The vision of sustainable 6G connectivity infrastructure, from space to ground, critically relies on green communication protocols that can efficiently operate within the constraints of wireless power transfer (WPT). Backscatter communication emerges as a cornerstone for such protocols, yet its potential is hindered by the inability to adapt to the dynamic spectrum and energy conditions inherent to WPT-powered networks. This paper presents ChannelDance, an edge-assisted adaptive hopping system for Bluetooth Low Energy (BLE) backscatter, architected specifically for green and sustainable connectivity in WPT-enabled environments. By leveraging real-time excitation channel intelligence from a low-latency edge server, ChannelDance dynamically configures the tag modulation clock, enabling robust and spectrally agile frequency hopping. This agility is paramount for maintaining reliable communication links amidst the interference and intermittent energy supply characteristic of integrated WPT systems. Our prototype demonstrates a median hopping success rate of 93% across 40 channels, a 3.5× goodput gain with channel optimization, and the ability to establish connections with commodity BLE devices under hopping conditions. ChannelDance thus establishes a foundational green communication primitive for future sustainable 6G networks, where seamless coexistence with energy transfer signals is not a feature but a fundamental requirement.

Information security is a non-negligible issue for wireless transmission. Covert communication provides high security by concealing the transmitted signals within environmental noise. However, existing solutions suffer from low transmission efficiency. Ambient backscatter, concealing data within ubiquitous ambient signals, provides a promising way to achieve high-efficiency covert communication. In this paper, we propose CoScatter, an efficient covert transmission system based on OFDM WiFi backscatter. Current studies rely on redundant modulation, resulting in low throughput. This paper is to increase throughput and shorten transmission time, thereby reducing exposure risk. This is the first work to realize single-sample level demodulation, efficiently eliminating the redundancy, increasing the throughput, and reducing the transmission time. We discover that the main obstacles are the additional phase offsets introduced by three independent wireless channels in backscatter systems. Based on this, we design a new backscatter channel equalization procedure to remove the channel influences while preserving all the covert information embedded by the tag, realizing an efficient covert transmission. Evaluation results show that Coscatter achieves a throughput exceeding 15.7 Mbps, which is around 64x of that of RapidRider, and 16x of that of Tscatter. Consequently, the exposure risk of CoScatter is reduced to 1/64 of that of RapidRider and 1/16 of that of Tscatter.

We present HEScatter, a high-efficiency ambient backscatter system that simultaneously improves carrier, power, and transmission efficiency. To improve carrier efficiency, we choose cellular signal as the carrier due to its continuous transmission nature. Specifically, to ensure low power, we design low-power periodic template matching based on the periodicity of cellular signals to trade time for synchronization accuracy. Further, we calibrate the drift introduced by Sampling Frequency Offset (SFO) to increase carrier utilization. In addition, we exploit Reference Signal (RS)-based demodulation to demodulate tag and ambient data from backscattered signals alone in various traffic patterns for efficient transmission. We prototype HEScatter using off-the-shelf FPGAs and SDRs. Extensive experiments show that HEScatter performs well in carrier utilization, power consumption and data transmission. The carrier utilization rate of HEScatter is as high as 99.97%, which is 3.0x higher than the counterpart of SyncLTE. In end-to-end transmission, the energy efficiency of HEScatter is 1.6x and 19.2x higher than LScatter+ and SyncLTE, while LScatter suffers from transmission failures. We also demonstrate the high transmission efficiency of HEScatter, as its aggregate goodput is 1.5x and 3.8x better than LScatter+ and SyncLTE respectively.

Video object detection stands as a pivotal element within the burgeoning landscape of autonomous driving systems. The exigency to fulfill stringent real-time requisites, while upholding both precision and efficiency in detection, underscores its significance. Although extant methodologies enhance either accuracy or efficiency through the exploitation of spatio-temporal inter-dependencies within the video context, their propensity to conduct detection on discrete frames begets superfluous computations and curbed real-time efficacy. This paper introduces a pioneering approach, called Real-Batch, tailored explicitly to redress this quandary. Real-Batch ingeniously processes batches of video frames uniformly, effectually winnowing out repetitive object detection occurrences. Our methodology is rigorously evaluated on the real-word datasets, scrutinizing four key metrics: accuracy, efficiency, informational value, and adherence to timing constraints. The comprehensive findings substantiate that Real-Batch yields an unparalleled maximal surge in accuracy and efficiency, increasing of 4.2%-13.2% and 24.7%-43.9%, respectively, offering promising advancements for autonomous driving systems.

