propose to adopt SDN to address the challenges in implementing our method in MEC systems. To evaluate our method, we build an MEC simulator based on ns3. Our simulations demonstrate that our method performs better than existing methods. It can effectively reduce FCT with little congestion caused.

Nowadays, smart wristbands have become one of the most prevailing wearable devices, as they are small and portable. However, due to the limited size of the touch screens, smart wristbands typically have poor interactive experience. There are a few works appropriating the human body as a surface to type on. Yet, by using multiple sensors at high sampling rates, they are not portable and are energy-consuming in practice. To break this stalemate, we proposed a portable, cost efficient text-entry system, termed ViType, which first leverages a single small form factor sensor to achieve a practical user input with much lower sampling rates. To enhance the input accuracy with less vibration information introduced by lower sampling rates, ViType designs a set of novel mechanisms, including a fine-grained feature extraction to process the vibration signals, and a runtime calibration and adaptation scheme to recover from the error due to temporal instability. Extensive experiments have been conducted on 30 human subjects. The results demonstrate that ViType is robust against various confounding factors. The average recognition accuracy is 95 percent with an initial training sample size of 20 for each key. The accuracy is 1.54 times higher than the state-of-the-art on-body typing system. Furthermore, when turning on the runtime calibration and adaptation system to update and enlarge the training sample size, the accuracy can reach around 98 percent on average during one month.

This paper investigates a spectrum sharing based cognitive unmanned aerial vehicle (UAV) network. To protect the primary users (PUs) from harmful co-channel interference from the UAVs' transmissions, and at the same time, to control the resultant downlink interference at the secondary users (SUs), each UAV is equipped with a directional antenna of adjustable beamwidth. We adopt a probability-based air-to-ground (AtG) channel model to reflect both line-of-sight (LoS) and non-line-of-sight (NLoS) effects from the UAVs' transmissions to the users on ground. By studying the interference distributions in both primary network and cognitive UAV network, we successfully characterize the coverage probabilities for both PUs and SUs. We then optimally design the UAV density and the UAV beamwidth, so as to maximize the UAV coverage probability under a PU coverage probability constraint. Despite of the complicated network performance characterizations that make this problem non-convex, we find an efficient method to solve this problem. Finally, numerical results are provided to validate the theoretical analysis, and show the performance of our proposed spectrum sharing method via UAV beamwidth over the benchmark scheme.

In this paper, we investigate the uplink of a millimeter-wave (mmWave) communication system, in which small base stations (SBSs) are densely placed. The positions of the SBSs are modeled as a homogeneous Poisson point process (PPP) and a user equipment (UE) connects to multiple SBSs opportunistically and adopts the spatial modulation (SM) scheme. Two detection methods, namely the centralized detection and distributed detection, are first proposed for the SBSs to demodulate the received signal and the power allocation scheme minimizing the system symbol error rate (SER) is investigated. Following this, by ignoring the side lobe of the UE, the probability that the SM scheme outperforms the conventional transmission scheme in which the UE only communicates with the SBS with a stronger channel is approximated to evaluate the performance of the SM scheme. Numerical results show that the SM scheme can effectively improve reliability of the system.

Internet of Things (IoT) plays an indispensable role in our daily life, in many cases, IoT systems are implemented following the client-server paradigm, which are vulnerable to single point of failures and malicious attacks. Due to the resilience and security promise of blockchain, the idea of combining blockchain and IoT has gained considerable attention in recent years. However, blockchains are power-intensive and low-throughput, which may not suitable for power-constrained IoT devices. To tackle these challenges, we present B-IoT, a blockchain based IoT system with credit-based consensus mechanism. We propose a credit-based proof-of-work (PoW) mechanism for IoT devices, which enhances security and improves transaction efficiency simultaneously. In order to protect the confidentiality of sensitive IoT data, we design a data authority management method to regulate the access to sensor data. In addition, our system is built based on a directed acyclic graph (DAG)-structured blockchain, which is more efficient than the satoshi-style blockchain. We implement a prototype of B-IoT on Raspberry Pi, and conduct case studies of a smart factory. Extensive evaluation and analysis results demonstrate that the proposed credit-based PoW mechanism and data access control are practical for IoT.

Recently, wearable devices have become increasingly popular in our lives because of their neat features and stylish appearance. However, their tiny sizes bring about new challenges to human-device interaction such as texts input. Although some novel methods have been put forward, they possess different defects and are not applicable to deal with the problem. As a result, we propose an acoustic-based texts-entry system, i.e., EchoWrite, by which texts can be entered with a finger writing in the air without wearing any additional device. More importantly, different from many previous works, EchoWrite runs in a training-free style which reduces the training overhead and improves system scalability. We implement EchoWrite with commercial devices and conduct comprehensive experiments to evaluate its texts-entry performance. Experimental results show that EchoWrite enables users to enter texts at a speed of 7.5 WPM without practice, and 16.6 WPM after about 30- minute practice. This speed is better than touch screen-based method on smartwatches, and comparable with previous related works.

Fuzzy rough set based feature selection is a widely adopted technique for big data analysis. However, the high accuracy of this technique depends on all the data correlations, so that it always works in the centralized computing mode. With the increasing data volume, the centralized server, especially its computation capability and memory space, cannot afford the computing of fuzzy rough set. To enable the fuzzy rough set for big data analysis, in this paper, we propose the novel Distributed Fuzzy Rough Set (DFRS) based feature selection in cloud computing, which separates and assigns the tasks to multiple nodes for parallel computing. The key challenge is to maintain the global information on each distributed node without conserving the entire fuzzy relation matrix. We tackle this challenge by a dynamic data decomposition algorithm and a data summarization process on each distributed node. Extensive experiments based on multiple real datasets demonstrate that DFRS significantly improves the runtime and its feature selection accuracy is nearly the same as the traditional centralized computing.

Smart wristband has become a dominant device in the wearable ecosystem, providing versatile functions such as fitness tracking, mobile payment, and transport ticketing. However, the small form-factor, low-profile hardware interfaces and computational resources limit their capabilities in security checking. Many wristband devices have recently witnessed alarming vulnerabilities, e.g., personal data leakage and payment fraud, due to the lack of authentication and access control. To fill this gap, we propose a secure text pin input system, namely Taprint, which extends a virtual number pad on the back of a user's hand. Taprint builds on the key observation that the hand "landmarks'', especially finger knuckles, bear unique vibration characteristics when being tapped by the user herself. It thus uses the tapping vibrometry as biometrics to authenticate the user, while distinguishing the tapping locations. Taprint reuses the inertial measurement unit in the wristband, "overclocks'' its sampling rate to extrapolate fine-grained features, and further refines the features to enhance the uniqueness and reliability. Extensive experiments on 128 users demonstrate that Taprint achieves a high accuracy (96%) of keystrokes recognition. It can authenticate users, even through a single-tap, at extremely low error rate (2.4%), and under various practical usage disturbances.

In this poster, we present our recent work, a wearable system for achieving real-time 3D arm skeleton. We have coped with the major challenge that the skeleton of each arm is determined from the locations of the elbow and wrist, whereas a wearable device only senses a single point from the wrist. Result shows that the potential solution space is huge. This underconstrained nature fundamentally challenges the achievement of accurate and real-time arm skeleton tracking. In this study, we propose Hidden Markov Model (HMM) state reorganization and hierarchical search two methods to improve the heavyweight computation of the state-of-art arm tracking model and achieve real-time tracking even on mobile phone.

This paper presents ArmTroi, a wearable system for understanding and analyzing the detailed arm motions of people primarily by using the motion sensors from wrist-worn wearable devices. ArmTroi can achieve real-time 3D arm skeleton tracking and reliable gesture inference tolerant to missing wearable sensors for enabling numerous useful application designs. We have coped with two major challenges through ArmTroi. First, the skeleton of each arm is determined from the locations of the elbow and wrist, whereas a wearable device only senses a single point from the wrist. We find that the potential solution space is huge. This underconstrained nature fundamentally challenges the achievement of accurate and real-time arm skeleton tracking. Second, wearable sensors may not reliably provide sensory data. For example, devices are not worn by the user, yet the learning tools for gesture inference, such as deep learning, typically have static network structures, which require nontrivial network adaptation to match the input's varying availability and ensure reliable gesture inference. We propose effective techniques to address above challenges, and all computations can be conducted on the user's smartphone. ArmTroi is thus a fully lightweight and portable system. We develop a prototype and extensive evaluation shows the efficacy of the ArmTroi design.

Accurate traffic prediction can benefit many smart city applications. Existing works mainly consider traffic prediction on each individual road segment, and heavily rely on some statistical or machine learning models, which suffer from either poor prediction accuracy or high computation overheads for predictions of the whole road network. In this paper, we instead consider the region-level traffic prediction that is still useful for many applications. To describe the regional traffic conditions and capture their spatio-temporal dependencies, we present a deep learning based model - DeepRTP. Specifically, we use a novel metric called Traffic State Index (TSI) to measure regional traffic conditions, and carefully classify traffic data into three categories that are used to capture hourly, daily, and weekly traffic patterns. Furthermore, we employ the convolutional and residual neural networks to model both spatial and temporal dependencies. Experimental results from real-world traffic data demonstrate that DeepRTP outperforms five baseline methods and can achieve higher prediction accuracy.

Recently, wearable devices have become increasingly popular in our lives because of their neat features and stylish appearance. However, due to the tiny size, it is inconvenient for users to interact with a device using conventional methods, especially for text entry. Although some methods have been proposed to handle this problem, they have different limitations and are not applicable to many existing mobile devices. As a result, we take the first step to propose a digits-entry system, i.e., AcouDigits, in which digits can be entered in the air using a finger without taking help from any additional hardware. We implement AcouDigits on two commercial devices and conduct experiments to evaluate its performance in recognizing ten basic digits. Experimental results show that AcouDigits can achieve average accuracies of 91.7% and 87.4% in recognizing basic digits and 26 English alphabets, respectively.

Underwater channels are considered challenging media in communication due to the harsh nature of such environments. However, dynamic transmission can assist in finding sub-optimal solutions by adaptively changing the employed techniques, e.g. the forwarding scheme between nodes and the transmitted signal intensity control, to compromise for the instantaneous fluctuations in various underwater environments. Additionally, Machine Learning (ML) techniques can provide appropriate solutions for various problems e.g. routing, resource allocation, and energy-efficiency to further enhance the quality of the communication systems. In this paper, we propose a novel dynamical transmission framework for multi-hop Internet of Underwater Things (IoUT) and underwater networks to fit for various conditions. The proposed framework employs a heuristic forwarding scheme selection approach beside an adaptive transmission signal intensity method. We also propose a decision-tree based ML-model that adaptively learns the proper forwarding method beside the appropriate amount of the transmitted signal intensity for each relay node to minimize the transmission error rate and the power consumption depending on numerous parameters e.g. node location, link reliability and certain water quality metrics such as water temperature, depth, and pH measurements. The model achieves remarkable accuracy for training and testing patterns beyond the 99%.

Wearable wristbands have become prevailing in the recent days because of their small and portable property. However, the limited size of the touch screen causes the problems of fat fingers and screen occlusion. Furthermore, it is not available for users whose hands are fully occupied with other tasks. To break this bottleneck, we propose a portable, hand-free and secure text-entry system, called FaceInput, which firstly uses a single small form factor sensor to accomplish a practical user input via facial vibrations. To sense the tiny facial vibration signals, we design and implement a double-stage amplifier whose maximum gain is 225. To enhance the input accuracy and robustness, we design a set of novel schemes for FaceInput based on the Mel-frequency cepstral coefficient (MFCC) concept and a hidden Markov model (HMM) to process the vibration signals, and an online calibration and adaptation scheme to recover the error due to temporal instability. Extensive experiments have been conducted on 30 human subjects during the period of one month. The results demonstrate that FaceInput can be successful to sense the tiny facial vibrations and robust to fight against various confounding factors. The average recognition accuracy is 98.2%. Furthermore, by enabling the runtime calibration and adaptation scheme that updates and enlarges the training data set, the accuracy can reach 100%.

