Model serving systems are critical for deep learning inference, managing GPU infrastructure to deliver end-to-end services. Current frameworks typically treat operators as basic compilation and scheduling units, which often fail to maximize GPU utilization due to hardware-unfriendly kernels and coarse-grained scheduling. To address these limitations, we propose a cooperative compilation and scheduling scheme that statically generates multiple kernel variants and dynamically schedules them based on runtime context. We present Infera, a high-performance model serving system that implements this approach. Experimental results demonstrate that Infera improves inference throughput by at least 1.6× compared to state-of-the-art baselines.

Rotation is a fundamental form of motion and rotation speed measurement holds paramount importance for assessing the health and performance of machinery with rotating components. However, existing measurement systems often face challenges such as limited measurement distance, low accuracy, and complex installation or maintenance processes. In this paper, we propose RoLEX, a LoRa-based rotation speed measurement system for long-distance and contactless monitoring of rotating machinery in ubiquitous scenarios. RoLEX employs a novel Signal Selection method to eliminate chirp interference and adapt to varying rotation speeds, along with a Boost Sensing method to enhance sampling rates and an advanced feature processing algorithm for precise rotation speed estimation and tracking. Comprehensive experiments validate that RoLEX achieves a measurement distance of 50 m, approximately 17 times farther than the latest wireless rotation speed measurement systems. Moreover, RoLEX is robust to interference and obstructions (including through-wall scenarios) and achieves an average measurement error less than 0.69% across different rotation speeds (100 - 5100 Revolutions Per Minute). For tracking performance, RoLEX achieves a relative error less than 2.8% in 90% of cases. We also present a case study to highlight RoLEX's practical applicability in real-world scenarios.

The rapid development of large language models (LLMs) has significantly advanced code completion capabilities, giving rise to a new generation of LLM-based Code Completion Tools (LCCTs). Unlike general-purpose LLMs, these tools possess unique workflows, integrating multiple information sources as input and prioritizing code suggestions over natural language interaction, which introduces distinct security challenges. Additionally, LCCTs often rely on proprietary code datasets for training, raising concerns about the potential exposure of sensitive data. This paper exploits these distinct characteristics of LCCTs to develop targeted attack methodologies on two critical security risks: jailbreaking and training data extraction attacks. Our experimental results expose significant vulnerabilities within LCCTs, including a 99.4% success rate in jailbreaking attacks on GitHub Copilot and a 46.3% success rate on Amazon Q. Furthermore, We successfully extracted sensitive user data from GitHub Copilot, including 54 real email addresses and 314 physical addresses associated with GitHub usernames. Our study also demonstrates that these code-based attack methods are effective against general-purpose LLMs, highlighting a broader security misalignment in the handling of code by modern LLMs. These findings underscore critical security challenges associated with LCCTs and suggest essential directions for strengthening their security frameworks.

Offsite-tuning is a privacy-preserving method for tuning large language models (LLMs) by sharing a lossy compressed emulator from the LLM owners with data owners for downstream task tuning. This approach protects the privacy of both the model and data owners. However, current offsite tuning methods often suffer from adaptation degradation, high computational costs, and limited protection strength due to uniformly dropping LLM layers or relying on expensive knowledge distillation. To address these issues, we propose ScaleOT, a novel privacy-utility-scalable offsite-tuning framework that effectively balances privacy and utility. ScaleOT introduces a novel layerwise lossy compression algorithm that uses reinforcement learning to obtain the importance of each layer. It employs lightweight networks, termed harmonizers, to replace the raw LLM layers. By combining important original LLM layers and harmonizers in different ratios, ScaleOT generates emulators tailored for optimal performance with various model scales for enhanced privacy protection. Additionally, we present a rank reduction method to further compress the original LLM layers, significantly enhancing privacy with negligible impact on utility. Comprehensive experiments show that ScaleOT can achieve nearly lossless offsite tuning performance compared with full fine-tuning while obtaining better model privacy.

Kai Yao, Zhaorui Tan, Penglei Gao, Lichun Li, Kaixin Wu, Yinggui Wang, Yuan Zhao, Yixin Ji, Jianke Zhu, Wei Wang. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Existing deep neural network (DNN) accelerator design automation (ADA) methods adopt architecture templates to predetermine parts of design choices and then explore the remaining design choices beyond templates. Based on the architecture hierarchy at the processing unit (PU) level, these templates can be classified into intra-PU templates and inter-PU templates. Since templates limit the flexibility of ADA, designing effective ADA methods without templates has become an important research topic. Although there have appeared some works to enhance the flexibility of ADA by removing intra-PU templates, to the best of our knowledge no existing works have studied ADA methods without inter-PU templates. ADA with predetermined inter-PU templates is typically inefficient in terms of resource utilization, especially for DNNs with complex topology. In this paper, we propose a novel method, called hardware computation graph (HCG), for ADA without inter-PU templates. In HCG, a novel inter-PU architecture exploration strategy is proposed to optimize on-chip memory utilization. This strategy mainly depends on an appearing-frequency guided pruning method and an appearing-frequency first generation method. Experiments show that HCG can achieve competitive latency while using only 13% 90% of on-chip memory, compared with existing state-of-the-art ADA methods.

This paper proposes CSS, a built-in channel state scrambling scheme to provide always-on protection for channel state information. CSS uses randomly generated scrambling vectors to emulate human activities, preventing eavesdroppers from recovering human physical activity from the channel state. By carefully designing the scrambling scheme based on physical channel models, we ensure that legacy receivers can successfully decode the frames and all frames transmitted over the air are scrambled. Furthermore, legitimate Wi-Fi sensors can still recover the true activity with a pre-shared secret key. Implementation on a real communication system shows that CSS can retain the same frame-error rate for commercial wireless receivers while misleading eavesdroppers with a success rate of over 95%.

Eye movements play a significant role in human-computer interaction and are widely recognized as an essential health indicator, making their detection both appealing and technically challenging. In this paper, we present a system named USEE that achieves high-precision capture of weak and aperiodic eye movements by utilizing fine-grained and ubiquitous ultrasound signals, capturing both blinking and more subtle saccades. We first identify signal changes associated with eye movements by capturing the unique impact of blinking. Further, we establish a pioneering relationship between the residuals from signal decomposition and subtle eye movements. Utilizing inno-vative signal processing architectures, we mitigate interference and effectively extract eye movement features. Subsequently, we employ one-dimensional convolutional operations in place of signal cross-correlation, designing filters for motion category identification and a lightweight convolutional neural network for saccade direction classification. This enables our system to serve as a foundational sensing layer for eye movement tracking, applicable across diverse applications. We implement USEE on both a research-purpose platform and a commodity Raspberry Pi. Extensive experimental results demonstrate the effectiveness of our system, achieving 91% accuracy in saccade recognition and 94% in blink detection. The system proves robust, even in challenging scenarios with strong interference, such as the presence of moving pedestrians.

Abstract In this article, we propose a mobile edge computing (MEC)‐related system named PD‐Gait, which can measure gait parameters of Parkinson's disease patients in a contactless and privacy‐preserving manner. We utilize inaudible acoustic signals and band‐pass filters to achieve privacy data protection in the physical layer. The proposed framework can be easily deployed in the mobile end of MEC, and hence release the edge server in cybersecurity attacks fighting. The gait parameters include stride cycle time length and moving speed, and hence providing an objective basis for the doctors' judgment. PD‐Gait utilizes acoustic signals in bands from 16 to 23 kHz to achieve device‐free sensing, which would release both doctors and patients from the tedious wearing process and psychological burden caused by traditional wearable devices. To achieve robust measurement, we propose a novel acoustic ranging method to avoid “broken tones” and “uneven peak distribution” in the received data. The corresponding ranging accuracy is 0.1 m. We also propose auto‐focus micro‐Doppler features to extract robust stride cycle time length, and can achieve an accuracy of 0.052 s. We deployed PD‐Gait in a brain hospital and collected data from 8 patients. The total walked distance is over 330 m. From the overall trend, our results are highly correlated with the doctor's judgment.