The demand for ultra-low-power wireless connectivity motivates the study of backscatter communication technology. There are already some related products on the market based on excitation signals from commercial radios. However, their modulation techniques, which are the key to backscatter communications, mainly focus on the phase or frequency domain while amplitude modulation is largely ignored. Most research works either deploy one finely tuned RF impedance port for every needed reflection state or connect a nonlinear device to antenna and tune the reflection amplitude by adjusting its biasing voltage, where the former is too complex and the latter is unstable under variable incident signal power, limiting the backscatter applications. For this reason, we introduce AMscatter to leverage the duty ratio of the frequency-shift wave (FS-wave) to design a novel amplitude modulation. Both theoretical analysis and experimental results show that the reflection amplitude approximates a sinusoidal function of the duty ratio. This method requires only two fixed RF impedances, making AMscatter simple and stable. Moreover, we show how to use AMscatter to design quadrature amplitude modulation (QAM) and pulse shaping to improve backscatter communication performance. Extensive experimental results show that the throughput can be as high as 3.9 Mbps, the supported operational range can reach 20 m with 16-QAM modulation, and the out-of-band interference can be suppressed by 15 dB without negatively affecting the communication performance through our pulse shaping.

Backscatter is promising to deliver near-zero-power communications for billions of ZigBee devices. However, existing backscatter tags face two practical challenges. Firstly, the deployment depends on dedicated excitors or additional receivers, driving up costs. Secondly, the modulator couples phase modulation with multiple high-frequency shifting clocks, boosting tag power. We propose BumbleBee, a pervasive, universal, and efficient ZigBee backscatter design. It reuses uncontrolled Bluetooth and proprietary FSK devices as excitors and demodulates with a single commodity receiver, slashing deployment cost. The key innovation of BumbleBee lies in overwrite modulation, which embeds tag data through dominant phase shifts, effectively overwriting ambient excitation content. The implementation of overwrite modulation depends on a novel Multi-Phase Shift (MPS) modulator that decouples baseband modulation from frequency shifting control, cutting clock requirements. These techniques benefit not only ZigBee but also other protocols that decode via the sign of phase shifts, as shown by a BLE5 backscatter extension. Our prototype uses commodity BLE4/FSK-based SDR excitors, an off-the-shelf FPGA, and commodity ZigBee/BLE5 receivers. Experiments show that BumbleBee achieves 223.2 kbps for ZigBee backscatter and 890.7 kbps for BLE5, 10x better than FreeRider [1], while MPS reduces power by 3.5x.

Throughput is a critical performance metric in backscatter communication systems. Existing approaches either suffer from limited throughput or necessitate modifications to transmitters or receivers, leading to incompatibility with commodity radios. In this paper, we introduce SubScatter, a system that achieves both high throughput and excellent compatibility. It employs a single Complementary Code Keying (CCK)-modulated 802.11b WiFi symbol to transmit eight tag bits by manipulating the phase of the backscattered signal across eight discrete time slots within the symbol, thereby enhancing throughput. To ensure compatibility with commercial-off-the-shelf (COTS) radios, SubScatter exclusively utilizes the physical service data unit (PSDU) for recovering the backscatter modulation that conveys the tag bits. Additionally, SubScatter employs real-time Hamming distance calculations to synchronize the binary envelope from the synchronization circuit with a reference sequence, facilitating the sub-symbol backscatter modulation. In addition, we emphasize SubScatter’s versatility in adapting its modulation scheme to optimize performance across various channel conditions. Extensive experiments conducted with our prototype validate its effectiveness, achieving a throughput approximately 11 times higher than that of leading backscatter systems compatible with COTS radios. Our Hamming-distance-based synchronization method outperforms conventional designs that rely solely on signal power detection, successfully reducing the bit error rate (BER) from over 10% to below 1%. Moreover, SubScatter’s throughput can be flexibly adjusted from 11 Mbps to 1.57 Mbps, while the BER improves from 0.29% to 0.04%.