The inevitable aging trend of the world's population brings a lot of challenges to the health care for the elderly. For example, it is difficult to guarantee timely rescue for a single-resided elder who falls at home. Under this circumstance, a reliable automatic fall detection machine is in great need for emergent rescue. However, the state-of-the-art fall detection systems are suffering from serious privacy concerns, having a high false alarm or being cumbersome for users. In this paper, we propose a device-free fall detection system, namely G-Fall, based on geophones. We first decompose the falling mode and characterize it with time-dependent floor vibration features. By leveraging Hidden Markov Model (HMM), our system is able to recognize the fall event precisely and achieve training-free recognition. It requires no training from the elderly but only an HMM template learned in advance through a small number of training samples. To reduce the false alarm rate, we propose a novel reconfirmation mechanism, namely Energy-of-Arrival (EoA) positioning to assist in recognizing a human's fall. Extensive experiments have been conducted on 12 human subjects. The results demonstrate that G-Fall achieves a 95.74% recognition precision with a false alarm rate of 5.30% on average. Furthermore, with the assistance of EoA, the false alarm rate is reduced to nearly 0%.

Mobile crowdsourcing (MCS) becomes an emerging paradigm for various useful urban sensing application designs by assigning the crowdsourcing tasks to the participants with rich-sensor equipped mobile devices. To effectively assign MCS tasks, many research efforts have been made in the literature. However, most prior schemes mainly optimize certain performance metrics in the assignment, yet overlooking other metrics, which thus cannot guarantee the overall system performance. This also limits the applicability of the proposed solution dedicated to the targeted performance metrics only. In this paper, we present UniTask, a unified task assignment design to address these issues. UniTask jointly considers the representative MCS performance metrics, including coverage, latency, and accuracy, to optimize the overall system utility. We mathematically formulate this problem and prove its NP-hardness. To efficiently schedule tasks, we also propose a utility-aware heuristic algorithm in UniTask. Moreover, a set of optimization techniques are further designed to enhance UniTask. Extensive evaluations are performed on two real-world datasets. Experimental results demonstrate that utility is an effective indicator of the system's overall performance. With an optimization on the system utility, UniTask can outperform the baseline methods on these individual performance metrics.

Recent years have witnessed a steep grow in the multimedia-oriented Internet of Things (IoT) over vehicular networks. Huge volume of multimedia traffic generated from the in-built IoT devices should be delivered among vehicles and immediate surroundings in real time. However, as network nodes with higher mobility, vehicles often experience more unpredictable wireless channels. Such time-frequency diversity poses substantial challenges to achieve pervasive and real-time multimedia connectivity. The hurdle lies in the inability of automatically approaching the subcarrier level channel variations in the existing video codecs. With coarse-grained traffic delivery rate, video decoding fails, and intermittent connection occurs. To break this stalemate, we propose a fine-grained wireless video streaming strategy, namely iCast, that intelligently achieves the most appropriate data rate and frame protection for multimedia traffic in highly mobile vehicular environments. The insight of iCast is a simple joint source-channel rateless code. It reaps the benefits of the frequency diversity to provide fine-grained data rate for the channel in conjunction with suitable protection for the source. Our experiments show that, by harnessing frequency diversity in mobile environments, iCast outperforms the existing competitive wireless video delivery schemes by up to 5 dB peak signal-to-noise ratio.

This paper presents StrLight, the first practical VLC system that leverages widely-deployed string lights to transmit data. The data transmission is imperceptible to human eyes. Users can decode the data by mobile devices (e.g., smartphones) equipped with cameras. StrLight primarily differs from existing VLC systems in using string lights which are composed of a large number of small LEDs and thus the unique design to address practical issues including a special data modulation/encoding scheme, a data representation with unstructured/unknown topologies of LEDs in the string light, and a fault tolerance against broken and blocked LEDs. To the best of our knowledge, StrLight is the first practical VLC system of its kind. We build several prototypes of string light transmitters and test with different smartphone models and a customized mobile device as receivers. The experiment results show that StrLight provides an efficient and robust data broadcasting. A string light of 100 LEDs working in 450 Hz and a camera with a capture rate of 30 Hz and an image resolution of as low as 320 × 240 pixels, delivers data rate of ~1 kbps, without observable light flickers.

One of the main characteristics of the power-domain non-orthogonal multiple access (NOMA) technique is the possibility of users to decode the messages of the other paired users on the same resources. This technique exhibits a security threat particularly if the base station (BS) have to serve untrusted users or users with different security clearance. In this paper, we study the outage probability and the secrecy outage probability in a two-users NOMA system at which the BS is pairing a legitimate/trusted user with another untrusted user due to the non-uniform distribution of trusted and untrusted users in the cell. Through the NOMA concept, we investigate the NOMA pair outage behavior under secrecy outage probability constraint on the trusted user. More specifically, the pair outage probability (OP) and the secrecy outage probability (SOP) of the trusted user are derived in closed form expressions. We provide insights on the feasibility of achieving an outage-optimal performance for the pair under the SOP constraint. Through numerical simulations, we verify the correctness of our analytical derivations under different scenarios.

A key feature of the non-orthogonal multiple access (NOMA) technique is that users with better channel conditions have prior information about the messages of other users. The technique to exploit the prior knowledge of strong users in order to improve the performance of weak users is known as cooperative NOMA. In this paper, we study the physical layer security in such a cooperative NOMA system. In order to reduce the complexity, the considered system in this paper has two users. Through the cooperative NOMA concept, the performance of the weak user is enhanced by the strong user. Given that there is an eavesdropper in the system that can hear all transmissions, we study the secrecy rate of the strong and the weak users. More specifically, we make an attempt to derive the secrecy outage probability (SOP) of both the users. Due to the intractable nature of the exact analysis for the weak user, we provide the closed form expression for the SOP of this user in high SNR regime while keeping the exactness for the strong user. Through numerical simulations, we verify the correctness of our analytical derivations under different scenarios. Besides, we provide the insights of achieving optimal secrecy performance in such a system.

Recently, we have seen the unprecedented development in unmanned aerial vehicles (UAVs) from different aspects. Accordingly, an increasing number of applications have emerged based on UAVs. Among which, placing UAVs as Aerial Base Stations (ABSs) has received considerable interest in both the industrial and academic community. Existing solutions focus on the optimization of the UAV deployment problem for static user topology using the control information obtained from the Terrestrial Base Station (TBS), that makes hard for the controller to make real-time decisions. To break this stalemate, we propose a SemI-DistributEd system, named SIDE, for the UAV self-deployment. In SIDE, we introduce a mechanical equilibrium based approach, named EMech, via which the UAV positions are self-adapted according to users' attraction (e.g., user distance and traffic demand) within their transmission range. To facilitate the EMech, we propose a fine-grained area splitting strategy, termed KDivision, that partitions the service area in accordance with the user density. Finally, an area merging technique, namely RMerge, is exploited to approximately optimize the positions of the UAVs assisted by an Utility Function that strikes a balance amid the network performance and economic cost. We conduct field experiments to validate the feasibility of EMech. Extensive simulation results show that the proposed SIDE finds the optimal number of assigned UAVs, which not only reduces the cost of the system significantly, but also improves the achievable rate up to 74.6% compared to the existing solutions while consuming almost the same energy level.

Atmospheric humidity is a significantly important factor in our daily lives, as it is closely bound up with agriculture, industrial production, human health and so on. Therefore, efficient and precise humidity measurement techniques are indispensable. However, the existing off-the-shelf techniques, including the dry and wet bulb hygrometer, humidity sensor as well as WiFi based detector, all fail to achieve a sensitive, accurate and convenient humidity measurement, especial for a large scale deployment. In this paper, we observe that different levels of water vapor have certain impact on millimeter wave (mmWave) signals in indoor environments. Accordingly, we propose an fine-grained environmental humidity sensing technology via wireless signals in the mmWave band. However, mmWave signals are not only sensitive to humidity, but also other environmental factors, such as oxygen. To establish a linear relationship between humidity and mm Wave signal propagation, we exploit a subspace projection technique to remove the environmental noise. Upon extracting the humidity-associated features in the noise-free signal, we utilize support vector machine (SVM) to model the humidity measurement classifier of a certain place. Extensive experiments have been conducted in different scenarios in order to verify the effectiveness of the proposed system. Results show that the average accuracy of humidity measurement is up to 85 % when the humidity interval is 3 %, and is 95 % when the humidity interval is 5%. We further show that the proposed method is very sensitive to the humidity dynamics and is 63.2 times faster compared to the traditional hygrometers.

The emergence of wearable devices has brought great simplicity and convenience to people's daily lives. However, due to the small form-factor, low-profile hardware interfaces, the input scheme for such wearable devices becomes a bottleneck and even sabotages their functionalities. The state-of-the-art interaction schemes, including voice input, inertial measurement unit (IMU)based input, or acoustic based input, all require a stable environment, which is critical for wearable device. To break this stalemate, we propose a stable QWERTY keyboard input for wearable devices based on bone-conduction models. Using the characteristics of human anatomy, we achieve a low-cost and high precision text input system, named Osteoacusis input (Oinput), with the help of human bones. To be specific, we first investigate a new set of bone-conduction theories. Through this set of theories, we combine a strong anti-noise cyclic neural network to achieve a high-precision QWERTY keyboard recognition for text input. Furthermore, in order to improve the user experience, we leverage slightly keyboard layout changing, dimensionality and feature selection to reduce the power consumption while preserving the convenience and stability. We have conducted experiments on 30 volunteers. The results show that Oinput has superior robustness with a high recognition accuracy of 93.3% in average. Moreover, Oinput's calibration mechanism increases the accuracy by more than 99%.

EMG is becoming an emerging interface for human-computer interface and has been applied to gesture recognition in previous work. However, those existing EMG-based interfaces can only recognize gestures at a coarse-grained level such as hand and arm gestures, which constraints their usage in applications involving fine-grained activities such as text entry via keystrokes. As a result, in this paper, we attempt to push the limit of existing EMG-based interfaces and propose the first wearable text-entry system, named ArmIn, with EMG signals. ArmIn is designed to recognize keystroke gestures with the help of a finger on printed and physical keyboards. We implement ArmIn using commodity EMG sensors and custom hardware board, and conduct experiments to evaluate its performance. By carefully designing the data processing scheme, ArmIn can recognize keystrokes on both kinds of keyboard, with 89.5% and 87.5% accuracy respectively, when it is worn on a user's left arm.

Recently, tiny smart devices have become increasingly popular in our lives because of their neat features and stylish appearance. However, their small form factors, especially screens, make it inconvenient for users to enter texts with conventional methods such as soft keyboards, which need a fairly large screen. To address this problem, we propose a novel texts-input system, called EchoType, with which users can enter texts with a finger writing in the air. EchoType makes use of acoustic sensors (i.e., microphone and speaker) to sense finger gestures and infer texts based on mapping relation between gestures and basic letters. We take a step to enable users to input texts with acoustic signals. Compared with existing approaches, EchoType enjoys merits of low hardware requirements and high scalability to different mobile devices.

Nowadays, smart wristbands have become one of the most prevailing wearable devices as they are small and portable. However, due to the limited size of the touch screens, smart wristbands typically have poor interactive experience. There are a few works appropriating the human body as a surface to extend the input. Yet by using multiple sensors at high sampling rates, they are not portable and are energy-consuming in practice. To break this stalemate, we proposed a portable, cost efficient text-entry system, termed ViType, which firstly leverages a single small form factor sensor to achieve a practical user input with much lower sampling rates. To enhance the input accuracy with less vibration information introduced by lower sampling rate, ViType designs a set of novel mechanisms, including an artificial neural network to process the vibration signals, and a runtime calibration and adaptation scheme to recover the error due to temporal instability. Extensive experiments have been conducted on 30 human subjects. The results demonstrate that ViType is robust to fight against various confounding factors. The average recognition accuracy is 94.8% with an initial training sample size of 20 for each key, which is 1.52 times higher than the state-of-the-art on-body typing system. Furthermore, when turning on the runtime calibration and adaptation system to update and enlarge the training sample size, the accuracy can reach around 98% on average during one month.