Acoustic ranging has been viewed as a promising Human-Computer Interaction (HCI) technology in many scenarios, such as Augmented Reality (AR)/Virtual Reality (VR) and smart appliances. Most ranging systems with distributed devices undergo an extra calibration process to remove the timing errors. However, the calibration process needs user intervention. Furthermore, it should assume that the clock drifts are linear and stable, which is disabled within tens of minutes. In this paper, we introduce a self-calibrated acoustic ranging system that achieves sub-millimeter accuracy on distributed asynchronous devices. Based on our theoretical timing model, we precisely cancel both the system delay and the nonlinear clock drift with carefully designed Orthogonal Frequency-Division Multiplexing (OFDM) ranging signals. Our synchronization scheme achieves a timing accuracy of 1.9 microseconds, which allows us to build large-scale virtual acoustic arrays. Based on such a calibration scheme, our localization system achieves a ranging error of$\rm{0.39}~mm$within three meters in real-world experiments.

With the rising of demands for novel Human-Computer Interaction (HCI) approaches in the 3D space, a number of intelligent approaches have been proposed to achieve the HCI by tracking the translation and rotation of the target devices. In this paper, we propose to realize a light-weight, battery-free, 3D motion tracking solution by leveraging a spinning linearly polarized antenna to track a passive RFID tag array. Instead of using the fixed antennas, which can only receive stable signal in some specific environments due to the unpredictable multipath effect, we propose to mitigate the multipath effect and the ambient interference by continuously spinning a linearly polarized antenna, and then extract the most distinctive features based on the optimal reading conditions of the spinning antenna. In particular, because the phase variation around the matching direction is more stable while the RSSI variation around the mismatching direction is more distinctive, we leverage such matching/mismatching property of the linearly polarized antenna to extract the most distinctive features for motion tracking. To depict the property, we build a theoretical model to explain the RSSI and the phase variation of the RFID tag along with the spinning of the antenna, and further extend the model from a single RFID tag to an RFID tag array. Based on the model, we can extract the distinctive RSSI features for the rotation tracking and the stable phase features for the translation tracking. Moreover, to tackle the low rate of feature extraction due to the spinning of antenna, we further propose to enhance the unstable phase features based on the overall trend of other tags with interpolation, such that the sampling rate can be efficiently improved. Finally, we propose a LSTM (Long Short Term Memory)-based network to track the 3D motion based on the signal features extracted based on the polarization model. The experimental results show that our system can achieve an average error of 10.45 cm in the translation tracking, and an average error of$6.02^\circ$in the rotation tracking in the 3D space.

We propose UltraCLR, a new contrastive learning framework that fuses dual modulation ultrasonic sensing signals to enhance gesture representation. Most existing ultrasound-based gesture recognition tasks rely on a large amount of manually labeled samples to learn task-specific representations via end-to-end training. However, they cannot exploit unlabeled continuous gesture signals that are easy to collect. Inspired by recent self-supervised learning techniques, UltraCLR aims to autonomously learn a ubiquitous gesture signal representation that can benefit all tasks from low-cost unlabeled signals. We use the STFT heatmap as a secondary input and leverage the contrastive learning framework to improve the high-quality Channel Impulsive Response heatmap input representations. The learned representations can better represent the spatial-position information and intermediate states of gesture movement. With the representation learned by UltraCLR, we can greatly reduce the complexity of downstream gesture recognition tasks so that they can be completed using a simple classifier trained with a small training set and a lower computational cost. Our experimental results show that UltraCLR outperforms state-of-the-art gesture recognition systems with only a few labeled samples and achieves more than 85% reduction in computational complexity and over 9× improvement in inference speed.

Deep learning models rely on highly optimized tensor libraries for efficient inference on heterogeneous hardware. Current deep compilers typically predetermine layouts of tensors and then optimize loops of operators. However, such unidirectional and one-off workflow strictly separates graph-level optimization and operator-level optimization into different system layers, missing opportunities for unified tuning.

The unmanned aerial vehicle (UAV) network is popular these years due to its various applications. In the UAV network, routing is significantly affected by the distributed network topology, leading to the issue that UAVs are vulnerable to deliberate damage. Hence, this paper focuses on the routing plan and recovery for UAV networks with attacks. In detail, a deliberate attack model based on the importance of nodes is designed to represent enemy attacks. Then, a node importance ranking mechanism is presented, considering the degree of nodes and link importance. However, it is intractable to handle the routing problem by traditional methods for UAV networks, since link connections change with the UAV availability. Hence, an intelligent algorithm based on reinforcement learning is proposed to recover the routing path when UAVs are attacked. Simulations are conducted and numerical results verify the proposed mechanism performs better than other referred methods.

This paper describes our submission to ICASSP 2023 MUG Challenge Track 4, Keyphrase Extraction, which aims to extract keyphrases most relevant to the conference theme from conference materials. We model the challenge as a single-class Named Entity Recognition task and developed techniques for better performance on the challenge: For the data preprocessing, we encode the split keyphrases after word segmentation. In addition, we increase the amount of input information that the model can accept at one time by fusing multiple preprocessed sentences into one segment. We replace the loss function with the multi-class focal loss to address the sparseness of keyphrases. Besides, we score each appearance of keyphrases and add an extra output layer to fit the score to rank keyphrases. Exhaustive evaluations are performed to find the best combination of the word segmentation tool, the pre-trained embedding model, and the corresponding hyperparameters. With these proposals, we scored 45.04 on the final test set.

Device-free gesture recognition schemes based on acoustic sensing signals are promising solutions for next-generation human-computer interaction systems. However, existing gesture recognition frameworks reuse visual neural networks to perform feature extraction. These approaches ignore the time sequence nature of the acoustic signal and treat acoustic echo profiles solely as 2D images. In this paper, we propose a time-sequence-based deep learning framework that can exploit the spatio-temporal information of sensing signals. The framework first fuses multi-channel acoustic signals to extract spatial gesture features from a single acoustic frame and then uses the Transformer network to discover the timing relations between spatial features. Our extensive evaluations with real-world datasets show that our light-weighted framework outperforms the state-of-the-art in classifying 14 gestures and achieves an average accuracy of 95.85%.

Traditional deep learning compilers rely on heuristics for subgraph generation, which impose extra constraints on graph optimization, e.g., each subgraph can only contain at most one complex operator. In this paper, we propose AGO, a framework for graph optimization with arbitrary structures to boost the inference performance of deep models by removing such constraints. To create new optimization opportunities for complicated subgraphs, we propose intensive operator fusion, which effectively stitches multiple complex operators together for better performance. Further, we design a graph partitioning scheme that allows an arbitrary structure for each subgraph while guaranteeing the acyclic property among all generated subgraphs. Additionally, to enable efficient performance tuning for complicated subgraphs, we devise a divide-and-conquer tuning mechanism to orchestrate different system components. Through extensive experiments on various neural networks and mobile devices, we show that our system can improve the inference performance by up to 3.3× when compared with state-of-the-art vendor libraries and deep compilers.

This paper is on designing a compact data structure for multi-set membership testing that allows fast set querying. Multi-set membership testing is a fundamental operation for computing systems. Most existing schemes for multi-set membership testing are built upon Bloom filter and fall short in either storage space cost or query speed. To address this issue, we propose Noisy Bloom Filter (NBF), Error Corrected Noisy Bloom Filter (NBF-E), and Data-driven Noisy Bloom Filter (NBF-D) in this paper. We optimize their misclassification and false positive rates by theoretical analysis and present criteria for selection between NBF, NBF-E, and NBF-D. The key novelty of the three schemes is to store set ID information in a compact but noisy way that allows fast recording and querying and use a denoising method for querying. Especially, NBF-E incorporates asymmetric error-correcting coding techniques into NBF, and NBF-D encodes set ID based on their cardinality. To evaluate NBF, NBF-E, and NBF-D in comparison with the prior art, we conducted experiments using real-world network traces. The results show that NBF, NBF-E, and NBF-D significantly advance the state-of-the-art on multi-set membership testing.

In this paper, we propose a Dynamic Speed Warping (DSW) algorithm to enable one-shot learning for device-free acoustic gesture signals performed by different users. The design of DSW is based on the observation that the gesture type is determined by the trajectory of hand components rather than the movement speed. By dynamically scaling the speed distribution and tracking the movement distance along the trajectory, DSW can effectively match gesture signals from different domains with a ten-fold difference in speeds. Our experimental results show that DSW can achieve a recognition accuracy of 97% for gestures performed by unknown users while only using one training sample of each gesture type from four training users.