Improving transmission efficiency is a key objective in OFDM WiFi backscatter systems. A promising direction is sub-symbol-level tag modulation, which embeds more tag data within each OFDM symbol. However, we observe that fine-grained tag modulation is coupled with channel variation, which distorts the cascade structure between the two channels, transmitter-to-tag and tag-to-receiver, making the conventional channel estimation method in WiFi ineffective. Although recent systems have explored new channel estimation methods, their accuracy is limited and the modulation redundancy remains necessary. To address this problem, we present Fascatter, a high-throughput OFDM WiFi backscatter system that enables single-sample-level tag modulation without modulation redundancy. The key enabler is a new channel estimation method that independently estimates the two channels at per-subcarrier granularity. We construct channel observations from the LTF fields and reference symbols, and accurately solve the two channels through matrix decomposition. We further introduce polynomial smoothing and multi-symbol fine-tuning modules to improve estimation robustness. Experimental results demonstrate that the channel estimation results are close to the actual channel responses, and our method shows robust performance under a variety of complex channel conditions. In particular, Fascatter achieves a throughput of up to 15.9 Mbps, which is at least 3.2× over state-of-the-art systems. © 2026 IEEE.

Long-Tailed (LT) recognition has been widely studied to tackle the challenge of imbalanced data distributions in real-world applications. However, the design of neural architectures for LT settings has received limited attention, despite evidence showing that architecture choices can substantially affect performance. This paper aims to bridge the gap between LT challenges and neural network design by providing an in-depth analysis of how various architectures influence LT performance. Specifically, we systematically examine the effects of key network components on LT handling, such as topology, convolutions, and activation functions. Based on these observations, we propose two convolutional operations optimized for improved performance. Recognizing that operation interactions are also crucial to network effectiveness, we apply Neural Architecture Search (NAS) to facilitate efficient exploration. We propose LT-DARTS, a NAS method with a novel search space and search strategy specifically designed for LT data. Experimental results demonstrate that our approach consistently outperforms existing architectures across multiple LT datasets, achieving parameter-efficient, state-of-the-art results when integrated with current LT methods.

The proliferation of healthcare IoT devices and the resulting rich healthcare data sprout new possibilities for intelligent healthcare applications. Patients in intensive care units (ICUs) rely on various networked gadgets to continuously monitor their health and manage critical situations. Among the common therapeutic interventions in ICUs, invasive mechanical ventilation and injecting sedatives during ventilation play crucial roles in maintaining respiratory function and enhancing patient care. While existing therapeutic interventions largely depend on experience and intuition, we propose a federated inverse reinforcement learning framework, termed FERRY, which automatically and intelligently learns optimal therapeutic intervention policies across networked ICUs while keeping raw data local. Specifically, our federated approach overcomes limitations in medical data privacy and facilitates collaboration; our proposed inverse reinforcement learning framework learns the variational posterior distribution from historical trajectories to handle the unknown reward. Additionally, we enhance our framework with distributionally robust optimization to ensure worst-case performance and adaptively filter out noisy data through joint loss learning. Extensive experiments on the real-world dataset demonstrate that FERRY improves the overall ventilation and sedation decision-making accuracy by 36.75% compared to other state-of-the-art baselines.

Multi-task learning (MTL) has emerged as a promising approach for deploying deep learning models in real-life applications. Recent studies have proposed optimization-based learning paradigms to establish task-shared representations in MTL. However, our paper empirically argues that these studies, specifically gradient-based ones, primarily emphasize the conflict issue while neglecting the potentially more significant impact of imbalance/dominance in MTL. In line with this perspective, we enhance the existing baseline method by injecting imbalance-sensitivity through the imposition of constraints on the projected norms. To demonstrate the effectiveness of our proposed IMbalance-sensitive Gradient (IMGrad) descent method, we evaluate it on multiple mainstream MTL benchmarks, encompassing supervised learning tasks as well as reinforcement learning. The experimental results consistently demonstrate competitive performance.