We are on the threshold of the explosive growth in the global Internet-of-Things (IoT) market. Comparing with the legacy human-centric applications, machine type communication scenarios exhibit totally different characteristics, such as low throughput, delay insensitivity, occasional transmission, and deep coverage. Meanwhile, it also requires the terminal devices to be cheap enough and sustain long battery life. These demands hasten the prosperity of low power wide area (LPWA) technologies. Narrowband IoT (NB-IoT) is the newest Long Term Evolution (LTE) specification ratified by the third generation partner project as one of the LPWA solutions to achieve the objectives of super coverage, low power, low cost, and massive connection. Working in the licensed frequency band, it is designed to reuse and coexist with the existing LTE cellular networks, which endows it with outstanding advantages in decreasing network installation cost and minimizing product time-to-market. In this backdrop, it has been extensively regarded as one of the most promising technologies toward the IoT landscape. However, as a new LTE standard, there are still a lot of challenges that need to overcome. This paper surveys its evolutions, technologies, and issues, spanning from performance analysis, design optimization, combination with other leading technologies, to implementation and application. The goal is to deliver a holistic understanding for the emerging wireless communication system, in helping to spur further research in accelerating the broad use of NB-IoT.

Non-orthogonal multiple access (NOMA) is envisioned to be one of the most beneficial technologies for next generation wireless networks due to its enhanced performance compared with other conventional radio access techniques. Although the principle of NOMA allows multiple users to use the same frequency resource, due to decoding complication, the information of users in practical systems cannot be decoded successfully if many of them use the same channel. Consequently, assigned spectrum of a system needs to be split into multiple subchannels in order to multiplex that among many users. Uplink resource allocation for such systems is more complicated compared with the downlink ones due to the individual users' power constraints and the discrete nature of subchannel assignment. In this paper, we propose an uplink subchannel and power allocation scheme for such systems. Due to the NP-hard and non-convex nature of the problem, the complete solution, that optimizes both subchannel assignment and power allocation jointly, is intractable. Consequently, we solve the problem in two steps. First, based on the assumption that the maximal power level of a user is subdivided equally among its allocated subchannels, we apply many-to-many matching model to solve the subchannel-user mapping problem. Then, in order to enhance the performance of the system further, we apply iterative water-filling and geometric programming two power allocation techniques to allocate power in each allocated subchannel-user slot optimally. Extensive simulation has been conducted to verify the effectiveness of the proposed scheme. The results demonstrate that the proposed scheme always outperforms all existing works in this context under all possible scenarios.

The wireless local area networks (WLANs) have been widely deployed with the rapid development of mobile devices and have further been brought into new applications with infrastructure mobility due to the growth of unmanned aerial vehicles (UAVs). However, the WLANs still face persistent challenge on increasing the network throughput to meet the customer’s requirement and fight against the node mobility. Interference is a well‐known issue that would degrade the network performance due to the broadcast characteristics of the wireless signals. Moreover, with infrastructure mobility, the interference becomes the key obstacle in pursuing the channel capacity. Legacy interference management mechanism through the channel access control in the MAC layer design of the 802.11 standard has some well‐known drawbacks, such as exposed and hidden terminal problems, inefficient rate adaptation, and retransmission schemes, making the efficient interference management an everlasting research topic over the years. Recently, interference management through exploiting physical layer mechanisms has attracted much research interest and has been proven to be a promising way to improve the network throughput, especially under the infrastructure mobility scenarios which provides more indicators for node dynamics. In this paper, we introduce a series of representative physical layer techniques and analyze how they are exploited for interference management to improve the network performance. We also provide some discussions about the research challenges and give potential future research topics in this area.

Localization in complex indoor environments with RF signal is a challenging task. Via this technique, the signal is easily affected by obstacles and environmental noise due to the broadcast nature of RF signal transmission. In this paper, we observe that seismic signal is more oriented than the RF signal, and thus is more suitable for localization in an complex indoor environment with obstacles. Motivated by this observation, we propose a device-free passive indoor localization system, namely Floc, that can fight against environmental impact and achieve high localization accuracy. FLoc is composed of three modules, which are sensing module that collects seismic signal from footsteps, footstep detection module that remove the environmental impact and recover the clean footsteps, and localization module that leverages seismic signal from footsteps for positioning. We implement FLoc on credit-card sized single-board computer Raspberry Pi equipped with geophones. To verify the effectiveness of our system, we conduct extensive experiments for different scenario in a complex indoor environment with the area of 6 × 8 square meter. The experimental results demonstrates that Floc can achieve up to 7cm localiztion accuracy on average, and can outperform the existing acoustic signal-based localization techniques.

This paper represents the design, feasibility evaluation and performance validation of ICMR, a novel cross layer protocol that can maximize concurrent transmissions and avoid data frame interference in wireless networks, achieving higher throughput comparing with the 802.11 standard and other state-of-the-art protocols. Observations on the 802.11 standard reveal that nodes degrade the network throughput from two aspects, including the so-called CF-CA problem and varied-IR problem, and these problems will make nodes around both the transmitter and receiver of the ongoing link waste concurrent transmission opportunities. A state-of-the-art protocol IRMA is proposed to improve the network throughput through solving the two problems at the transmitter side. In this paper, a new ICMR protocol is proposed to solve both problems at the receiver side to further improve the network throughput through discernible interference cancellation, a physical layer mechanism that can successfully detect data frames when collided by control frames. Hardware experiments based on USRP2 demonstrate the feasibility of the discernible interference cancellation mechanism, and simulations based on ns-2 confirm that ICMR outperforms the 802.11 standard and other protocols significantly.

Interference management through power control in wireless ad hoc networks has both the hidden terminal problem which induces collisions, and the exposed terminal problem which prohibits concurrent transmissions. Both problems are caused by the varied interference range induced by the adjusted transmission power. Through observing that the nodes adopt the power control mechanism induces collisions in one scenario and miss concurrent transmission opportunities in two scenarios, this paper presents IAPC (Interference-Aware Power Control), a novel protocol to improve the network throughput from these aspects. IAPC makes the interference range of each link covered by its CTS (Clear-To-Send) transmission through utilizing a signature detection process, so as to avoid interference. Meanwhile, it lets CTS convey the transmission power information of this link to make the neighboring node determine a limited transmission power, which tries to make the neighboring node outside the interference range of the ongoing link, thus can exploit concurrent transmissions. Simulation results based on ns-2 show that IAPC can outperform the other protocols significantly.

Conflagration is one of the major disasters that threatens human life and property. If the proper action is not taken in detecting the symptom of conflagration events ahead of time, the number of such disasters will keep increasing. An effective solution in this context will alleviate many fire-related global problems to a great extent. Although fire detectors are not available in many places, WiFi networks are increasingly prevalent nowadays. Motivated by the previous works that used WiFi signals for the purpose of environment monitoring and activity recognition, we make an attempt to use WiFi signals to detect fire. Through several experiments, we find that fire influences the transmission of wireless signals uniquely, and consequently it affects the amplitude and phase of the resultant Channel State Information (CSI). Based on this observation, in this paper, we propose a device-free fire detection system, namely Wi-Fire, using commercial WiFi devices. To the best of our knowledge, this is the first work that leverages CSI of radio frequency (RF) signal to detect fire events using existing wireless infrastructure without requiring any additional device. We implement our proposed system on desktop computers equipped with commercial 802.11n network interface cards (NICs). Comprehensive experiments have been conducted for different scenarios in different environments to verify the effectiveness of our proposed system. The results verify that the fire detection accuracy of this training-based system is up to 96.67% on average.

Blind mobility aid is a primary part in the daily life of blind people. Although plenty of systems or devices are invented to make the navigation of blind people easier, those are generally expensive and hardly affordable for them. To solve these issues, we introduce ABAid, a novel system designed for blind or visually impaired people to navigate, with commercial off-the-shelf (COTS) mobile devices. Based on in-depth acoustic localization and gyroscope techniques, this system is not only the means of huge convenience to carry, but also is capable of detecting obstacles before reaching them. In our experiments designed to detect the distance of wall, the proposed system achieves 3.24% average error rate. It can further measure the direction of wall, and the average error in this case is 2.73°. With high accuracy and stable measurement, ABAid is able to help blind people move independently in fairly uncomplicated scenarios.

Object differentiation plays a vital role in our daily life and such systems are widely deployed with RFID tags or bar codes attached on goods. In certain scenarios, however, attaching tags to objects may be impractical due to cost and protection issues. In this paper, we propose TagFree, a novel object differentiation scheme without attaching tags. Instead of relying on external tags, we exploit the inherent radiometric properties of different objects as their signatures. To improve the robustness and efficiency of TagFree, we empirically determine a spatial safe zone and harness successive cancellation to distinguish multiple objects simultaneously. We prototype TagFree on commercial WiFi infrastructure and evaluate its performance in various indoor scenarios. Experimental results demonstrate that TagFree achieves single object distinguishing accuracy of 96% measured at the same location, and over 80% within the safe zone range of up to 3m along a 7m link. TagFree can also differentiate up to 3 objects with acceptable accuracy.

The wearable devices are small and easy to carry but typically with poor interaction experience. For example, Apple iWatch does not support instant text message input feature because of the lack of keyboard availability on the tiny touch screen. To address this problem, we develop a novel system, termed iKey, which enables users to use the back of one of their hands as virtual keyboard for wearable wristbands. iKey recognizes keystrokes based on a location-based training model via body vibration. We will demonstrate a real time functional prototype of iKey in this demo.

Packet corruption caused by collision is a critical problem that hurts the performance of wireless networks. Conventional medium access control (MAC) protocols resort to collision avoidance to maintain acceptable efficiency of channel utilization. According to our investigation and observation, however, collision avoidance comes at the cost of miscellaneous overhead, which oppositely hurts channel utilization, not to mention the poor resiliency and performance of those protocols in face of dense networks or intensive traffic. Discovering the ability to tolerate collisions at the physical layer implementations of wireless networks, we in this paper propose Coco, a protocol that advocates simultaneous accesses from multiple senders to a shared channel, i.e., optimistically allowing collisions instead of simply avoiding them. With a simple but effective design, Coco addresses the key challenges in achieving collision tolerance, such as precise sender alignment and the control of transmission concurrency. We implement Coco in 802.15.4 networks and evaluate its performance through extensive experiments with 21 TelosB nodes. The results demonstrate that Coco is light-weight and enhances channel utilization by at least 20 percent in general cases, compared with state-of-the-arts protocols.

Fixed channelization configuration in today's wireless devices falls inefficient in the presence of growing data traffic and heterogeneous devices. In this regard, a number of fairly recent studies have provided spectrum adaptation capabilities for current wireless devices, however, they are limited to inband adaptation or incur substantial coordination overhead. The target of this paper is to fill the gaps in spectrum adaptation by overcoming these limitations. We propose SEER, a frame-level wideband spectrum adaptation solution which consists of two major components: i) a specially-constructed preamble that can be detected by receivers with arbitrary RF bands, and ii) a spectrum detection algorithm that identifies the desired transmission band in the context of multiple asynchronous senders by exploiting the preamble's temporal and spectral properties. SEER can be realized on commodity radios, and can be easily integrated into devices running different PHY/MAC protocols. We have prototyped SEER on the GNURadio/USRP platform to demonstrate its feasibility. Furthermore, using 1.6GHz channel measurements and trace-driven simulations, we have evaluated the merits of SEER over state-of-the-art approaches.

Injuries that are caused by falls have been regarded as one of the major health threats to the independent living for the elderly. Conventional fall detection systems have various limitations. In this work, we first look for the correlations between different radio signal variations and activities by analyzing radio propagation model. Based on our observation, we propose WiFall, a truly unobtrusive fall detection system. WiFall employs physical layer Channel State Information (CSI) as the indicator of activities. It can detect fall of the human without hardware modification, extra environmental setup, or any wearable device. We implement WiFall on desktops equipped with commodity 802.11n NIC, and evaluate the performance in three typical indoor scenarios with several layouts of transmitter-receiver (Tx-Rx) links. In our area of interest, WiFall can achieve fall detection for a single person with high accuracy. As demonstrated by the experimental results, WiFall yields 90 percent detection precision with a false alarm rate of 15 percent on average using a one-class SVM classifier in all testing scenarios. It can also achieve average 94 percent fall detection precisions with 13 percent false alarm using Random Forest algorithm.