Continuous domestic respiration monitoring provides vital information for diagnosing assorted diseases. In this article, we introduce RespTracker , the first continuous, multiple-person respiration tracking system in domestic settings using acoustic-based COTS devices. RespTracker uses a multi-stage algorithm to separate and recombine respiration signals from multiple paths so that it can track the respiration rate of multiple moving subjects. And it leverages features from multiple dimensions to separate different users in the same area. Our experimental results show that our two-stage algorithm can distinguish the respiration of at least four subjects and cover a distance of three meters.

In recent years, automatic guided vehicles (AGVs) have developed rapidly and been widely applied in intelligent transportation, cargo assembly, military testing, and other fields. One of the key issues in these applications is path planning. Global path planning results based on known environmental information are used as the ideal path for AGVs combined with local path planning to achieve safe and fast arrival at the destination. The global planning method planning results as the ideal path should meet the requirements of as few turns as possible, short planning time, and continuous path curvature. We propose a global path-planning method based on an improved A * algorithm. And the robustness of the algorithm is verified by simulation experiments in typical multi obstacles and indoor scenarios. To improve the efficiency of pathfinding time, we increase the heuristic information weight of the target location and avoided the invalid cost calculation of the obstacle areas in the dynamic programming process. Then, the optimality of the number of turns in the path is ensured based on the turning node backtracking optimization method. Since the final global path needs to satisfy the AGV kinematic constraints and the curvature continuity condition, we adopt a curve smoothing scheme and select the optimal result that meets the constraints. Simulation results show that the improved algorithm proposed in this paper outperforms the traditional method and can help AGVs improve the efficiency of task execution by efficiently planning a path with low complexity and smoothness. Additionally, this scheme provides a new solution for global path planning of unmanned vehicles.

Existing deep neural network (DNN) accelerator design automation (ADA) methods adopt architecture templates to predetermine parts of design choices and then explore the left design choices beyond templates. These templates can be classified into intra-PU templates and inter-PU templates according to the architecture hierarchy. Since templates limit the flexibility of ADA, designing effective ADA methods without templates has become an important research topic. Although there have appeared some works to enhance the flexibility of ADA by removing intra-PU templates, to the best of our knowledge no existing works have studied ADA methods without inter-PU templates. ADA with predetermined inter-PU templates is typically inefficient in terms of resource utilization, especially for DNNs with complex topology. In this paper, we propose a novel method, called hardware computation graph (HCG), for ADA without inter-PU templates. Experiments show that HCG method can achieve competitive latency while using only 1.4× ~ 5× fewer on-chip memory, compared with existing state-of-the-art ADA methods.

Voice assistant has been widely used for human-computer interaction and automatic meeting minutes. However, for multiple sound sources, the performance of speech recognition in voice assistant decreases dramatically. Therefore, it is crucial to separate multiple voices efficiently for an effective voice assistant application in multi-user scenarios. In this paper, we present a novel voice separation system using a 2D microphone array in multiple sound source scenarios. Specifically, we propose a spatial filtering-based method to iteratively estimate the Angle of Arrival (AoA) of each sound source and separate the voice signals with adaptive beamforming. We use BeamForming-based cross-Correlation (BF-Correlation) to accurately assess the performance of beamforming and automatically optimize the voice separation in the iterative framework. Different from cross-correlation, BF-Correlation further performs cross-correlation among the after-beamforming voice signals processed with each linear microphone array. In this way, the mutual interference from voice signals out of the specified direction can be effectively suppressed or mitigated via the spatial filtering technique. We implement a prototype system and evaluate its performance in real environments. Experimental results show that the average AoA error is 1.4 degree and the average ratio of automatic speech recognition accuracy is 90.2% in the presence of three sound sources.

With the rising of AR/VR technology and miniaturization of mobile devices, gesture recognition is becoming increasingly popular in the research area of human-computer interaction. Some pioneer ultrasound-based gesture recognition systems have been proposed. However, they mostly rely on low-resolution Doppler Effect, with the focus on whole hand motion and fail to deal with minor finger motions. This paper is to present UltraGesture, an ultrasonic finger motion perception and recognition system based on Channel Impulse Response (CIR). CIR measurements can provide with 7 mm resolution, which is sufficient for minor finger motion recognition. UltraGesture encapsulates CIR measurements into image, and builds a Convolutional Neural Network model to classify these images into different categories corresponding to distinct gestures. Furthermore, we use a sliding-window based method to improve accuracy and reduce response latency. UltraGesture can run on the already existed commercial speakers and microphones on most mobile devices without any hardware modification. Our results demonstrate that UltraGesture can achieve an average accuracy ofgreater than 99 percent for 12 gestures including finger click and rotation.

Among all the road accidents, speeding is the most deadly factor. To reduce speeding, it is essential to devise efficient schemes for ubiquitous speed monitoring. Traditional approaches either suffers from using special equipment(e.g., radar speed gun) or special deployment(e.g., position-fixed cameras). In this article, we propose SpeedTalker, a mobile phone-based approach to perform speed detection on automobiles. By leveraging the built-in microphones and camera from the mobile phone, SpeedTalker estimates the automobile speed by passively sensing the acoustic and image signals. We propose an integrated solution to effectively estimate the automobile’s speed based on COTS devices, and provide a platform for every pedestrian to help report the speeding event of automobiles. Specifically, we use the time difference of arrivals (TDOA) model based on acoustic signals to figure out the candidate trajectories of automobile, and use the pin-hole model based on image frames to figure out the vertical distance between the user’s position and the automobile’s trajectory, thus to estimate the unique trajectory. Combined with the time stamp of the trajectory, the automobile speed can be estimated. Besides, we propose a method to effectively mitigate the influence of the movement jitters of mobile phone. We implemented a system prototype for SpeedTalker and estimated the automobile speed with high accuracy. Experiment results show that in the scenario of single automobile, SpeedTalker can achieve an average estimation error of 6.1 percent compared to radar speed guns. In the scenario of multiple automobiles, SpeedTalker can achieve an average estimation error of 9.8 percent, which is acceptable for usage.

Continuous domestic respiration monitoring provides vital information for diagnosing assorted diseases. In this paper, we introduce RESPTRACKER, the first continuous, multiple-person respiration tracking system in domestic settings using acoustic-based COTS devices. RESPTRACKER uses a two-stage algorithm to separate and recombine respiration signals from multiple paths in a short period so that it can track the respiration rate of multiple moving subjects. Our experimental results show that our two-stage algorithm can distinguish the respiration of at least four subjects at a distance of three meters.

Gesture tracking is a basic Human-Computer Interaction mechanism to control devices, such as IoT and VR/AR devices. However, prior OFDM signal based systems focus on gesture recognition and provide results with insufficient accuracy, and thus, cannot be applied for high-precision gesture tracking. In this paper, we propose a CSI based device-free gesture tracking system, called WiTrace, which leverages the CSI values extracted from OFDM signals to enable accurate gesture tracking. For 1D tracking, WiTrace derives the phase of the signals reflected by the hand from the composite signals, and measures the phase changes to obtain the movement distance. For 2D tracking, WiTrace proposes the first CSI based scheme to accurately estimate the initial position, and adopts the Kalman Filter based on continuous Wiener process acceleration model to further filter out tracking noise. Our results show that WiTrace achieves an average accuracy of 6.23 cm for initial position estimation and achieves cm-level accuracy with average tracking errors of 1.46 cm and 2.09 cm for 1D tracking and 2D tracking, respectively.

Cellular network operators deploy base stations with a high density to ensure radio signal coverage for 4G/5G networks. While users enjoy the high-speed connection provided by cellular networks, an adversary could exploit the dense cellular deployment to detect nearby human movements and even recognize keystroke movements of a victim by passively listening to the CRS broadcast from base stations. To demonstrate this, we develop SpiderMon, the first attempt to perform passive continuous keystroke monitoring using the signal transmitted by commercial cellular base stations. Our experimental results show that SpiderMon can detect keystroke movements at a distance of 15 meters and can recover a 6-digits PIN input with a success rate of more than 51% within ten trials when the victim is behind the wall.

In this paper, we propose a Dynamic Speed Warping (DSW) algorithm to enable one-shot learning for device-free gesture signals performed by different users. The design of DSW is based on the observation that the gesture type is determined by the trajectory of hand components rather than the movement speed. By dynamically scaling the speed distribution and tracking the movement distance along the trajectory, DSW can effectively match gesture signals from different domains that have a ten-fold difference in speeds. Our experimental results show that DSW can achieve a recognition accuracy of 97% for gestures performed by unknown users, while only use one training sample of each gesture type from four training users.