The modeling of complicated time-evolving physical dynamics from partial observations is a long-standing challenge. Particularly, observations can be sparsely distributed in a seemingly random or unstructured manner, making it difficult to capture highly nonlinear features in a variety of scientific and engineering problems. However, existing data-driven approaches are often constrained by fixed spatial and temporal discretization. While some researchers attempt to achieve spatio-temporal continuity by designing novel strategies, they either overly rely on traditional numerical methods or fail to truly overcome the limitations imposed by discretization. To address these, we propose CoPS, a purely data-driven methods, to effectively model continuous physics simulation from partial observations. Specifically, we employ multiplicative filter network to fuse and encode spatial information with the corresponding observations. Then we customize geometric grids and use message-passing mechanism to map features from original spatial domain to the customized grids. Subsequently, CoPS models continuous-time dynamics by designing multi-scale graph ODEs, while introducing a Markov-based neural auto-correction module to assist and constrain the continuous extrapolations. Comprehensive experiments demonstrate that CoPS advances the state-of-the-art methods in space-time continuous modeling across various scenarios. The source code is available at~\url{https://github.com/Sunxkissed/CoPS}.

Motivated by the pursuit of greater efficiency, companies, such as Amazon and JD, are shifting toward a warehouse-distribution integration model to optimize logistics operations. In general full-link logistics scenarios, the collaboration between warehouses and sorting centers managed by different enterprises leads to data silos, posing challenges in securely sharing information and accurately predicting delivery times across the entire logistics network. Current delivery time prediction methods often overlook the heterogeneity of logistics networks and face data sharing constraints. We aim to address these issues by facilitating secure internode relationship analysis and leveraging distinct spatio-temporal characteristics to enhance efficiency. However, challenges remain in overcoming data isolation while maintaining protection and integrating diverse node characteristics for optimized modeling. To address these challenges, we propose the federated heterogeneous graph transformer (Fed-HGT) framework. This framework includes a federated training module that integrates local and central gradients by exchanging node representations and model parameters between logistics nodes and the central server. Additionally, it features a federated prediction module where local nodes compute time representations using their local data and transmit these to the central server. The central server then uses these representations to make accurate full-link delivery time predictions. Our method was evaluated on a dataset from a major e-commerce platform in China, demonstrating significant performance improvements over existing solutions.

Most Personalized Federated Learning (PFL) algorithms merge the model parameters of each client with other (similar or generic) model parameters to optimize the personalized model (PM). However, the merged model parameters in these algorithms may fit low relevance data, thereby limiting the performance of PM. In this paper, we generate similar data for each client through the collaboration of a generic model (GM) on the server, rather than merging model parameters. To train a generator capable of generating data for all classes on the server without real data, we employ the GM as the discriminator in adversarial training with the generator. Additionally, we introduce a similarity assessment metric, which allows for the assessment of the similarity between local data and data from other classes. Nevertheless, the presence of non-IID data among clients can weaken the performance of the GM, consequently impacting the training of the generator and similarity assessment. To address this issue, we design a directive mechanism so that GM can be optimized during adversarial training without the need for additional training. The experimental results validate the superiority of our algorithm over state-of-the-art algorithms in terms of accuracy, loss, and convergence speed.

WiFi-based gesture recognition has emerged as a promising alternative to computer vision, enabling seamless integration and enhanced interaction in human-computer interaction systems. Simultaneously identifying users during gesture recognition is vital for improving security and personalization. However, existing WiFi-based dual-task recognition approaches often rely on handcrafted features, which hinder precision and introduce delays in cross-domain scenarios. To address these challenges, we propose WiDual, a real-time system for cross-domain gesture recognition and user identification using WiFi signals. By integrating spatial and channel attention mechanisms, WiDual adaptively extracts crucial features for dual-task recognition. The system employs Channel State Information (CSI) visualization to convert WiFi signals into images, facilitating efficient feature extraction and minimizing information loss and latency. Furthermore, a collaborative module fuses gesture and user identity features, enhancing recognition performance. Experimental evaluations on a public dataset with six gestures and six users across diverse environments demonstrate WiDual's effectiveness. It achieves 96% accuracy in cross-domain gesture recognition and 91.27% in user identification. Compared to state-of-the-art methods, WiDual improves user identification accuracy by 26%, gesture recognition by 8%, and reduces processing time sixfold, showcasing its potential for real-time applications.