Gesture recognition has emerged recently as a promising application in our daily lives. Owing to low cost, prevalent availability, and structural simplicity, RFID shall become a popular technology for gesture recognition. However, the performance of existing RFID-based gesture recognition systems is constrained by unfavorable intrusiveness to users, requiring users to attach tags on their bodies. To overcome this, we propose GRfid, a novel device-free gesture recognition system based on phase information output by COTS RFID devices. Our work stems from the key insight that the RFID phase information is capable of capturing the spatial features of various gestures with low-cost commodity hardware. In GRfid, after data are collected by hardware, we process the data by a sequence of functional blocks, namely data preprocessing, gesture detection, profiles training, and gesture recognition, all of which are well-designed to achieve high performance in gesture recognition. We have implemented GRfid with a commercial RFID reader and multiple tags, and conducted extensive experiments in different scenarios to evaluate its performance. The results demonstrate that GRfid can achieve an average recognition accuracy of 96.5 and 92.8 percent in the identical-position and diverse-positions scenario, respectively. Moreover, experiment results show that GRfid is robust against environmental interference and tag orientations.

Recent advances in rateless codes facilitate the utilization of dense constellation. By exploiting more constellation points with controlled symbol distance, they ensure higher transmission date rate and better diversity gain against fading. In this paper, we observe that there exists certain redundancy in the existing dense constellation diagram. With proper design, such redundancy can be can be optimized, and exploited to enable a new orthogonal dimension for multiple access. We termed it as hash division multiple access (HDMA). HDMA incorporates a orthogonal, linear and random encoder in PHY layer to construct orthogonal hash space for user separation. An AP-driven MAC protocol is then proposed to fully utilize this transmission concurrency for uplink and downlink multiple access. We verify the feasibility of HDMA via a GNU radio testbed, and further conduct trace-driven simulations to evaluate the multiplexing gain of HDMA. The results reveal that HDMA provides a maximum number of 3 to 5 concurrent transmissions, with performance gain over 131% and 39% compared to 802.11 and AutoMAC.

Emerging Wi-Fi Standards have incorporated growing channel widths to meet the proliferation of wireless services. Wider channel, unfortunately, increases the probability of partial channel overlap, and thus introduces more adjacent channel interference (ACI) than ever. Prior work either tries to avoid ACI, or neglects it in a brute-force way. In this paper, we carefully investigate the interference pattern at physical layer (PHY), and propose interference randomization (InterRandom) to strategically harness ACI for simultaneous transmission. The key insight behind InterRandom is to randomize the interference according to the overlapped ratio, and utilize coding redundancy to recover the corrupted data. To demonstrate the effectiveness of InterRandom, we further propose an InterRandom-enabled Media Access Control layer (MAC) protocol to facilitate Wi-Fi infrastructure mode transmission. We verify the feasibility of InterRandom on GNU radio testbed. Furthermore, our trace-driven simulations show that, InterRandom-aware MAC achieves 190% throughput gain compared to the legacy 802.11ac.

With an increasing need for worldwide counter-terrorism, security is becoming a top priority on the agenda of local governments. Governments has enforced patrols, security forces, and security equipments such as door frame detectors, screens and hand-held detectors in public places like railway stations and airports. The main purpose of these security equipments is to detect potentially dangerous metal objects. And the way to detect metal is mainly based on electromagnetic induction and X-rays. On the one hand, detectors based on electromagnetic induction needs to be placed close to the examined objects and on the other hand, detectors based on X-rays are usually large, expensive, complex, harmful and the detecting distance is limited on many occasions. For the above reasons, current metal detection equipments cannot be deployed on a large scale with a low cost scheme. But with the popular use of Wi-Fi in public areas, low-cost and large-scale deployment is possible if we could use Wi-Fi to detect metal. Since the dielectric constant and the reflection coefficient of metal are different from those of nonmetal objects, the effect of Wi-Fi signal reflection caused by metal objects is different from nonmetal items as well. Thus we attempt to use this observation to enable metal detection. But using Wi-Fi to detect metal will encounter some challenges. Factors such as the interference caused by multi-path effect and other signal sources, shapes, materials and surface smoothness of the target, as well as the angle of reflection will cause great impact on a detecting result. However, we implement a system by using a series of methods to eliminate most of the interferences and use machine learning algorithm to identify metal. We implement this system in commercial products. We conduct several experiments and their average accuracy and false alarm are respectively more than 90% and less than 10% in 1 meter detecting distance.

Ocean sensor networks have many applications ranging from oceanographic data collection, pollution monitoring, disasters prevention to tactical surveillance. However, due to the harsh ocean environment, it is difficult to deploy three-dimensional sensor networks underwater for sustainable monitoring tasks. In this paper, we propose a novel floating three-dimensional sensor networks for ocean monitoring and surveillance applications, which leverage nodes' restricted movement to enlarge the monitoring area. The networks are deployed on both the surface of the ocean and underwater, with many chains of nodes attached to mooring lines. Compared with deployments requiring self-adjusted nodes, our scheme is simpler and more cost-efficient, and the sustainable monitoring cost is low. We design deployment strategies, and provide in-depth mathematical analysis for network coverage. Then we derive the minimum number of sensors to achieve full surface coverage. Evaluated by simulations of a few underwater scenarios, the effectiveness and efficiency of the proposed scheme is proved.

With the development of acoustic localization schemes using mobile devices, keystroke detection has received tremendous attention from the academia and industry. Currently, most of the existing systems focus on the recognition of a single keystroke and they are limited by several restrictions. In this paper, we consider the idea of signal variation caused by the combination of two combined keystrokes, and propose an acoustic-based scheme that can detect the combined keystrokes effectively. Our system exploits the blind signal separation technique to deal with the mixed signals, resultant from typing two separate keys simultaneously. Then, we apply feature extraction and pattern recognition algorithms to recognize the combined keystrokes. Extensive experiments have been conducted in a laboratory environment with mobile phones equipped with two microphones. Our results show that for several combinations of two keystrokes, on average, we can achieve 78.4% recognition accuracy.

A typical scenario for public WLANs is large audience environment where Wi-Fi hotspots serve scores of mobile devices. The performance of those Wi-Fi hotspots is extremely poor in terms of low goodput and severe delay due to heavy contention and MAC inefficiency. After carefully investigating the traffic patterns in public WLANs, we proposeCarpool, a practical design that facilitates transmission sharing among multiple receivers, to tackle this problem. The key idea is to reduce contention by feeding frames for multiple destinations into one transmission at physical layer (PHY). As such, each downlink transmission carries payloads for multiple receivers, which reduces contention overhead and enables in-time response to concurrent requests from multiple users. To achieve efficient and reliable transmission in Carpool, we propose i) a lightweight frame structure to support multiple receivers, and ii) a real-time channel estimation scheme to continuously calibrate channel estimation during the transmission of a Carpool frame. We have implemented the entire PHY of Carpool on the GNURadio/USRP platform and tested it in various indoor environments. Furthermore, our trace-driven MAC evaluation shows that Carpool achieves up to$3.2 \times$goodput gain and reduces up to$75$percent delay compared to the IEEE 802.11n MAC frame aggregation scheme.

Channel condition varies frequently in wireless networks. To achieve good performance, devices need rate adaptation. In rate adaptation, choosing proper modulation schemes based on channel conditions is vital to the transmission performance. However, due to the natural character of discrete modulation types and continuous varied link conditions, we cannot make a one-to-one mapping from modulation schemes to channel conditions. This matching gap causes either over-select or under-select modulation schemes which limits throughput performance. To fill-in the gap, we propose time-line modulation (TiM), a novel three-Dimensional modulation scheme by adding time dimension into current amplitude-phase domain schemes. With estimation of channel condition, TiM changes base-band data transmission time by artificially interpolating values between original data points without changing amplitude-phase domain modulation type. We implemented TiM on USRP2 and conducted comprehensive simulations. Results show that, compared with rate adaptation choosing from traditional modulation schemes, TiM can improve channel utilization up to 200 percent.

Recent literature advances Wi-Fi signals to “see” people's motions and locations. This paper asks the following question: Can Wi-Fi “hear” our talks? We present WiHear, which enables Wi-Fi signals to “hear” our talks without deploying any devices. To achieve this, WiHear needs to detect and analyze fine-grained radio reflections from mouth movements. WiHear solves this micro-movement detection problem by introducing Mouth Motion Profile that leverages partial multipath effects and wavelet packet transformation. Since Wi-Fi signals do not require line-of-sight, WiHear can “hear” people talks within the radio range. Further, WiHear can simultaneously “hear” multiple people's talks leveraging MIMO technology. We implement WiHear on both USRP N210 platform and commercial Wi-Fi infrastructure. Results show that within our pre-defined vocabulary, WiHear can achieve detection accuracy of 91 percent on average for single individual speaking no more than six words and up to 74 percent for no more than three people talking simultaneously. Moreover, the detection accuracy can be further improved by deploying multiple receivers from different angles.

Seeing through walls and knowing clearly what exist inside just like a superman are not only fantastic wishes for humans, but also of much practical significance. For example, you would like to know whether there are pipes, or rebars inside a wall before drilling into it. Moreover, knowing how pipes are configured in a wall before attempting to fix defects would definitely prevent unnecessary damages. Existing methods that intend to address this issue are either costly due to the use of high-end technology, or restrictive for reasons of some strong assumptions. However, in this paper, we present a novel system, SmartScanner, which is based on off-the-shelf sensors embedded in a smartphone. SmartScanner makes full use of in-built sensors, namely, the accelerometer, gyroscope, and magnetometer to achieve this goal inexpensively and conveniently. Specifically, by combining these sensors, we are able to clearly distinguish certain objects inside a wall and map out the layout of an in-wall pipeline system. We implement SmartScanner on two smartphone platforms, namely iPhone 4 and Xiaomi Mi2S, and conduct extensive experiments to evaluate its performance. Experiments show that SmartScanner can achieve high accuracies in distinguishing objects in various scenarios. Meanwhile, as for layout mapping, 90 percent of length errors are limited to several centimeters for horizontal and vertical pipeline segments, respectively. Also, SmartScanner can achieve centimeter-level position errors of turning points in horizontal and vertical directions in the testbed.

Rate adaptation is an essential component in today's wireless standards, which help approach the channel capacity and maximize the throughput. However, how to estimate the optimal data rate in a fluctuated channel remains of great concern. Previous wisdoms leverage PHY layer information for rate estimation, including confidence information like SoftPHY hints, and channel state information (CSI) measurements. However, when experiencing rapid time varying and frequency selective fading channel, the above metrics can be inaccurate. The reason roots from the fact that there are not enough cost-efficient pilots, which are pre-known symbols inserted in a packet for channel estimation. In this paper, we observe that by digging into both PHY layer decoder and upper layer protocol headers, more reliable data bits with high confidence level can be exploited. These data bits, termed smart pilot, can be used to calibrate the channel estimation measurements cost-efficiently. Based on the calibrated estimation, we further propose a novel greedy rate selection algorithm to track the optimal data rate, which successfully avoids the impact of deep fading subcarriers in both legacy 802.11a/g and 802.11n MIMO systems. Our experiments on GNU radio testbed show that SmartPilot quickly tracks the link variance, and improve the channel estimation accuracy by 87%. Furthermore, the trace driven simulation reveals that greedy rate selection algorithm predicts the data rate as good as the optimal rate adaptation algorithms for 802.11 standards.

Dynamic video streaming protocols allow video clients to adaptively choose video rates to improve viewer Quality-of- Experience (QoE). Existing works on rate adaptation choose the video rate according to network capacity or buffer state, however, they didn't consider the early quitting problem. Most viewers quit before the video ends, in which case optimal solutions for complete video sessions may become suboptimal. In this paper, we design a video rate adaptation scheme which considers incomplete video sessions and aims at maximizing the quality of the video chunks that are actually watched by viewers. We formulate the video rate adaptation problem as a Markov Decision Process (MDP), in which the viewers' decision of quitting depends on the video quality and the rebuffer time. To solve the MDP problem, we propose two online learning algorithms for regular viewers and new viewers with no viewing history, respectively. The simulation results show that the proposed scheme increases the video quality {truly} consumed by the viewers, thus improving the viewer QoE.

Most recently, gesture recognition has increasingly attracted intense academic and industrial interest due to its various applications in daily life, such as home automation, mobile games. Present approaches for gesture recognition, mainly including vision-based, sensor-based and RF-based, all have certain limitations which hinder their practical use in some scenarios. For example, the vision-based approaches fail to work well in poor light conditions and the sensor-based ones require users to wear devices. To address these, we propose WiG in this paper, a device-free gesture recognition system based solely on Commercial Off-The-Shelf (COTS) WiFi infrastructures and devices. Compared with existing Radio Frequency (RF)-based systems, WiG stands out for its systematic simplicity, extremely low cost and high practicability. We implemented WiG in indoor environment and conducted experiments to evaluate its performance in two typical scenarios. The results demonstrate that WiG can achieve an average recognition accuracy of 92% in line-of-sight scenario and average accuracy of 88% in the none-line-of sight scenario.