Device-free hand gesture is one of the most natural ways to interact with everyday objects. However, existing WiFi-based gesture recognition solutions are typically restricted to indoor environments due to limited outdoor coverage. Furthermore, to achieve high sampling rates, they may interfere with normal data transmissions. In this paper, we aim to develop a robust dynamic gesture interaction system that can be ubiquitously deployed using Long-term Evolution (LTE) mobile terminals. Through both empirical studies and in-depth analysis using the Fresnel zone model, we reveal the key factors that contribute to the repeatability and discernibility of gestures. We show that the optimal location and orientation to perform gestures indeed exist and can be identified without prior knowledge of the position of LTE base stations (BSs) relative to a terminal. Guided by the design principles derived from Fresnel zone characteristics around a 4G terminal, we design highly repeatable and discernible gestures with salient received signal profiles. A gesture interaction system has been developed and implemented to achieve robust recognition with this careful design. Extensive experiments have been conducted in both indoor and outdoor environments, for different relative placements of mobile terminal and BS, and with different users. The proposed system can automatically identify the direction of BSs with a median error of less than 15 degrees and achieve gesture recognition accuracy as high as 98% in all scenarios without the need to acquire any training data.

Logistics monitoring is a fundamental application that utilizes RFID systems to manage numerous tagged-objects. Due to the frequent rearrangement of tagged-objects, a fast RFID-based tracking approach is highly desired for accurate logistics distribution. However, traditional RFID systems usually take tens of seconds to interrogate hundreds of RFID tags, not to mention the time delay involved to locate all the tags, which severely prevents from in-time tracking. To address this issue, we reduce the problem domain by first distinguishing the motion status of the tagged-objects, i.e., “stationary” or “moving”, and then tracking the moving objects with the state-of-the-art localization schemes, which significantly reduces the efforts of tracking all the objects. Toward this end, we propose a moving tag detection mechanism, which achieves the time efficiency by exploiting the useless collision signal in RFID systems. In particular, we extract two kinds of physical-layer features (namely, phase profile and backscatter link frequency) from the collision signal received by the USRP to distinguish tags at different positions. We further develop the Graph Matching (GM) method and Coherent Phase Variance (CPV) method to detect the moving tagged-objects. Experiment results show that our approach can accurately detect the moving objects while reducing 80 percent inventory time compared with the state-of-art solutions.

In memory computing and high-end distributed storage demand low latency, high throughput, and zero data loss simultaneously from datacenter networks. Existing reactive congestion control approaches cannot both minimize queuing latency and ensure zero data loss. A token-oriented proactive approach can achieve them together by controlling congestion even before sending data packets. However, state-of-the-art token-oriented approaches only strive to optimize network-level metrics: maximizing throughput while achieving flow-level fairness. This article answers the question of how to support objective-aware traffic scheduling in token-oriented approaches. The novelty of Token-Oriented in-network Prioritization (TOP) is that it prioritizes tokens instead of data packets. We make three contributions. Via simulations over a hypothetical TOP system, our first contribution is demonstrating the potential performance gain that can be brought by TOP. Second, we investigate the applicability of TOP. Although the overhead of enabling necessary TOP features in switches is trivial, we find that mainstream commodity datacenter switches do not support them. We hence propose a readily-deployable remedy to achieve in-network prioritization by pushing both switch and end-host hardware capacity to an extreme end. Lastly, we implement a running TOP system with Linux hosts and commodity switches, and evaluate TOP in testbeds and with large-scale simulations for various scenarios.

Remote Direct Memory Access(RDMA) technology rapidly changes the landscape of nowadays datacenter applications. Congestion control for RDMA networking is a critical challenge. As an end-to-end layer 3 congestion control mechanism, Datacenter QCN (DCQCN) alleviates the unfairness and head-of-the-line blocking problems of Priority-based Flow Control (PFC). However, a lossless network does not guarantee low latency even with DCQCN enabled. When network congestion happens, switch queues still build-up due to the response latency of end-to-end solutions. In this article, we propose Predictive PFC (P-PFC) to reduce tail latency in RDMA networks. P-PFC monitors the derivative of buffer occupation, predicts the happening of PFC trigger in the future, and proactively triggers PFC pause in advance. The benefit is that buffer usage can be maintained at a low level, hence the tail latency can be controlled. Preliminary evaluation results demonstrate that P-PFC can reduce tail latency by more than half of that in standard PFC in many scenarios, without hurting the throughput and average latency. P-PFC can also protect innocent flows compared with standard PFC according to our experiments. To our best knowledge, this is the first work of using derivative to improve PFC in lossless RDMA networks.

Nowadays, the growing demand for the 3D human-computer interaction (HCI) has brought about a number of novel approaches, which achieve the HCI by tracking the motion of different devices, including the translation and the rotation. In this paper, we propose to use a spinning linearly polarized antenna to track the 3D motion of a specified object attached with the passive RFID tag array. Different from the fixed antenna-based solutions, which suffer from the unavoidable signal interferences at some specific positions/orientations, and only achieve the good performance in some feasible sensing conditions, our spinning antenna-based solution seeks to sufficiently suppress the ambient signal interferences and extracts the most distinctive features, by actively spinning the antenna to create the optimal sensing condition. Moreover, by leveraging the matching/mismatching property of the linearly polarized antenna, i.e., in comparison to the circularly polarized antenna, the phase variation around the matching direction is more stable, and the RSSI variation in the mismatching direction is more distinctive, we are able to find more distinctive features to estimate the position and the orientation. We build a model to investigate the RSSI and the phase variation of the RFID tag along with the spinning of the antenna, and further extend the model from a single RFID tag to an RFID tag array. Furthermore, we design corresponding solutions to extract the distinctive RSSI and phase values from the RF-signal variation. Our solution tracks the translation of the tag array based on the phase features, and the rotation of the tag array based on the RSSI variation. The experimental results show that our system can achieve an average error of 13. 6cm in the translation tracking, and an average error of 8.3° in the rotation tracking in the 3D space.

This paper addresses the fundamental problem of identifying driver talking directions using a single smartphone, which can help drivers by warning distraction of having conversations with passengers in a vehicle and enable safety enhancement. The basic idea of our system is to perform talking status and direction identification using two microphones on a smartphone. We first use the sound recorded by the two microphones to identify whether the driver is talking or not. If yes, we then extract the so-called channel fingerprint from the speech signal and classify it into one of three typical driver talking directions, namely, front, right and back, using a trained model obtained in advance. The key novelty of our scheme is the proposition of channel fingerprint which leverages the heavy multipath effects in the harsh in-vehicle environment and cancels the variability of human voice, both of which combine to invalidate traditional TDoA, DoA and fingerprint based sound source localization approaches. We conducted extensive experiments using two kinds of phones and two vehicles for four phone placements in three representative scenarios, and collected 23 hours voice data from 20 participants. The results show that our system can achieve 95.0% classification accuracy on average.

Voice has been used as biometrics for human authentication because different people have different voice characteristics due to different vocal tract shapes and intonations. However, traditional voice based human authentication is subject to four types of attacks: impersonation, voice conversion, synthesis and voice replay. In this paper, we propose SpeakPrint, an ultrasound based human speech authentication scheme for smartphones which is resistant for these attacks. Compared with traditional speech authentication system which focuses on what a user speaks, SpeakPrint captures how a user speaks by recording mouth and vocal movement through ultrasound signal at the same time. Our key insight is that for the valid user, features extracted from voice signal should be consistent with his mouth and vocal movement recorded from ultrasound signal, while an imitator or an audio player can't produce the same signals in ultrasound domain. SpeakPrint extracts MFCC feature in normal voice frequency and MMSI features from ultrasound signal. An SVM classifier is trained to detect these attacks by comparing above feature differences. We implemented SpeakPrint on Samsung S5 and conducted experiments on 40 users. Experimental results show that SpeakPrint can detect replay attacks with 100% accuracy and replay attack with lip synching for 99.12% for passphrases longer than five words. This technology can be used in multi-factor authentication systems, where multiple authentication mechanisms are used to achieve defense in depth.

Machine slot-time spent on data transmission has direct impact on average job completion time (JCT). In this paper, we propose Macroflow, a networking abstraction that can capture the primitive scheduling granularity of machine slot-time. We demonstrate that minimizing machine slot-time is equivalent to minimizing the average macroflow completion time (MCT). We prove that minimizing MCT to be strongly NP-hard and focus on developing effective heuristics. We propose the Smallest-Macroflow-First (SMF) and Smallest-Average-Macroflow-First (SAMF) heuristics that greedily schedule macroflows based on their network footprint. To work with existing commodity switches, priority discretization is performed to classify macroflows into a small number of priority queues.