Domain generalization remains a key challenge in human activity recognition based on channel state information (CSI). Different domains correspond to distinct data distributions, deviating from the typical assumption of independent and identically distributed (i.i.d.) data, which leads to significant performance degradation when models are applied to unseen domains. To address this issue, we propose a novel domain generalization model that integrates meta-learning initialization and an adaptive channel grouping attention mechanism. First, a meta-learning strategy is employed to acquire well-initialized parameters from multiple source domain tasks, enabling the model to implicitly enhance its cross-domain generalization ability. Second, an adaptive grouping attention mechanism is designed in the feature extraction stage to effectively capture the sensitivity differences of different subcarriers to human activities. Meanwhile, a random masking training mechanism is introduced to simulate real-world domain variations and improve model robustness. In addition, a domain adversarial training framework based on the gradient reversal layer (GRL) is adopted to mitigate domain-specific feature dependency, further enhancing the model's generalization capability. We evaluate our proposed method on both a self-collected dataset, which includes human activity data from nine volunteers across six different environments, and a public CSI dataset. The experimental results demonstrate that our method significantly outperforms existing approaches in domain generalization performance, verifying its effectiveness and practical applicability.

Ambient backscatter is one of the most promising solutions for the widespread deployment of low-power IoT. We present CAB, a content-agnostic backscatter system that can demodulate both tag and ambient data from ambient backscattered WiFi alone. In contrast to prior ambient backscatter systems that use ambient data (content) to demodulate tag data, we focus on zero-subcarriers, which are invariant and independent for any ambient OFDM WiFi. The idea of using zero-subcarriers to convey tag data is simple and elegant. Not only does it for the first time remove the dependency of tag-data demodulation on ambient data, but it also significantly improves the practicality of ambient backscatter. CAB is also universal for various OFDM-WiFi signals because the zero-subcarrier we use is virtual and is essentially the phase estimation of the pilot, which is not affected by different pilot patterns. In order to verify the universality, we prototype CAB using off-the-shelf FPGAs and SDRs. Extensive experiments show CAB is universal as it can work with multi-band, multi-stream, and multi-user ambient traffic, including WiFi 3/4/5/6. In addition, CAB can work with excitations at different data rates over various commercial NICs. As the first content-agnostic backscatter system with 340.9 Mbps aggregate throughput, we believe CAB takes a crucial step forward on ubiquitous battery-free IoTs.

We present EchScatter, a novel backscatter system that takes productive ZigBee signals as excitations and enables high-throughput ZigBee backscatter communication. Compared with the existing ZigBee backscatter, EchScatter does not need to control the carrier, ensuring the universality of the transmitter. EchScatter has realized chip-level modulation of productive ambient ZigBee backscatter for the first time. It is capable of transmitting more tag bits simultaneously, thus increasing the throughput of the system. We first design 16 different 32-chip phase modulation sequences to realize the translation of any two ZigBee symbols. Therefore, our system can transmit four tag bits through one symbol. After that, we use an average energy detection-based synchronization method to make sure the synchronization error of the EchScatter meets the requirements of chip-level modulation. We prototyped EchScatter using an off-the-shelf FPGA and commodity ZigBee transceivers. Through extensive experiments and field studies, we show that EchScatter can work universally on commodity ZigBee transceivers. In line-of-sight and non-line-of-sight scenarios, EchScatter is able to achieve 247 kbps and 245 kbps throughput at a signal strength of around -60 dBm, respectively. The throughput of EchScatter is 32 times higher than that of FreeRider. We believe it will have more pervasive applications.

WiFi-based human activity recognition (WiFi-based HAR) has emerged as a technology in recent decades, offering convenient and privacy-friendly applications. However, existing frameworks designed for stable environments encounter challenges when faced with changing circumstances and imbalanced training datasets in realistic scenarios. In this article, we address both issues from a unified perspective by exploring a more generalized local minima. Initially, we revisit existing solutions and empirically observe the presence of sharp minima in trained long-tailed WiFi-based HAR models. Consequently, we propose a novel method called Class Region Flattening ( CRF ) to identify class-conditional flat minima. This approach effectively mitigates bias caused by the long-tailed distribution and enhances generalization capabilities in the face of changing circumstances. Furthermore, we introduce a selective flattening operation to prevent optimization conflicts among different activity categories and reduce computational overhead. We integrate CRF into mainstream WiFi-based HAR models and evaluate their performance using our collected WiFi-based HAR dataset. Through extensive experiments, we demonstrate that the incorporation of CRF leads to significant improvements in performance. These findings underscore the effectiveness of CRF in addressing the challenges posed by changing circumstances and imbalanced training datasets in WiFi-based HAR.