The proliferation of WiFi hotspots in public places enables ubiquitous Internet access. These public WiFi hotspots usually serve scores of mobile devices and suffer from extremely poor performance in terms of low good put and severe delay. In this paper, we first study the traffic characteristics in public WiFi networks, and demonstrate that the main causes of such poor performance are media access control (MAC) inefficiency and downlink-uplink traffic asymmetry. To cope with these issues, we call attention to transmission carpool, which facilitates an access point (AP) to send multiple frames for different mobile stations (STAs) in a single transmission. It reduces contention and conveys more frames in each channel access. As such, each downlink transmission carries more payload and thus improves efficiency and solves traffic asymmetry simultaneously.

Partially overlapped channels are barely used for concurrent transmission in WiFi networks, since they lead to collisions where the collided packets cannot be decoded successfully. In this paper, we observe that the actual corrupted symbols by partial-channel interference in OFDM-based WiFi networks are not as severe as we expected. There remains extra coding redundancy that can be exploited from the corrupted symbols, and utilized for packet recovery. Accordingly, we present a novel paradigm termed Piros, in order to Push the lImits of partially concurrent transmission in WiFi networks. Piros strategically leverages the coding redundancy according to the overlap portion in a distributed manner, and extracts useful decoding information from the corrupted symbols to decode the packet with partial-channel interference.

Fixed channelization configuration in today's wireless devices falls inefficient in the presence of growing data traffic and heterogeneous devices. In this regard, a number of fairly recent studies have provided spectrum adaptation capabilities for current wireless devices, however, they are limited to inband adaptation or incur substantial coordination overhead. The target of this paper is to fill the gaps in spectrum adaptation by overcoming these limitations. We propose Seer, a frame-level wideband spectrum adaptation system which consists of two major components: i) a specially-constructed preamble that can be detected by receivers with arbitrary RF bands, and ii) a spectrum detection algorithm that identifies the intended transmission band in the context of multiple asynchronous senders by exploiting the preamble's temporal and spectral properties. Seer can be realized on commodity radios, and can be easily integrated into devices running different PHY/MAC protocols. We have prototyped Seer on the GNURadio/USRP platform and tested it under various environments. Furthermore, our evaluation using 1.6GHz spectrum measurements shows that Seer largely improves system throughput over fixed channel configuration and state-of-the-art spectrum adaptation approaches.

Reliable people counting is crucial to many urban applications. However, most existing people counting systems are sensor-based and can only work in some fixed gateways or checkpoints where sensors have been installed. This high dependence on the exact locations of sensors leads to low accuracy. To overcome these limitations, in this paper, we propose a smartphone-based people counting system, Wi-Counter, by leveraging the pervasive Wi-Fi infrastructure. To collect comprehensive Wi-Fi signals and people count information based on crowdsource, Wi-Counter first adopts a preprocessor to overcome the noisy, discrepant, and fragile data based on the Wiener filter and Newton interpolation. It then makes use of the designated five-layer neural network to learn the relation model between the Wi-Fi signals and the number of people. By analyzing the received Wi-Fi signals, Wi-Counter can estimate the number of people based on the resulting model. We have conducted experiments by implementing a prototype of Wi-counter based on smartphones and evaluated the system in terms of accuracy and power consumption in an indoor testbed covering an area of 96 m $^2$. Wi-Counter achieved a counting accuracy of up to 93% and exhibited reliable and robust performance resisting temporal environmental changes with negligible power usage.

Recently, there is a massive growth in the amount of wireless video traffics. To increase the network efficiency and cater to the needs of Quality of Experience (QoE), researchers propose hybrid MAC schemes, with multiple MAC protocols anchored in a single MAC layer. Different MAC protocols are selected for diverse users' requirements. However, the existing hybrid MAC mainly combines various MAC protocols in time domain. Only one type of traffic can be satisfied at a time, and others have to endure poor QoE. Meanwhile, the channel resources are not fully utilized due to coarse usage in time domain. Motivated by this, we propose FC-MAC, a Fine-grained Cognitive MAC for video streaming in wireless networks. Instead of dividing channel time into slots, FC-MAC splits the channel frequency into fine-grained subchannels, and hybrid different MAC protocols in frequency domain. By dynamically adjusting the bandwidth and the MAC protocols according to the users' needs and channel condition, FC-MAC ensures the QoE at low cost and achieves high multiplexing gain. We conduct extensive simulations to verify the effectiveness of FC-MAC. Numerous results show that compared with time domain hybrid MAC, FC-MAC achieves 190% performance gain.

For wireless service providers (WSP), the emerging femtocell market brings new opportunities as well as challenges. It is difficult for a WSP to have complete information of the technical strength of other WSP in the market, and it is hard to estimate the user demand, which is influenced by the price and service quality of all competing WSP in the market. To address these problems, in this paper, we propose an economic framework for the WSP to maximize their utility, via a joint pricing and spectrum allocation strategy, under the condition of incomplete information of other rival WSP in the market. We study two scenarios: 1) all WSP enter the market at the same time; 2) the WSP enter the market at different times. In both scenario, we formulate the problem as a Bayesian game, and derive the Bayesian Nash equilibrium. The simulation results verify that the proposed joint pricing and spectrum allocation strategy outperforms sole pricing or sole spectrum allocation strategy. Interestingly, when the WSP enter the market at different times, the later entered WSP chooses more aggressive pricing and spectrum allocation strategy than the early entered WSP.

Channel condition varies frequently in wireless networks. To achieve good performance, devices need rate adaptation. In rate adaptation, choosing proper modulation schemes based on channel conditions is vital to the transmission performance. However, due to the natural character of discrete modulation types and continuous varied link conditions, we cannot make a one-to-one mapping from modulation schemes to channel conditions. This matching gap causes either over-select or under-select modulation schemes which limits throughput performance. To fill-in the gap, we propose TiM (Time-line Modulation), a novel 3-Dimensional modulation scheme by adding time dimension into current amplitude-phase domain schemes. With estimation of channel condition, TiM changes base-band data transmission time by artificially interpolating values between original data points without changing amplitude-phase domain modulation type. We implemented TiM on USRP2 and conducted comprehensive simulations. Results show that, compared with rate adaptation choosing from traditional modulation schemes, TiM can improve channel utilization up to 200%.

Newly popular indoor location-based services (ILBS), when integrated with commerce and public safety, offer a promising land for wireless indoor localization technologies. WLAN is suggested to be one of the most potential candidates owing to its prevalent infrastructure (i.e., access points (APs)) and low cost. However, the overall performance can be greatly degraded by the spatial localizability variance problem, i.e., the localization accuracy across various locations may have significant differences given any fixed AP deployment. As a result, it brings in user experience inconsistency which is unfavorable for ILBS. In this paper, we propose NomLoc - an indoor localization system using nomadic APs to address the performance variance problem. The key insight of NomLoc is to leverage the mobility of nomadic APs to dynamically adjust the WLAN network topology. A space partition (SP)-based localization algorithm is tailored for NomLoc to perform calibration-free positioning. Moreover, fine-grained channel state information (CSI) is employed to mitigate the performance degradation of the SP-based method due to multipath and none-line-of-sight (NLOS) effects. We have implemented the NomLoc system with off-the-shelf devices and evaluated the performance in two typical indoor environments. The results show that NomLoc can greatly mitigate spatial localizability variance and improve localization accuracy with the assistance of nomadic APs as compared with the corresponding static AP deployment. Moreover, it is robust to the position error of nomadic APs.

Rate adaptation is an essential component in today's wireless standards, for its ability to adaptively approach the channel capacity, and maximize the system throughput. The difficulty in rate adaptation stems from estimating the optimal data rate in a fluctuated channel. Previous wisdoms leverage PHY layer information for rate estimation, such as Soft PHY hints or Channel State Information. These information solely comes from one same layer, which are insufficient to track the optimal data rate. We observe that by investigating the information in both PHY layer decoder and upper layer protocol headers, more pilots can be exploited to estimate the optimal data rate across both time and frequency domain. These smart pilots help remove the residual channel effect and calibrate the CSI with minimum overhead. Based on the calibrated CSI, we propose a novel greedy rate selection algorithm to harness frequency diversity, which obtains the optimal data rate over all the subcarriers. Our experiments on GNU radio test bed show that Smart Pilot quickly tracks the link variance, and reduces the residual channel effect by 87%. Further, the trace driven simulation reveals that greedy rate selection algorithm predicts the data rate as good as the optimal rate adaptation algorithms for 802.11 standards.

Recently, directional antenna has shown great potential to deploy in indoor environments and significantly improve wireless network capacity. Due to its directionality, the orientation of antenna becomes a critical issue especially for moving devices while exchanging data with directional AP. Previous work largely focused on scenarios, in which the clients are stable at most of the time and addressed the issue of moving device by remeasurement of the whole network information such as received signal strength, conflict graph. However, re-collection of such information will introduce significant overhead. To adjust directional antenna quickly and wisely in real-time, in this paper, we introduce ADAS, a system that adjust orientation of directional antenna with sensor hints from mobile devices. The key idea of ADAS system is to obtain the movement behavior and location information from sensors equipped in mobile devices, then the AP adapts its orientation accordingly. We implement ADAS system on commercial directional antenna and evaluate its performance under different configurations. The experiment results demonstrate that directional AP in ADAS system could adapt to mobility with less overhead.

Ocean covers nearly 71% of our planet's surface, yet 95% of the ocean remains unexplored by human being, and wireless sensor networks are envisioned to perform monitoring tasks over the large portion of our world. However, deploying wireless sensor networks on the sea poses many challenges and for maritime surveillance security applications we may need to deploy sensors both on the sea surface and underwater for three-dimensional detection. In this paper, we propose a hybrid ocean sensor networks called Double-head maritime Sensor Networks (DSNs), which combine the advantages of wireless sensor networks and underwater acoustic sensor networks. By leveraging the unique characteristics of DSNs, we design a localization scheme LDSN which is consisted of two algorithms SML and FLA. We first use SML to localize moored anchor nodes as seed nodes. After the underwater sensor networks have been localized, the floating double-head nodes can figure out its instant position via FLA algorithm. We evaluate the scheme by simulations and the results show that the scheme can achieve a high localization accuracy.

Due to the rapid growth of the smartphone applications and the fast development of the Wireless Local Area Networks (WLANs), numerous indoor location-based techniques have been proposed during the past several decades. Floorplan, which defines the structure and functionality of a specific indoor environment, becomes a hot topic nowadays. Conventional floorplan techniques leverage smartphone sensors combined with WiFi signals to construct the floorplan of a building. However, existing approaches with sensors cannot detect the shape of a corner, and the sensors cost huge amount of energy during the whole floorplan constructing process. In this paper, we propose a sensor-free approach to detect the shape of a certain corner leveraging WiFi signals without using sensors on smartphones. Instead of utilizing traditional wireless communication indicator Received Signal Strength (RSS), we leverage a finer-grained indicator Channel State Information (CSI) to detect the shape of a certain corner. The evaluation of our approach shows that CSI is more robust in sensor-free corner shape detection, and we have achieved over 85% detection accuracy in simulation and over 70% detection accuracy in real indoor experiments.

Sensing data from mobile phones provide us exciting and profitable applications. Recent research focuses on sensing indoor environment, but suffers from inaccuracy because of the limited reachability of human traces or requires human intervention to perform sophisticated tasks. In this paper, we present ShopProfiler, a shop profiling system on crowdsourcing data. First, we extract customer movement patterns from traces. Second, we improve accuracy of building floor plan by adopting a gradient-based approach and then localize shops through WiFi heat map. Third, we categorize shops by designing an SVM classifier in shop space to support multi-label classification. Finally, we infer brand name from SSID by applying string similarity measurement. Based on over five thousand traces in three big malls in two different countries, we conclude that ShopProfiler achieves better accuracy in building refined floor plan, and characterizes shops in terms of location, category and name with little human intervention.