Target group distribution pattern analysis has a wide potential application in various domains, i.e., weather forecast based on cloud system distribution, target correlation and tracking based on distribution relationships, forest sustainable management based on tree distribution patterns and so on. However, existing work in target group distribution pattern analysis generally concentrates on either the distribution tendency or the distribution shape of the group while ignore the subtle difference in the density variation in the patterns. To address the above issue, we propose an effective target group distribution pattern discovery method via convolutional neural network (CNN) to discriminate such delicate target group distribution patterns. Firstly, we transform the spatial target group distribution samples into 2D images. Upon that, we design a bagged convolutional neural network (CNN) model. Finally, we apply the bagged CNN model for target group distribution pattern identification. Extensive experiments on synthetic data sets indicate that our method has outperformed the classical machine learning methods significantly1.

Recently, gesture recognition has gained considerable attention in emerging applications (e.g., AR/VR systems) to provide a better user experience for human-computer interaction. Existing solutions usually recognize the gestures based on wearable sensors or specialized signals (e.g., WiFi, acoustic and visible light), but they are either incurring high energy consumption or susceptible to the ambient environment, which prevents them from efficiently sensing the fine-grained finger movements. In this paper, we present RF-finger, a device-free system based on Commercial-Off-The-Shelf (COTS) RFID, which leverages a tag array on a letter-size paper to sense the fine-grained finger movements performed in front of the paper. Particularly, we focus on two kinds of sensing modes: finger tracking recovers the moving trace of finger writings; multi-touch gesture recognition identifies the multi-touch gestures involving multiple fingers. Specifically, we build a theoretical model to extract the fine-grained reflection feature from the raw RF -signal, which describes the finger influence on the tag array in cm- level resolution. For the finger tracking, we leverage K-Nearest Neighbors (KNN) to pinpoint the finger position relying on the fine-grained reflection features, and obtain a smoothed trace via Kalman filter. Additionally, we construct the reflection image of each multi-touch gesture from the reflection features by regarding the multiple fingers as a whole. Finally, we use a Convolutional Neural Network (CNN) to identify the multi-touch gestures based on the images. Extensive experiments validate that RF -finger can achieve as high as 88% and 92% accuracy for finger tracking and multi-touch gesture recognition, respectively.

Enabling touch gesture sensing on all surfaces of the mobile device, not limited to the touchscreen area, leads to new user interaction experiences. In this paper, we propose VSkin, a system that supports fine-grained gesture-sensing on the back of mobile devices based on acoustic signals. VSkin utilizes both the structure-borne sounds, i.e., sounds propagating through the structure of the device, and the air-borne sounds, i.e., sounds propagating through the air, to sense finger tapping and movements. By measuring both the amplitude and the phase of each path of sound signals, VSkin detects tapping events with an accuracy of 99.65% and captures finger movements with an accuracy of 3.59mm.

For AR/VR systems, tapping-in-the-air is a user-friendly solution for interactions. Most prior in-air tapping schemes use customized depth-cameras and therefore have the limitations of low accuracy and high latency. In this paper, we propose a fine-grained depth-aware tapping scheme that can provide high accuracy tapping detection. Our basic idea is to use light-weight ultrasound based sensing, along with one COTS mono-camera, to enable 3D tracking of user's fingers. The mono-camera is used to track user's fingers in the 2D space and ultrasound based sensing is used to get the depth information of user's fingers in the 3D space. Using speakers and microphones that already exist on most AR/VR devices, we emit ultrasound, which is inaudible to humans, and capture the signal reflected by the finger with the microphone. From the phase changes of the ultrasound signal, we accurately measure small finger movements in the depth direction. With fast and light-weight ultrasound signal processing algorithms, our scheme can accurately track finger movements and measure the bending angle of the finger between two video frames. In our experiments on eight users, our scheme achieves a 98.4% finger tapping detection accuracy with FPR of 1.6% and FNR of 1.4%, and a detection latency of 17.69ms, which is 57.7ms less than video-only schemes. The power consumption overhead of our scheme is 48.4% more than video-only schemes.

Activity sensing has become a key technology for many ubiquitous applications, such as exercise monitoring and elder care. Most traditional approaches track the human motions and perform activity recognition based on the waveform matching schemes in the raw data representation level. In regard to the complex activities with relatively large moving range, they usually fail to accurately recognize these activities, due to the inherent variations in human activities. In this paper, we propose a wearable approach for logic cognition-based activity sensing scheme in the logical representation level, by leveraging the meta-activity recognition. Our solution extracts the angle profiles from the raw inertial measurements, to depict the angle variation of limb movement in regard to the consistent body coordinate system. It further extracts the meta-activity profiles to depict the sequence of small-range activity units in the complex activity. By leveraging the least edit distance-based matching scheme, our solution is able to accurately perform the activity sensing. Based on the logic cognition-based activity sensing, our solution achieves lightweight-training recognition, which requires a small quantity of training samples to build the templates, and user-independent recognition, which requires no training from the specific user. The experiment results in real settings shows that our meta-activity recognition achieves an average accuracy of 92% for user-independent activity sensing.

Keystroke privacy is critical for ensuring the security of computer systems and the privacy of human users as what is being typed could be passwords or privacy sensitive information. In this paper, we show for the first time that WiFi signals can also be exploited to recognize keystrokes. The intuition is that while typing a certain key, the hands and fingers of a user move in a unique formation and direction and thus generate a unique pattern in the time-series of channel state information (CSI) values, which we call CSI-waveform for that key. In this paper, we propose a WiFi signal-based keystroke recognition system called WiKey. WiKey consists of two commercial off-the-shelf WiFi devices, a sender (such as a router) and a receiver (such as a laptop). The sender continuously emits signals and the receiver continuously receives signals. When a human subject types on a keyboard, WiKey recognizes the typed keys based on how the CSI values at the WiFi signal receiver end. We implemented the WiKey system using a TP-Link TL-WR1043ND WiFi router and a Lenovo X200 laptop. WiKey achieves over 97.5% detection rate for detecting the keystroke and 96.4% recognition accuracy for classifying single keys. In real-world experiments, WiKey can recognize keystrokes in a continuously typed sentence with an accuracy of 93.5%. WiKey can also recognize complete words inside a sentence with over 85% accuracy.

Since human bodies are good reflectors of wireless signals, human activities can be recognized by monitoring changes in WiFi signals. However, existing WiFi-based human activity recognition systems do not build models that can quantify the correlation between WiFi signal dynamics and human activities. In this paper, we propose a Channel State Information (CSI)-based human Activity Recognition and Monitoring system (CARM). CARM is based on two theoretical models. First, we propose a CSI-speed model that quantifies the relation between CSI dynamics and human movement speeds. Second, we propose a CSI-activity model that quantifies the relation between human movement speeds and human activities. Based on these two models, we implemented the CARM on commercial WiFi devices. Our experimental results show that the CARM achieves recognition accuracy of 96% and is robust to environmental changes.

In this paper, we propose WifiU, which uses commercial WiFi devices to capture fine-grained gait patterns to recognize humans. The intuition is that due to the differences in gaits of different people, the WiFi signal reflected by a walking human generates unique variations in the Channel State Information (CSI) on the WiFi receiver. To profile human movement using CSI, we use signal processing techniques to generate spectrograms from CSI measurements so that the resulting spectrograms are similar to those generated by specifically designed Doppler radars. To extract features from spectrograms that best characterize the walking pattern, we perform autocorrelation on the torso reflection to remove imperfection in spectrograms. We evaluated WifiU on a dataset with 2,800 gait instances collected from 50 human subjects walking in a room with an area of 50 square meters. Experimental results show that WifiU achieves top-1, top-2, and top-3 recognition accuracies of 79.28%, 89.52%, and 93.05%, respectively.

In a number of RFID-based applications such as logistics monitoring, the RFID systems are deployed to monitor a large number of RFID tags. They are usually required to track the movement of all tags in a real-time approach, since the tagged-goods are moved in and out in a rather frequent approach. However, a typical cycle of tag inventory in COTS RFID system usually takes tens of seconds to interrogate hundreds of RFID tags. This hinders the system to track the movement of all tags in time. One critical issue in such type of tag monitoring is to efficiently distinguish the motion status of all tags, i.e., stationary or moving. According to the motion status of different tags, the state-of-art localization schemes can further track those moving tags, instead of tracking all tags. In this paper, we propose a real-time approach to detect the moving tags in the monitoring area, which is a fundamental premise to support tracking the movement of all tags. We achieve the time efficiency by decoding collisions from the physical layer. Instead of using the EPC ID, which cannot be decoded in collision slots, we are able to extract two kinds of physical-layer features of RFID tags, i.e., the phase profile and the backscatter link frequency, to distinguish among different tags in different positions. By resolving the two physical-layer features from the tag collisions, we are able to derive the motion status of multiple tags simultaneously, and greatly improve the time-efficiency. Experiment result shows that our solution can accurately detect the moving tags while reducing 80% of inventory time compared with the state-of-art solutions.

Device-free gesture tracking is an enabling HCI mechanism for small wearable devices because fingers are too big to control the GUI elements on such small screens, and it is also an important HCI mechanism for medium-to-large size mobile devices because it allows users to provide input without blocking screen view. In this paper, we propose LLAP, a device-free gesture tracking scheme that can be deployed on existing mobile devices as software, without any hardware modification. We use speakers and microphones that already exist on most mobile devices to perform device-free tracking of a hand/finger. The key idea is to use acoustic phase to get fine-grained movement direction and movement distance measurements. LLAP first extracts the sound signal reflected by the moving hand/finger after removing the background sound signals that are relatively consistent over time. LLAP then measures the phase changes of the sound signals caused by hand/finger movements and then converts the phase changes into the distance of the movement. We implemented and evaluated LLAP using commercial-off-the-shelf mobile phones. For 1-D hand movement and 2-D drawing in the air, LLAP has a tracking accuracy of 3.5 mm and 4.6 mm, respectively. Using gesture traces tracked by LLAP, we can recognize the characters and short words drawn in the air with an accuracy of 92.3% and 91.2%, respectively.

In this demo, we present LLAP, a hand tracking system that uses ultrasound to localize the hand of the user to enable device-free gesture inputs. LLAP utilizes speakers and microphones on Commercial-Off-The-Shelf (COTS) mobile devices to play and record sound waves that are inaudible to humans. By measuring the phase of the sound signal reflected by the hands or fingers of the user, we can accurately measure the gesture movements. With a single pair of speaker/microphone, LLAP can track hand movement with accuracy of 3.5 mm. For devices with two microphones, LLAP enables drawing-in-the air capability with tracking accuracy of 4.6 mm. Moreover, the latency for LLAP is smaller than 15 ms for both the Android and the iOS platforms so that LLAP can be used for real-time applications.

This paper is on designing a compact data structure for multi-set membership testing allowing fast set querying. Multi-set membership testing is a fundamental operation for computing systems and networking applications. Most existing schemes for multi-set membership testing are built upon Bloom filter, and fall short in either storage space cost or query speed. To address this issue, in this paper we propose Noisy Bloom Filter (NBF) and Error Corrected Noisy Bloom Filter (NBF-E) for multi-set membership testing. For theoretical analysis, we optimize their classification failure rate and false positive rate, and present criteria for selection between NBF and NBF-E. The key novelty of NBF and NBF-E is to store set ID information in a compact but noisy way that allows fast recording and querying, and use denoising method for querying. Especially, NBF-E incorporates asymmetric error-correcting coding technique into NBF to enhance the resilience of query results to noise by revealing and leveraging the asymmetric error nature of query results. To evaluate NBF and NBF-E in comparison with prior art, we conducted experiments using real-world network traces. The results show that NBF and NBF-E significantly advance the state-of-the-art on multi-set membership testing.

Existing work on localization in wireless rechargeable sensor networks (WRSNs) assumes that the charger is equipped with an omnidirectional antenna. We notice that the current available COTS (Commercial off-the-shelf) energy harvesting kit such as the one manufactured by Powercast includes wireless charger with directional antenna. We consider the localization problem in a wireless rechargeable sensor network with a mobile wireless charger. We propose two efficient region dividing methods, Angle Division and Grid Division. Both exploit the field angle of the directional charger to localize individual sensor nodes. The Angle Division is more suitable for general convex polygons. Grid Division algorithm is efficient when the nodes are distributed uniformly and the area is approximate to a rectangle. Further, we extend the two methods by combining the charging time to improve the localization precision. To verify the design, we have extensively evaluated our design by both field experiments and large- scale simulations. The experiment and simulation results show that our algorithm can achieve a localization distance error of less than a quarter meter by as less as four stops.

Heterogeneous cellular networks use small base stations, such as femtocells and WiFi APs, to offload traffic from macrocells. While network operators wish to globally balance the traffic, users may selfishly select the nearest base stations and make some base stations overcrowded. In this paper, we propose to use an auction-based algorithm - Femto-Matching, to achieve both load balancing among base stations and fairness among users. Femto-Matching optimally solves the global proportional fairness problem in polynomial time by transforming it into an equivalent matching problem. Furthermore, it can efficiently utilize the capacity of randomly deployed small cells. Our trace-driven simulations show Femto-Matching can reduce the load of macrocells by more than 30% compared to non-cooperative game based strategies.

In the age of big data, the emitter parameter measurement data is generally characteristic of uncertainty in the form of normally-distributed intervals, enormous size and continuous growth. However, existing interval-valued data analysis methods generally assume a uniform distribution instead and are unable to adapt to the rapid growth of volume. To address the above problems, we have brought forward an incremental distributed weighted class discriminant analysis method on interval-valued emitter parameters. Extensive experiments indicate that our method is able to cope with these new characteristics effectively.

Keystroke privacy is critical for ensuring the security of computer systems and the privacy of human users as what being typed could be passwords or privacy sensitive information. In this paper, we show for the first time that WiFi signals can also be exploited to recognize keystrokes. The intuition is that while typing a certain key, the hands and fingers of a user move in a unique formation and direction and thus generate a unique pattern in the time-series of Channel State Information (CSI) values, which we call CSI-waveform for that key. In this paper, we propose a WiFi signal based keystroke recognition system called WiKey. WiKey consists of two Commercial Off-The-Shelf (COTS) WiFi devices, a sender (such as a router) and a receiver (such as a laptop). The sender continuously emits signals and the receiver continuously receives signals. When a human subject types on a keyboard, WiKey recognizes the typed keys based on how the CSI values at the WiFi signal receiver end. We implemented the WiKey system using a TP-Link TL-WR1043ND WiFi router and a Lenovo X200 laptop. WiKey achieves more than 97.5\% detection rate for detecting the keystroke and 96.4% recognition accuracy for classifying single keys. In real-world experiments, WiKey can recognize keystrokes in a continuously typed sentence with an accuracy of 93.5%.

Some pioneer WiFi signal based human activity recognition systems have been proposed. Their key limitation lies in the lack of a model that can quantitatively correlate CSI dynamics and human activities. In this paper, we propose CARM, a CSI based human Activity Recognition and Monitoring system. CARM has two theoretical underpinnings: a CSI-speed model, which quantifies the correlation between CSI value dynamics and human movement speeds, and a CSI-activity model, which quantifies the correlation between the movement speeds of different human body parts and a specific human activity. By these two models, we quantitatively build the correlation between CSI value dynamics and a specific human activity. CARM uses this correlation as the profiling mechanism and recognizes a given activity by matching it to the best-fit profile. We implemented CARM using commercial WiFi devices and evaluated it in several different environments. Our results show that CARM achieves an average accuracy of greater than 96%.

With the rapid proliferation of RFID-based applications, RFID tags have been deployed into pervasive spaces in increasingly large numbers, e.g., the shelves of super markets are filled with tag-labeled items. Conventional localization schemes usually leverage precise anchor nodes to help compute the position of objects. However, it is usually difficult to find or deploy enough anchor nodes for accurate localization. In this paper, we propose solutions to locate the mobile users based on imprecise anchors in RFID systems. A large number of tags with approximate locations are used as anchor nodes to compute the user's locations. We thus present a time-efficient localization scheme to continuously tracking the mobile users. Experimental results indicate that our solutions can accurately locate the mobile users in a real-time approach. The improved method's accuracy is more than 30% better than the base solution.