The world population is in the midst of a unique and irreversible process of aging. Fall, which is one of the major health threats and obstacles to independent living of elders, will aggravate the global pressure in elders' health care and injury rescue. Thus, automatic fall detection is highly in need. Current proposed fall detection systems either need hardware installation or disrupt people's daily life. These limitations make it hard to widely deploy fall detection systems in residential settings. In this work, we analyze the wireless signal propagation model considering human activities influence. We then propose a novel and truly unobtrusive detection method based on the advanced wireless technologies, which we call as WiFall. WiFall employs the time variability and special diversity of Channel State Information (CSI) as the indicator of human activities. As CSI is readily available in prevalent in-use wireless infrastructures, WiFall withdraws the need for hardware modification, environmental setup and worn or taken devices. We implement WiFall on laptops equipped with commercial 802.11n NICs. Two typical indoor scenarios and several layout schemes are examined. As demonstrated by the experimental results, WiFall yielded 87% detection precision with false alarm rate of 18% in average.

Wireless video stream delivery is choppy. This problem becomes much severer in MU-MIMO's simultaneous video transmission within the same band. Conventional schemes achieve graceful video delivery by harnessing from high data redundancy. However, with concurrent transmission in the same band, the leverage of high data redundancy leads to high probability of collisions and packets loss, which limits the performance. In concurrent video transmission, achieving efficiency over varied link condition is the main issue. To address this issue, this paper presents SimCast (Simultaneous), a cross-layer design for achieving efficient concurrent video uploading/downloading in MU-MIMO WLANs. The key idea of SimCast is to harness frequency diversity of the channel and spatial similarity of users. We implemented SimCast on USRP2 and conducted extensive simulations. Result shows that SimCast achieves higher throughput than traditional schemes by 1.2× on average. Video quality of SimCast outperforms competitive schemes, which is up to 5 dB in PSNR.

Seeing through walls and knowing clearly what exist inside just like a superman are not only fantastic wishes for humans, but also of much practical significance. For example, you would like to know whether there are pipes, or rebars inside a wall before drilling into it. Moreover, knowing how pipes are configured in a wall before attempting to fix defects would definitely prevent unnecessary damages. Existing methods that intend to address this issue are either costly due to the use of high-end technology, or too restrictive for reasons of some strong assumptions. However, in this paper, we present a novel system, SmartSening, which is based on off-the-shelf sensors embedded in smartphones. SmartSensing makes full use of in-built sensors, namely, the accelerometer, the gyroscope, and the magnetometer to achieve this goal inexpensively and conveniently. Specifically, by combining these sensors, we are able to clearly distinguish certain objects inside a wall. In addition, the layout of a pipeline system can be mapped out automatically in an economical and laborsaving way. We implement this system on two different kinds of smartphone platforms, namely iPhone4 and Xiaomi Mi2S. We conduct experiments in a proof-of-concept testbed of size 1.8m×1.0m. Experimental results show that SmartSensing can achieve no less than an average accuracy of 96%, 89% and 77% in distinguishing objects under three different depths, respectively. Also, as for layout mapping, it can achieve less than 32cm and 28cm length error with 90% probability on average for whole horizontal and vertical pipeline segments, with a 6.8m and 4.0m total length, respectively.

Recent literature advances Wi-Fi signals to "see" people's motions and locations. This paper asks the following question: Can Wi-Fi "hear" our talks? We present WiHear, which enables Wi-Fi signals to "hear" our talks without deploying any devices. To achieve this, WiHear needs to detect and analyze fine-grained radio reflections from mouth movements. WiHear solves this micro-movement detection problem by introducing Mouth Motion Profile that leverages partial multipath effects and wavelet packet transformation. Since Wi-Fi signals do not require line-of-sight, WiHear can "hear" people talks within the radio range. Further, WiHear can simultaneously "hear" multiple people's talks leveraging MIMO technology. We implement WiHear on both USRP N210 platform and commercial Wi-Fi infrastructure. Results show that within our pre-defined vocabulary, WiHear can achieve detection accuracy of 91% on average for single individual speaking no more than 6 words and up to 74% for no more than 3 people talking simultaneously. Moreover, the detection accuracy can be further improved by deploying multiple receivers from different angles.

To accommodate the soaring mobile broadband traffic, the Federal Communications Commission (FCC) in the U.S. sets out to retrieve under-utilized spectrum (e.g., TV Whitespace) and lay the groundwork for spectrum redistribution. Auction is an efficient way to allocate resources to those who value them the most. The large pool of spectrums to be released, especially the ones in TV Whitespace, consist of wide-range frequencies. Apart from spatial reuse, spectrum heterogeneity imposes new challenges for spectrum auction design: 1) Wireless service providers with different targeted cell coverages have different spectrum frequency preferences; 2) interference relationship is frequency-dependent due to frequency-selective signal fading. Unfortunately, existing spectrum auction mechanisms either assume spectrum valuation is homogeneous or use homogeneous interference graph to group buyers who can reuse the same spectrum. In this paper, we propose TAMES, an auction framework for heterogeneous spectrum transaction. We consider a multi-seller-multi-buyer double auction, in which every buyer submits a bid, consisting of the spectrum demand and a bidding profile of prices for spectrums contributed by all sellers. A novel buyer grouping approach is proposed to tackle the problem of heterogeneous interference graph. TAMES is proved to be truthful as well as individually rational. The simulation results show that TAMES significantly improves spectrum utilization, sellers' revenue and buyers' utility by making smart use of spectrum heterogeneity, while keeping low running time comparable with existing auction mechanisms. Moreover, via simulation, we show how to help buyers obtain continuous spectrums which further improves buyers' satisfaction.

Radio frequency (RF) based technologies play an important role in indoor localization, since Radio Signal Strength (RSS) can be easily measured by various wireless devices without additional cost. Among these, radio map based technologies (also referred as fingerprinting technologies) are attractive due to high accuracy and easy deployment. However, these technologies have not been extensively applied on real environment for two fatal limitations. First, it is hard to localize multiple objects. When the number of target objects is unknown, constructing a radio map of multiple objects is almost impossible. Second, environment changes will generate different multipath signals and severely disturb the RSS measurement, making laborious retraining inevitable. Motivated by these, in this paper, we propose a novel approach, called Line-of-sight radio map matching, which only reserves the LOS signal among nodes. It leverages frequency diversity to eliminate the multipath behavior, making RSS more reliable than before. We implement our system MODLoc based on TelosB sensor nodes and commercial 802.11 NICs with Channel State Information (CSI) as well. Through extensive experiments, it shows that the accuracy does not decrease when localizing multiple targets in a dynamic environment. Our work outperforms the traditional methods by about 60 percent. More importantly, no calibration is required in such environment. Furthermore, our approach presents attractive flexibility, making it more appropriate for general RF-based localization studies than just the radio map based localization.

Improving channel utilization is a well-known issue in wireless networks. In traditional point-to-point wireless communication, significant efforts had been made by the existing studies on enhancing the utilization of the channel access time. However, in the emerging wireless network using MU-MIMO, considering only the time domain in channel utilization is not sufficient. As multiple transmitters are allowed to transmit packets simultaneously to the same AP, allowing more antennas at AP would lead to higher channel utilization. Thus, the channel utilization in MU-MIMO should consider both time and spatial domains, i.e., the channel access time and the antenna usage, which have not been considered in the existing methods. In this paper, we point out that the fundamental problem is lacking of the antenna information of contention nodes in channel contention. To address this issue, we propose a new MAC-PHY architecture design, CUTS, to allow distributed nodes effectively contend for the channel and utilize the channel in both domains. Particularly, CUTS adopts interference nulling to attach the antenna information in channel contention. Meanwhile, techniques such as channel contention in frequency domain and ACK in frequency domain using self-jamming are adopted. Through the software defined radio-based real experiments and extensive simulations, we demonstrate the feasibility of our design and illustrate that CUTS provides better channel utilization with the gain over IEEE 802.11 reaching up to 470 percent.

Energy efficiency is the main concern in wireless sensor networks (WSNs) due to devices' limited battery power. Because the heavy burden of nodes that near the sink, this "energy hole problem" makes nodes near the sink have faster energy depletion than others. Because of this, the lifetime of WSNs, to some extent, is determined by the power consumption of communication between sink and sensing nodes that near the sink. To address this issue, we propose CSMA/SF (Shortest First) protocol to reduce power consumption of sink-node communication by minimizing energy cost in carrier sense during nodes' channel contention. CSMA/SF modifies existed CSMA/CA MAC protocol. Instead of complete contention-based, CSMA/SF ensures nodes remaining shorter message has higher priority in contention by implementing a distributed scheduling algorithm and incorporating Length Detection scheme. Further, CSMA/SF employs an Anti-Starvation mechanism to solve the starvation problem of shortest-first protocol. CSMA/SF also optimizes channel utilization by reducing the probability of collisions. We have implemented CSMA/SF into USRP2 platform and also conducted comprehensive simulations. The experimental results show that CSMA/SF can reduce overall energy consumption by around 20%. CSMA/SF can improve channel utilization up to 40%.

It is known that current fixed spectrum assignment policy has made the spectrum resource significantly underutilized. As a promising solution, cognitive radio emerges and shows its advantages. It allows the unlicensed users to opportunistically access the spectrum not used by the licensed users. To ensure that the unlicensed users can identify the vacate spectrum fast and accurately without interfering the licensed users, cooperative sensing is explored to improve the sensing performance by leveraging spatial diversity. However, cooperation gain can be compromised dramatically with cooperation overhead. Furthermore, when sensing decisions are made, contention on spectrum access also contributes a lot to the control overhead, especially in the distributed networks. Motivated by this, we propose a novel MAC design, termed Frequency domain Cooperative sensing and Multi-channel contention (FCM) for Cognitive Radio Ad Hoc Networks (CRAHNs). FCM is proposed for OFDM (Orthogonal Frequency Division Multiplexing) modulation based communication systems, which moves cooperative sensing and multi-channel contention from time domain into frequency domain. Therefore, control overhead caused by cooperation and contention can be significantly reduced. Meanwhile, the sensing and access performance can be both guaranteed. Extensive simulation results show that FCM can effectively reduce the control overhead, and improve the average throughput by 220% over Traditional Cooperative MAC for CRAHNs.

Many emerging applications such as intruder detection and border protection drive the fast increasing development of device-free passive (DfP) localization techniques. In this paper, we present Pilot, a Channel State Information (CSI)-based DfP indoor localization system in WLAN. Pilot design is motivated by the observations that PHY layer CSI is capable of capturing the environment variance due to frequency diversity of wideband channel, such that the position where the entity located can be uniquely identified by monitoring the CSI feature pattern shift. Therefore, a ``passive'' radio map is constructed as prerequisite which include fingerprints for entity located in some crucial reference positions, as well as clear environment. Unlike device-based approaches that directly percepts the current state of entities, the first challenge for DfP localization is to detect their appearance in the area of interest. To this end, we design an essential anomaly detection block as the localization trigger relying on the CSI feature shift when entity emerges. Afterwards, a probabilistic algorithm is proposed to match the abnormal CSI to the fingerprint database to estimate the positions of potential existing entities. Finally, a data fusion block is developed to address the multiple entities localization challenge. We have implemented Pilot system with commercial IEEE 802.11n NICs and evaluated the performance in two typical indoor scenarios. It is shown that our Pilot system can greatly outperform the corresponding best RSS-based scheme in terms of anomaly detection and localization accuracy.

Packet corruption caused by collision is a critical problem that hurts the performance of wireless networks. Conventional medium access control (MAC) protocols resort to collision avoidance to maintain acceptable efficiency of channel utilization. According to our investigation and observation, however, collision avoidance comes at the cost of miscellaneous overhead, which oppositely hurts channel utilization, not to mention the poor resiliency and performance of those protocols in face of dense networks or intensive traffic. Discovering the ability to tolerate collisions at the physical layer implementations of wireless networks, we in this paper propose Coco, a MAC protocol that advocates simultaneous accesses from multiple senders to a shared channel, i.e., optimistically allowing collisions instead of simply avoiding them. With a simple but effective design, Coco addresses the key challenges in achieving collision tolerance, such as precise sender alignment and fine control of the transmission concurrency. We implement Coco in 802.15.4 networks and evaluate its performance through extensive experiments with 21 TelosB nodes. The results demonstrate that Coco is light-weight and enhances channel utilization by at least 20% in general cases, compared with state-of-the-arts protocols.