With the wide deployment of network facilities and the increasing requirement of network reliability, the disruptive event like natural disaster, power outage or malicious attack has become a non-negligible threat to the current communication network. Such disruptive event can simultaneously destroy all devices in a specific geographical area and affect many network based applications for a long time. Hence, it is essential to build disaster-resilient network for future highly survivable communication services. In this paper, we focus on the integrated approach through the technique of software defined networking to mitigate disaster risks while cut down the investment and management costs. Our design consists of a sub-graph based proactive protection approach for fast rerouting at the network nodes and a splicing approach at the controller for effective post-disaster restoration. Such a systematic design is implemented in OpenFlow framework through the Mininet emulator and Nox controller. Numerical results show that our approach can achieve high reliability, fast recovery and low control overhead.

In wireless sensor networks, the connectivity-based localization protocols are widely studied due to low cost and no requirement for special hardware. Many connectivity-based algorithms rely on distance estimation between nodes according to their hop count, which often yields large errors in anisotropic sensor network. In this paper, we propose a virtual potential field algorithm, in which the estimated positions of unknown nodes are iteratively adjusted by eliminating the inconsistency to the connectivity constraint. Unlike current connectivity-based algorithms, VPF effectively exploits the connectivity constraint information, regardless of distance estimation between nodes, thus achieving high localization accuracy in both isotropic and anisotropic sensor networks. Simulation results show that VPF improves the localization accuracy by an average of 47% compared with MDS in isotropic network, and 42% compared with PDM in anisotropic network. As a refinement procedure, the average improvement factor of VPF is 56% and 50%, based on MDS and PDM respectively.

The incremental classifier is superior in saving significant computational cost by incremental learning on continuously increasing training data. However, existing classification algorithms are problematic when applied for incremental learning for multi-class classification. First, some algorithms, such as neural network and SVM, are not inexpensive for incremental learning due to their complex architectures. When applied for multi-class classification, the computational cost would rise dramatically when the class number increases. Second, existing incremental classification algorithms are usually based on a heuristic scheme and sensitive to the training data input order. In addition, in case the test instance is an outlier and belongs to none of the existing classes, few classification algorithms is able to detect it. Finally, the feature selection and weighing schemes being utilized are generally risky for a "siren pitfall" for multi-class classification tasks. To address the above problems, we bring forward an incremental gray relational analysis algorithm (IGRA). Experimental results showed that, when applied for incremental multi-class classification, IGRA is stable in output, robust to training data input order, superior in computational efficiency, and also capable of detecting outliers and alleviating the "siren pitfall".

Consisting of sensors and vehicles, Underwater Sensor Networks (UWSNs) are deployed to perform collaborative monitoring tasks over a given region, such as water quality monitoring, mining equipment monitoring, oceanographic data collection, pollution surveillance, etc. However, comparing with terrestrial wireless sensor networks, it is more crucial to prolong network lifetime for UWSNs since the varying characteristics of the underwater environment and superior difficulty of underwater device maintenance. In this paper we present a three- dimensional hemisphere model for UWSNs and prove that the improvement in network lifetime by utilizing one Automatic Underwater Vehicle (AUV) is upper bounded by a factor of eight and the AUV only needs to stay within a two hop radius of the sink. We further propose an underwater aggregation routing algorithm UARA to prolong lifetime by utilizing one AUV. Finally, we perform extensive simulations to validate our conclusions.

Spatial multiple access holds the promise to boost the capacity of wireless networks when an access point has multiple antennas. Due to the asynchronous and uncontrolled nature of wireless LANs, conventional MIMO technology does not work efficiently when concurrent transmissions from multiple stations are uncoordinated. In this paper, we present the design and implementation of a crosslayer system, called SAM, that addresses the challenges of enabling spatial multiple access for multiple devices in a random access network like WLAN. SAM uses a chain-decoding technique to reliably recover the channel parameters for each device, and iteratively decode concurrent frames with misaligned symbol timings and frequency offsets. We propose a new MAC protocol, called CCMA, to enable concurrent transmissions by different mobile stations while remaining backward compatible with 802.11. Finally, we implement the PHY and MAC layer of SAM using the Sora high-performance software radio platform. Our evaluation results under real wireless conditions show that SAM can improve network uplink throughput by 70% with two antennas over 802.11.

Half duplex devices are widely used in today's wireless networks. These devices can only send or receive, but not do both at the same time. In this paper, we use cooperative decode-forward relay strategies to increase the throughput of half-duplex wireless networks. Due to the half duplex constraint, relays need to carefully choose their transmission states in order to maximize the throughput. We show that the transmission schedule optimization can be formulated as a linear programming problem. Although the number of possible states grows exponentially as the number of relays increases, only a small subset of these states needs to be used in the optimal transmission schedule. This observation allows us to use heuristic algorithms to solve for near-optimal schedule in large networks. Our numerical results show that the decode-forward strategy can provide nearly 3 times more throughput than the traditional multi-hop relaying strategy in half duplex wireless networks.

Abstract In this paper, we study the problem of scheduling sensor activity to cover a set of targets with known locations such that all targets can be monitored all the time and the network can operate as long as possible. A solution to this scheduling problem is to partition all sensors into some sensor covers such that each cover can monitor all targets and the covers are activated sequentially. In this paper, we propose to provide information coverage instead of the conventional sensing disk coverage for target. The notion of information coverage is based on estimation theory to exploit the collaborative nature of geographically distributed sensors. Due to the use of information coverage, a target that is not within the sensing disk of any single sensor can still be considered to be monitored (information covered) by the cooperation of more than one sensor. This change of the problem settings complicates the solutions compared to that by using a disk coverage model. We first define the target information coverage (TIC) problem and prove its NP‐completeness. We then propose a heuristic to approximately solve our problem. Simulation results show that our heuristic is better than an existing algorithm and is close to the upper bound when only the sensing disk coverage model is used. Furthermore, simulation results also show that the network lifetime can be significantly improved by using the notion of information coverage compared with that by using the conventional definition of sensing disk coverage. Copyright © 2008 John Wiley & Sons, Ltd.

Low latency anonymity systems are susceptive to traffic analysis attacks. In this paper, we propose a dependent link padding scheme to protect anonymity systems from traffic analysis attacks while providing a strict delay bound. The covering traffic generated by our scheme uses the minimum sending rate to provide full anonymity for a given set of flows. The relationship between user anonymity and the minimum covering traffic rate is then studied via analysis and simulation. When user flows are Poisson processes with the same sending rate, the minimum covering traffic rate to provide full anonymity to m users is O(log m). For Pareto traffic, we show that the rate of the covering traffic converges to a constant when the number of flows goes to infinity. Finally, we use real Internet trace files to study the behavior of our algorithm when user flows have different rates.

In centralized wireless networks, reducing the transmission power normally leads to higher network transport throughput. In this paper, we investigate power control in a different scenario, where the network adopts distributed MAC layer coordination mechanisms. We first consider widely adopted RTS/CTS based MAC protocols. We show that an optimal power control protocol should use higher transmission power than the "just enough" power in order to improve spatial utilization. The optimal protocol has a minimal transmission floor area of Theta(dijdmax), where dmaxis the maximal transmission range and dijis the link length. This surprisingly implies that if a long link is broken into several short links, then the sum of the transmission floors reserved by the short links is still comparable to that reserved by the long link. Thus, using short links does not necessarily lead to higher throughput. Another consequence of this is that, with the optimal RTS/CTS based MAC, rate control can at best provide a factor of 2 improvement in transport throughput. We then extend our results to other distributed MAC protocols which uses physical carrier sensing or busy-tone as the control signal. Our simulation results show that the optimal power controlled scheme outperforms other popular MAC layer power control protocols.

This paper considers the coverage problem for hybrid networks which comprise both static and mobile sensors. The mobile sensors in our network only have limited mobility, i.e., they can move only once over a short distance. In random static sensor networks, sensor density should increase as O(log L + k log log L) to provide k-coverage in a network with a size of L. As an alternative, an all-mobile network can provide k-coverage with a constant density of O(k), independent of network size L. We show that the maximum distance for mobile sensors is O( 1/radic(k) log3/4(kL)). We then propose a hybrid network structure, comprising static sensors and a small fraction of O( 1/radic(k)) of mobile sensors. For this network structure, we prove that k-coverage is also achievable with a constant sensor density of O(k). Furthermore, for this hybrid structure, we prove that the maximum distance which any mobile sensor has to move is bounded as O(log(3/4)L). We then propose a distributed relocation algorithm, where each mobile sensor only requires local information in order to optimally relocate itself. We verify our analysis via extensive numerical evaluations and show an implementation of the mobility algorithm on real mobile sensor platforms.