Spectrums are heterogeneous, especially from the aspect of their central frequency. According to signal propagation properties, low-frequency spectrum generally has lower path loss, thus longer transmission range, compared with high-frequency spectrum. Cellular operators with different targeted cell size will have different preferences for spectrums with different frequencies. Furthermore, the transmission range also affects the interference relationships among transmitters. Transmitters who can reuse the same high-frequency spectrum may interfere with each other when reusing the low-frequency spectrum, so it is difficult to decide how to construct the interference graph to exploit spectrum reusability among transmitters. Auction is considered as an efficient way for spectrum allocation. However, most of the previous works only considered homogenous spectrum auction, failing to address the problem of spectrum heterogeneity. In this paper, we propose TAMES, a Truthful Auction Mechanism for hEterogeneous Spectrum allocation, which allows buyers to freely express their different preferences towards different spectrums. Frequency-specific interference graphs are constructed to determine buyer groups. The proposed heterogeneous spectrum auction is theoretically proved to be truthful and individual rational. The simulation results verifies that the proposed auction mechanism outperforms other auction mechanisms with homogenous bid or homogenous interference graph. The proposed auction mechanism is able to yield higher buyers' satisfaction, seller's revenue and spectrum utilization.

Improving channel utilization is a well-known issue in wireless networks. In traditional point-to-point wireless communication, significant efforts had been made by the existing study on enhancing the utilization of the channel access time. However, in the emerging wireless network using MU-MIMO, considering only the time domain in channel utilization is not sufficient. As multiple transmitters are allowed to transmit packets simultaneously to the same AP, allowing more antennas at AP would lead to higher channel utilization. Thus the channel utilization in MU-MIMO should consider both time and spatial domains, i.e., the channel access time and the antenna usage, which has not been considered in the existing methods. In this paper, we point out that the fundamental problem is lacking of the antenna information of contention nodes in channel contention. To address this issue, we propose a new MAC-PHY architecture design, CUTS, to utilize the channel in both domains. Particularly, CUTS adopts interference nulling to attach the antenna information in channel contention. Meanwhile, techniques such as channel contention in frequency domain and ACK in frequency domain using self-jamming are adopted. Through the software defined radio based real experiments and extensive simulations, we demonstrate the feasibility of our design and illustrate that CUTS provides better channel utilization with the gain over IEEE 802.11 reaching up to 470%.

Effective coordination can dramatically reduce radio interference and avoid packet collisions for multistation wireless local area networks (WLANs). Coordination itself needs consume communication resource and thus competes with data transmission for the limited wireless radio resources. In traditional approaches, control frames and data packets are transmitted in an alternate manner, which brings a great deal of coordination overhead. In this paper, we propose a new communication model where the control frames can be "attachedâ to the data transmission. Thus, control messages and data traffic can be transmitted simultaneously and consequently the channel utilization can be improved significantly. We implement the idea in OFDM-based WLANs called hJam, which fully explores the physical layer features of the OFDM modulation method and allows one data packet and a number of control messages to be transmitted together. hJam is implemented on the GNU Radio testbed consisting of eight USRP2 nodes. We also conduct comprehensive simulations and the experimental results show that hJam can improve the WLANs efficiency by up to 200 percent compared with the existing 802.11 family protocols.

Leveraging concurrent transmission is a promising way to improve throughput in wireless networks. Existing media access control (MAC) protocols like carrier sense multiple access always try to minimize the number of concurrent transmissions to avoid collision, although collisions at sender sides are harmless to the overall performance. The reason for such conservative strategy is that those protocols cannot obtain accurate channel status (who is transmitting and receiving) with low cost. They can only avoid potential collisions through rough channel status (idle or busy). To obtain additional information in a cost-efficient way, we propose a novel coding scheme, Attachment Coding, to allow control information to be “attached” on data packet. Nodes then transmit two kinds of signals simultaneously, without degrading the effective throughput of the original data traffic. Based on Attachment Coding, we propose an Attached-RTS MAC (AR-MAC) to exploit exposed terminals for concurrent transmissions. The attached control information provides accurate channel status for nodes in real time. Therefore, nodes can identify exposed terminals and utilize them for concurrent transmission. We theoretically analyze the feasibility of Attachment Coding, and implement it on the GNU Radio testbed to further verify it. We also conduct extensive simulations to evaluate the performance of Attached-RTS. The experimental results show that by leveraging Attachment Coding, AR-MAC achieves up to 180 percent in dense deployed ad hoc networks.

Indoor positioning systems have received increasing attention for supporting location-based services in indoor environments. WiFi-based indoor localization has been attractive due to its open access and low cost properties. However, the distance estimation based on received signal strength indicator (RSSI) is easily affected by the temporal and spatial variance due to the multipath effect, which contributes to most of the estimation errors in current systems. In this work, we analyze this effect across the physical layer and account for the undesirable RSSI readings being reported. We explore the frequency diversity of the subcarriers in orthogonal frequency division multiplexing systems and propose a novel approach called FILA, which leverages the channel state information (CSI) to build a propagation model and a fingerprinting system at the receiver. We implement the FILA system on commercial 802.11 NICs, and then evaluate its performance in different typical indoor scenarios. The experimental results show that the accuracy and latency of distance calculation can be significantly enhanced by using CSI. Moreover, FILA can significantly improve the localization accuracy compared with the corresponding RSSI approach.

Wireless technology has become ever more popular in recent years, which results in a higher and higher density of wireless devices. In order to cope with this high density, researchers are proposing the provision of multiple concurrent transmissions by dividing a broadband channel into separate narrow band subchannels. In particular, a fine-grained channel access approach calls for efficient channel allocation mechanisms, especially in distributed networks. However, most of the current multi-channel access methods rely on costly coordination, which significantly degrades network performance. Motivated by this, we propose a cross layer design, termed Attachment Learning (AT-Learning), to achieve multi-channel allocation with low cost and high efficiency in distributed OFDMA based networks. AT-Learning utilizes a jamming and cancellation technique to attach identifier signals to data traffic, without degrading the effective throughput of the original data transmission. These identifier signals help mobile stations learn the allocation strategy by themselves. After the learning stage, mobile stations can achieve a TDMA-like performance, where stations will know exactly when to transmit and on which channel without further collisions. We conduct comprehensive simulations, comparing AT-Learning with a traditional multi-channel access method like Slotted ALOHA. The experimental results demonstrate that AT-Learning can improve the throughput by up to 300% over Slotted ALOHA.

Radio spectrum resource is shown to be significantly underutilized with fixed spectrum assignment policy. As a promising solution, cognitive radio allows unlicensed users to opportunistically access the spectrum not used by the licensed users. Cooperative sensing is further exploited to improve the sensing performance of unlicensed users by leveraging spatial diversity. However, cooperation gain can be compromised dramatically with cooperation overhead. Furthermore, when sensing decisions are made, contention on spectrum access also becomes an overhead, especially in the distributed networks. Motivated by this, we propose a novel MAC design, termed Frequency domain Cooperative sensing and Multi-channel contention (FCM). FCM moves cooperative sensing and multi-channel contention from time domain into frequency domain. Thus, the control overhead caused by cooperation and contention can be significantly reduced, without reducing the sensing and access performance. Extensive simulation results show that FCM can effectively reduce the control overhead, and improve the average throughput by 220% over Traditional Cooperative MAC for CRAHNs.

This paper presents the design, implementation, and evaluation of FAST (Full-duplex Attachment System), a cross layer system to solve both Hidden terminal problem and exposed terminal problems. The main reason that CSMA-like protocol cannot well solve hidden and exposed terminal problem both at once lies in the fact that they cannot obtain accurate Channel Usage Information (CUI, who is transmitting or receiving nearby) with low cost. FAST successfully obtains cost CUI in a cost-efficient way, thus can know exactly what are the neighborhood doing currently. FAST includes Attachment Coding in PHY layer to provide cost-effective CUI, and Attachment Sense in MAC layer to utilize CUI to identify hidden and exposed nodes in real time. Extensive simulation results show FAST can well solve both hidden and exposed terminal problems, and improves an average throughput to 200% over CSMA in practical ad-hoc networks.

WLAN-based indoor location fingerprinting has been attractive owing to the advantages of open access and high accuracy. Most fingerprinting-based systems so far rely on the received signal strength (RSS), which can be easily measured at the receiver with commercial WLAN equipment. However, RSS is a coarse value which simply measures the received power for a whole channel. Thus, it fluctuates over time in typical indoor environments with rich multipath effects and not unique for a specific location. In this paper, we present the design, implementation, and evaluation of a Fine-grained Indoor Fingerprinting System (FIFS). FIFS explores a PHYlayer Channel State Information (CSI) that specifies the channel status over all the subcarriers for location fingerprinting in WLAN. The system leverages the CSI values including different amplitudes and phases at multiple propagation paths, known as the frequency diversity, to uniquely manifest a location. Moreover, the multiple antennas provides the spatial diversity that can be further augmented in fingerprinting. We also present a coherence bandwidth-enhanced probability algorithm with a correlation filter to map object to the fingerprints. We conducted experiments in two typical indoor scenarios with commercial IEEE 802.11 NICs. The experimental results demonstrate that the overall positioning accuracy can be improved compared with the RSS-based Horus system.

Device-free passive (Dfp) motion detection seeks to monitor the position change of entities without actively carrying any physical devices. Recently, WLAN with a rich set of installed wireless infrastructures enables motion detection in the area of interest. WLAN-enabled DfP motion detection rely on received signal strength (RSS) is verified to be able to provide acceptable high accuracy. Although RSS can be easily measured with commercial equipments, it is suspectable to measurement itself due to multipath effect in indoor environment. In this paper, we present an Indoor device-free Motion Detection system (FIMD) to overcome the preceding RSS-based limitation. FIMD explores properties of Channel State Information (CSI) from PHY layer in OFDM system. FIMD is designed based on the insight that CSI maintains temporal stability in static environment, while exhibits burst patterns when motion takes place. Motivated by this observation, FIMD uses a novel feature extracted from CSI to leverage its temporal stability and frequency diversity. The motion detection is conducted with outliers identification from normal features in continuous monitoring using density-based DBSCAN algorithm. Moreover, we leverage two schemes including false alert filter and data fusion to enhance the detection accuracy. We implement FIMD system with commercial IEEE 802.11n NICs and evaluate its performance in two typical indoor scenarios. Experiment results show that FIMD can achieve high detection rate. Moreover, comparing with RSSI, the feature extracted from CSI enables better detection performance in accuracy and robustness to narrowband interference.

Nowadays, cyber-physical system (CPS) relies on wireless networks for devices control and information backhaul. But the mass deployment CPS devices make operators' spectrum scarce situations even more worse. Hence, cellular network operators anticipate the Dynamic Spectrum Access (DSA) technology to solve the spectrum shortage problem in the context of cognitive radio (CR). Femtocells, acting as gateways in CPS, integrate CPS devices into cellular networks in a seamless manner. The concept of cognitive femtocell can solve the spectrum congestion problem even within a massive network on the CPS scale. However, in practical systems, cellular white space spectrum should be quantitatively measured to guide cognitive femtocell access algorithms design. We are the first to conduct a comprehensive measurement study for the purpose of measurement, discovery and model features of GSM white space spectrum. We evaluate availabilities of extra 21.4 MHz capacity in GSM white space spectrum as a reason of artificial GSM network spectrum planning. In our study, we find out that perfect results can hardly be obtained because of inherent measurement trade-offs, even when extremely high sweep speed of 16 GHz/s is applied at the receiver. Based on statistical analysis of real-scene traces, we propose an Efficient Duty Cycle (EDC) model to accurately characterize the white space in GSM network by considering miss-detection probabilities. Cross-validating evaluation results show that the EDC model can well decrease interference probabilities at high time-granularity measurement periodicity scenarios. Our results confirm the feasibility of cognitive femtocells access in an intra-operator scenario and can be applied to future wireless networks.

Effective coordination can dramatically reduce radio interference and avoid packet collisions for multi-station wireless local area networks (WLANs). Coordination itself needs consume communication resource and thus competes with data transmission for the limited wireless radio resources. In traditional approaches, control frames and data packets are transmitted in an alternate manner, which brings a great deal of coordination overhead. In this paper we propose a new communication model where the control frames can be “attached” to the data transmission. Thus, control messages and data traffic can be transmitted simultaneously and consequently the channel utilization can be improved significantly. We implement the idea in OFDM-based WLANs called hJam, which fully explores the physical layer features of the OFDM modulation method and allows one data packet and a number of control messages to be transmitted together. hJam is implemented on the GNU Radio testbed consisting of eight USRP2 nodes. We also conduct comprehensive simulations and the experimental results show that hJam can improve the WLANs efficiency by up to 72% compared with the existing 802.11 family protocols.