Target tracking and localization are important applications in wireless sensor networks. Although the coverage problem for target detection has been intensively studied, few consider the coverage problem from the perspective of target localization. In this paper, we propose two methods to estimate the lower bound of sensor density to guarantee a bounded localization error over the sensing field. We first convert the coverage problem for localization to a conventional disk coverage problem, where the sensing area is a disk centered at the sensor. Our results show that the disk coverage model requires 4 times more sensors for localization compared to detection applications. We then introduce the idea of sector coverage to tighten the lower bound. The lower bound derived through sector coverage is 2 times less than through disk coverage. A distributed sector coverage algorithm is then proposed in this paper. Compared to disk coverage, sector coverage requires more computations. However, it provides more accurate density estimations than the disk model. Numerical evaluations show that the density bound derived through our sector coverage model is tight.

In this paper we consider the coverage problem for target detection applications in wireless sensor networks. Unlike conventional coverage problems which assume sensing regions are disks around sensors, we define the sensing region according to detection constraints in terms of false alarm probability and missing probability. We show that exploiting cooperation between sensors can extend the overall sensing region while maintain the same constraints on false alarm probability and missing probability. We then propose an energy efficient cooperative detection scheme and study the trade-offs on energy consumption between cooperative and non-cooperative schemes. The cooperative scheme can use half the number of sensors to monitor the whole fleld compared to disk model in networks deployed on grids. We also study the communication overheads incurred by the co-operative scheme, and show that only cooperation between limited number of nearby sensors is profitable in terms of energy consumption. In our simulations on randomly deployed networks, cooperation reduces the number of sensors to cover the area by 30% and nearly doubles the number of disjoint sensor sets where each can fully cover the area. Appropriately trading off energy consumption with coverage extension, our cooperative detection scheme can increase the network lifetime by nearly 70%.

This paper studies the problem of how to maximize the network lifetime while preserving network coverage for an energy constrained wireless sensor network. We consider network coverage from an our recently proposed information coverage model [1] other than the conventional sensing disk model. The lifetime maximization problem is modeled as a nonlinear programming problem and is shown NP-Complete. We then propose a family of greedy algorithms to allocate sensors different roles such that different sensors may consume different amount of energies in different intervals to prolong network lifetime while still guaranteeing application requirements. Simulation results suggest that the algorithm with the best balancing between communication energy consumption and area coverage requirement has the highest network lifetime.

In this paper, we study the coverage problem for hybrid networks which comprise both static and mobile sensors. We consider mobile sensors with limited mobility, i.e., they can move only once over a short distance. Such mobiles are simple and cheap compared to sophisticated mobile robots. In conventional static sensor networks, for a random deployment, the sensor density should increase as O(log L + k log log L) to provide k-coverage in a network with a size of L. As an alternative, an all mobile sensor network can provide k-coverage over the field with a constant density of O(k), independent of network size L. We show that the maximum distance that any mobile sensor will have to move is O(1 over √k log 3 over 4 (kL)). We then propose a hybrid network structure, comprising static sensors and a small fraction of O(1 over √(k)) of mobile sensors. For this network structure, we prove that k-coverage is achievable with a constant sensor density of O(k), independent of network size L. Furthermore, for this hybrid structure, we prove that the maximum distance which any mobile sensor has to move is bounded as O(log3 over 4 L). We then propose a distributed relocation algorithm, where each mobile sensor only requires local information in order to optimally relocate itself and characterize the algorithm's computational complexity and message overhead. Finally, we verify our analysis via extensive numerical evaluations.

In many delay tolerant applications, information is opportunistically exchanged between mobile devices who encounter each other. In order to effect such information exchange, mobile devices must have knowledge of other devices in their vicinity. We consider scenarios in which there is no infrastructure and devices must probe their environment to discover other devices. This can be an extremely energy consuming process and highlights the need for energy conscious contact probing mechanisms. If devices probe very infrequently, they might miss many of their contacts. On the other hand, frequent contact probing might be energy inefficient. In this paper, we investigate the trade-off between the probability of missing a contact and the contact probing frequency. First, via theoretical analysis, we characterize the trade-off between the probability of a missed contact and the contact probing interval for stationary processes. Next, for time varying contact arrival rates, we provide an optimization framework to compute the optimal contact probing interval as a function of the arrival rate. We characterize real world contact patterns via Bluetooth phone contact logging experiments and show that the contact arrival process is self-similar. We design STAR, a contact probing algorithm which adapts to the contact arrival process. Via trace driven simulations on our experimental data, we show that STAR consumes three times less energy when compared to a constant contact probing interval scheme.

Coverage is an important issue in wireless sensor networks. The most commonly used coverage model in the literature defines a point to be covered if its Euclidian distance to at least one sensor is less than a fixed threshold. This is a conservative definition of coverage which implicitly assumes that each sensor makes a decision independent of other sensors in the field. Sensors can cooperate to make an accurate estimation, even if any single sensor is unable to do so. We have previously proposed a new notion of information coverage and investigated its properties. In this paper, we study sensor density requirements for complete information coverage of a field with random sensor deployment. We provide an upper bound on the probability that an arbitrary point in a randomly deployed sensor field is not information covered and find the relationship between the sensor density and the average field vacancy. Simulation results validate our theoretical analysis and show that significant savings in terms of sensor density for complete coverage can be achieved with information coverage.

Target tracking and localization are important applications in wireless sensor networks. Although the coverage problem for target detection has been intensively studied, few consider the coverage problem from the perspective of target localization. In this paper, we propose two methods to estimate the necessary sensor density which can guarantee a localization error bound over the sensing field. In the first method, we convert the coverage problem for localization to a conventional disk coverage problem, where the sensing area is a disk centered around the sensor. Our results show that the disk coverage model requires 4 times more sensors for tracking compared to detection applications. We then introduce the idea of sector coverage, which can satisfy the same coverage conditions with 2 times less sensors over the disk coverage approach. This shows that conventional disk coverage model is insufficient for tracking applications, since it overestimates the sensor density by two times. Simulation results show that the network density requirements derived through sector coverage are close to the actual need for target tracking applications.

An important application of wireless sensor networks is to perform the monitoring missions, for example, to monitor some targets of interests at all times. Sensors are often equipped with non-rechargeable batteries with limited energy and energy saving is a critical aspect for wireless sensor networks. If a target is monitored simultaneously by serval sensors, some of them can be switched off to save energy without causing mission failure and by which their operational times as well as the network lifetime can be prolonged. In this paper, we study the problem of scheduling sensor activity to cover a set of targets with known locations such that all targets can be monitored all the time and the network can operate as long as possible. A solution to this scheduling problem is to partition all sensors into sensor covers such that each cover can monitor all targets and the covers are activated successively. In this paper, we propose to use the notion of information coverage which is based on the estimation theory to exploit the collaborative nature of wireless sensor networks, instead of using the conventional definition of coverage. Due to the use of information coverage, a target that is not within the sensing disk of any single sensor can still be considered to be monitored (information covered) by the cooperation of more than one sensor.

In this paper we investigate the benefits of a heterogeneous architecture for wireless sensor networks composed of a few resource rich mobile nodes and a large number of simple static nodes. These mobile nodes can either act as mobile relays or mobile sinks. To investigate the performance of these two options and the trade-offs associated with these two options, we first consider a finite network. We then compute the lifetime for different routing algorithms for three cases (i) when the network is all static (ii) when there is one mobile sink and (iii) when there is one mobile relay. We find that using the mobile node as a sink results in the maximum improvement in lifetime. We contend however that in hostile terrains, it might not always be possible for the sink to be mobile. We then investigate the performance of a large dense network with one mobile relay and show that the improvement in network lifetime over an all static network is upper bounded by a factor of four. Also, the proof implies that the mobile relay needs to stay only within a two hop radius of the sink. We then construct a joint mobility and routing algorithm which comes close to the upper bound. However this algorithm requires all the nodes in the network to be aware of the location of the mobile node. We then proposed an alternative algorithm, which achieves the same performance, but requires only a limited number of nodes in the network to be aware of the location of the mobile. We finally compare the performance of the mobile relay and mobile sink and show that for a densely deployed sensor field of radius R hops, we require O(R) mobile relays to achieve the same performance as the mobile sink.