Indoor positioning systems have received increasing attention for supporting location-based services in indoor environments. WiFi-based indoor localization has been attractive due to its open access and low cost properties. However, the distance estimation based on received signal strength indicator (RSSI) is easily affected by the temporal and spatial variance due to the multipath effect, which contributes to most of the estimation errors in current systems. How to eliminate such effect so as to enhance the indoor localization performance is a big challenge. In this work, we analyze this effect across the physical layer and account for the undesirable RSSI readings being reported. We explore the frequency diversity of the subcarriers in OFDM systems and propose a novel approach called FILA, which leverages the channel state information (CSI) to alleviate multipath effect at the receiver. We implement the FILA system on commercial 802.11 NICs, and then evaluate its performance in different typical indoor scenarios. The experimental results show that the accuracy and latency of distance calculation can be significantly enhanced by using CSI. Moreover, FILA can significantly improve the localization accuracy compared with the corresponding RSSI approach.

Interference is a critical issue in wireless communications. In a typical multiple-user environment, different users may severely interfere with each other. Coordination among users therefore is an indispensable part for interference management in wireless networks. It is known that coordination among multiple nodes is a costly operation taking a significant amount of valuable communication resource. In this paper, we have an interesting observation that by generating intended patterns, some simultaneous transmissions, i.e., "interference,” can be successfully decoded without degrading the effective throughput in original transmission. As such, an extra and "free” coordination channel can be built. Based on this idea, we propose a DC-MAC to leverage this "free” channel for efficient medium access in a multiple-user wireless network. We theoretically analyze the capacity of this channel under different environments with various modulation schemes. USRP2-based implementation experiments show that compared with the widely adopted CSMA, DC-MAC can improve the channel utilization efficiency by up to 250 percent.

IEEE 802.15.4 standard specifies physical layer (PHY) and medium access control (MAC) sublayer protocols for low-rate and low-power communication applications. In this protocol, every 4-bit symbol is encoded into a sequence of 32 chips that are actually transmitted over the air. The 32 chips as a whole is also called a pseudonoise code (PN-Code). Due to complex channel conditions such as attenuation and interference, the transmitted PN-Code will often be received with some PN-Code chips corrupted. In this paper, we conduct a systematic analysis on these errors occurring at chip level. We find that there are notable error patterns corresponding to different cases. We then show that recognizing these patterns enables us to identify the channel condition in great details. We believe that understanding what happened to the transmission in our way can potentially bring benefit to channel coding, routing, and error correction protocol design. Finally, we propose Simple Rule, a simple yet effective method based on the chip error patterns to infer the link condition with an accuracy of over 96 percent in our evaluations.

RFID has been gaining popularity due to its variety of applications, such as inventory control and localization. One important issue in RFID system is tag identification. In RFID systems, the tag randomly selects a slot to send a Random Number (RN) packet to contend for identification. Collision happens when multiple tags select the same slot, which makes the RN packet undecodable and thus reduces the channel utilization. In this paper, we redesign the RN pattern to make the collided RNs decodable. By leveraging the collision slots, the system performance can be dramatically enhanced. This novel scheme is called DDC, which is able to directly decode the collisions without exact knowledge of collided RNs. In the DDC scheme, we modify the RN generator in RFID tag and add a collision decoding scheme for RFID reader. We implement DDC in GNU Radio and USRP2 based testbed to verify its feasibility. Both theoretical analysis and testbed experiment show that DDC achieves 40 percent tag read rate gain compared with traditional RFID protocol.

RFID tag identification is a crucial problem in UHF RFID systems. Traditional tag identification algorithms can be classified into two categories, ALOHA-based and tree-based. Both of them are inefficient due to the incidental high coordination cost. In this paper, we bring CSMA into UHF RFID systems to enhance tag read rate by reducing coordination cost. However, it is not straightforward due to the simple hardware design of passive RFID tags, which is unable to sense the transmissions or collisions of other tags. To tackle this challenge, we propose receiver-based CSMA (RCSMA) in this paper. In RCSMA, the reader notifies the tags channel condition. According to different sensing results of reader's notifications, the tags take corresponding actions, e.g., random back off. RCSMA does not require special RFID tag hardware design. An absorbing Markov chain model is presented to analyze the performance of RCSMA and shown to be consistent with the simulation results. Compared with optimized ALOHA-based algorithms and optimized tree-based algorithms, RCSMA can enhance the tag read rate by 30-70 percent under different reader and tag data rates.

Surveillance is a critical problem for harbor protection, border control or the security of commercial facilities. The effective protection of vast near-coast sea surfaces and busy harbor areas from intrusions of unauthorized marine vessels, such as pirates smugglers or, illegal fishermen is particularly challenging. In this paper, we present an innovative solution for ship intrusion detection. Equipped with three-axis accelerometer sensors, we deploy an experimental Wireless Sensor Network (WSN) on the sea's surface to detect ships. Using signal processing techniques and cooperative signal processing, we can detect any passing ships by distinguishing the ship-generated waves from the ocean waves. We design a three-tier intrusion detection system with which we propose to exploit spatial and temporal correlations of an intrusion to increase detection reliability. We conduct evaluations with real data collected in our initial experiments, and provide quantitative analysis of the detection system, such as the successful detection ratio, detection latency, and an estimation of an intruding vessel's velocity.

The hidden terminal problem is known to degrade the throughput of wireless networks due to collisions, while the exposed terminal problem results in poor performance by wasting valuable transmission opportunities. As a result, extensive research has been conducted to solve these two problems, such as Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA). However, CSMA-like protocols cannot solve both of these two problems at once. The fundamental reason lies in the fact that they cannot obtain accurate Channel Usage Information (CUI, who is transmitting or receiving nearby) with a low cost. To obtain additional CUI in a cost-efficient way, we propose a cross layer design, FAST (Full-duplex Attachment System). FAST contains a PHY layer Attachment Coding, which transmits control information independently on the air, without degrading the effective throughput of the original data traffic, and a MAC layer Attachment Sense, which utilizes the PHY layer control information to identify the hidden and exposed nodes in real time. We theoretically analyze the feasibility of the Attachment Coding, and then implement it on a GNU Radio testbed consisting of eight USRP2 nodes. We also conduct extensive simulations to evaluate the performance of FAST, and the experimental results show that FAST can effectively solve both the hidden and the exposed terminal problems, and improve the average throughput by up to 200% over CSMA in practical ad-hoc networks.

Surveillance is a vital problem for harbor protection, border control or the security of other commercial facilities. It is particularly challenging to protect the vast near-coast sea surface and busy harbor areas from intrusions of unauthorized marine vessels, such as trespassing boats and ships. In this paper, we present an innovative solution for ship intrusion detection. Equipped with three-axis accelerometer sensors, we deploy an experimental wireless sensor network on the sea surface to detect ships. Using signal processing techniques and cooperative signal processing, we can detect the passing ships by distinguishing the ship-generated waves and the ocean waves. We design an intrusion detection system in which we propose to exploit spatial and temporal correlations of the intrusion to increase detection reliability. We conduct evaluations with real data collected by our initial experiments, and provide quantitative analysis on the detection system, such as the successful detection ratio and the estimation of the intruding ship velocity.

Wireless technologies have gained tremendous popularity in recent years, resulting in a dense deployment of wireless devices. Therefore, it is desired to provide multiple concurrent transmissions by dividing a broadband channel into separate sub channels. This fine-grained channel access calls for efficient channel allocation mechanisms, especially in distributed networks. However, most of the current multichannel access methods rely on costy coordination, which significantly degrade their performance. Motivated by this, we propose a cross layer design called Attachment Learning (AT-learning) in distributed OFDMA (Orthogonal Frequency Division Multiple Access) based networks. AT-learning utilizes jamming technique to attach identifier signals on data traffic, where the identifier signals can help mobile stations to learn allocation strategy by themselves. After the learning stage, mobile stations can achieve a TDMA-like performance, where stations can know when exactly to transmit on which channel without further collisions. We conduct comprehensive simulations and the experimental results show that AT-learning can improve the throughput by up to 300% compared with traditional multichannel access method which asks mobile stations to randomly choose channels without learning.

RFID has been gaining popularity due to its variety of applications, such as inventory control and localization. One important issue in RFID system is tag identification. In RFID systems, the tag randomly selects a slot to send a Random Number (RN16) packet to contend for identification. Collision happens when multiple tags select the same slot, which makes the RN packet undecodable and thus reduces the channel utilization. In this paper, we redesign the RN pattern to make the collided RNs decodable. By leveraging the collision slots, the system performance can be dramatically enhanced. This novel scheme is called DDC, which is able to directly decode the collisions without exact knowledge of collided RNs. In the DDC scheme, we modify the RN generator in RFID tag and add a collision decoding scheme for RFID reader. We implement DDC in GNU Radio and USRP2 based testbed to verify its feasibility. Both theoretical analysis and testbed experiment show that DDC achieves 40% tag read rate gain compared with traditional RFID protocol.

Recent years have seen an increasing need of wireless networks in a mobile environment serving for more complex tasks and applications. Mobility becomes an indispensable factor of the system design and has been widely recognized as a general cause of packet loss. Though many works have been done on mobility study, to the best of our knowledge, they are mainly based on simulations or analytical studies that assume idealized link conditions. In this work, we experimentally investigate the nature of the error characteristics of mobility-induced packet losses at "chip-level" in IEEE 802.15.4. We believe this more understanding of mobility-induced packet losses can bring great potential benefits for further study on channel coding, routing and protocol design. Toward this end, we design and implement an efficient algorithm to distinguish mobility-induced packet losses from other packet losses of static environments such as attenuation. Our algorithm is greatly advantaged as it needs no training data even when environment changes. Collecting three corrupted packets is sufficient to obtain a satisfactory performance. This feature makes our design in particular suitable for dynamic and mobile environments, allowing real-time mobility induced loss detection in an online manner. Experiments based on GNU Radio testbed show that our algorithm can provide an accuracy of up to 96%.

IEEE 802.15.4 standard specifies physical layer (PHY) and medium access control (MAC)sublayer protocols for low-rate and low-power communication applications. In this protocol, every 4-bit symbol is encoded into a sequence of 32 chips that are actually transmitted over the air. The 32 chips as a whole is also called a pseudo-noise code (PN-Code). Due to complex channel conditions such as attenuation and interference, the transmitted PN-Code will often be received with some PN-Code chips corrupted. In this paper, we conduct a systematic analysis on these errors occurring at chip-level. We find that there are notable error patterns corresponding to different cases. Recognizing these patterns will enable us to identify the channel condition in great details. We believe that understanding what happened to the transmission in our setup can potentially bring benefit to channel coding, routing and error correction protocol design.

Interference is a critical issue in wireless communications. In a typical multiple-user environment, different users may severely interfere with each other. Coordination among users therefore is an indispensable part for interference management in wireless networks. It is known that, coordination among multiple nodes is a costly operation taking a significant amount of valuable communication resource. In this paper, we have an interesting observation that by generating intended patterns, some simultaneous transmissions, i.e., "interference", can be successfully decoded without degrading the effective throughput in original transmission. As such, an extra and "free" coordination channel can be built. Based on this idea we propose a DC-MAC to leverage this "free" channel for efficient medium access in a multiple-user wireless network. We theoretically analyze the capacity of this channel under different environments with various modulation schemes. USRP2-based implementation experiments show that compared with the widely adopted CSMA, DC-MAC can improve the channel utilization efficiency by up to 250%.

Performance analysis for workflow models has been recognized as one of the most significant tasks in workflow management. Time performance evaluation methods are needed for workflow designers to estimate the average turnaround time of processes in the early stage of workflow system development. Based on extending fuzzy temporal workflow nets by introducing choice possibilities for transitions in structural conflict, a new workflow model named extended fuzzy temporal workflow nets (EFTWF-nets) has been proposed. And the calculation of temporal elements in EFTWF-nets is given. Then a decomposition algorithm for EFTWF-nets is put forward and a time performance evaluation method based on it is investigated. Finally a case study is given to illustrate how to use this approach