Backscatter communication has attracted significant attention due to its low power consumption and energy efficiency. Enabling concurrent backscatter allows multiple tags to operate simultaneously, and their data can be recovered from collided signals. This capability is crucial for enhancing management efficiency in smart logistics and mitigating multi-tag collisions in Internet-of-Things (IoT) scenarios where multiple tags work collaboratively. However, existing concurrent backscatter schemes are vulnerable to noise and asynchronous signals, causing limited performance. To address these challenges, we introduce Ortho-CodeA, a backscatter scheme that enables reliable concurrent backscatter communication despite high noise levels and asynchronous signals. The underlying concept is to take advantage of coding mechanisms to combat noise and employ time-shifted orthogonal codes to mitigate the effects of asynchronous signals. Specifically, we design a set of time-shifted orthogonal codes that maintain code orthogonality despite asynchronous signals. Built upon the designed codes, we develop a multi-tag decoding scheme to recover data from each tag. We theoretically analyze the feasibility of our scheme and validate its performance through extensive experimental simulations. The results demonstrate that Ortho-CodeA achieves a BER of about 0.0036% in the case of 7 tags with an SNR of 10 dB and a maximum time delay of$1 \,\mu \text{s}$.

When attacked by mainstream adversarial attack methods, the classical WiFi-based human activity recognition (HAR) system shows poor robustness. Nevertheless, popular standard adversarial training (SAT) conducts adversarial attacks on every data batch by selecting partial perturbation values, resulting in unfavorable model performance and robustness. To mitigate the issues, we propose cross-adversarial training (CAT), an adversarial training technique that cross-attacks the source domain data in each batch. We develop a filter to establish the perturbation threshold for the parameter (i.e., ϵ) of adversarial examples that are distinguishable to naked eyes. Further, we design cross-zero, which crosses the zero point of ϵ and randomly selects the value of ϵ from positive and negative intervals of the threshold range to generate adversarial examples. Our experiments demonstrate that when utilizing the fast gradient sign method (FGSM) and projected gradient descent (PGD), CAT outperforms SAT by an average of 10.37% and 16.33% in accuracy, while SAT results in a decrease of 6.37% and 12.53% compared to the original model (ORI), respectively. To evaluate robustness, we also use a series of ϵ to attack ORI, SAT, and CAT. Results show that the overall robustness of CAT prevails. We believe CAT has a bright future in WiFi-based HAR applications since it increases the system’s security and efficiency through mitigating prevalent adversarial attacks.

OFDM WiFi backscatter with uncontrolled ambient signals is a promising approach for realizing passive Internet of Things (IoT) systems. However, existing OFDM backscatter systems are often limited by coarse modulation granularity, typically constrained to the packet or OFDM symbol levels, and commonly require dual receivers for data demodulation. To address these challenges, we propose DFTScatter, a novel sub-symbol level backscatter system that utilizes a single receiver for demodulation. DFTScatter leverages the Discrete Fourier Transform (DFT) shift theorem to achieve sub-symbol level modulation through frequency domain cyclic shifts. This method enhances data transmission efficiency and operates within a single-symbol bandwidth, thereby optimizing spectrum utilization. Additionally, we introduce a single-receiver decoding technique that exploits invariant frequency domain information of the reflected signals for accurate demodulation. Extensive experiments demonstrate that DFTScatter outperforms existing methods and effectively operates with various ambient OFDM WiFi signals, including WiFi 3/4/5/6, paving the way for scalable, low-cost IoT ecosystems in smart cities, healthcare, industrial automation, and beyond.

The emergence of e-commerce with logistics service provides great convenience to people’s lives. However, platform usually receive seller claim for some reasons (e.g., damaged packages). Thus, it is important to predict seller claim risk in logistics scenarios. Existing solution for seller claim risk predict are challenging to address this problem due to two unique features including (i) multi-side collaborative risk factor caused by data sharing constraints of e-commerce and logistics platform, (ii) industry-specific risk factor caused by the dynamic similarity of sellers. To incorporate these new factors, we propose a novel seller claim risk prediction framework (Fed-SCRP) via federated multi-view learning, where we (i) design a federated multi-view learning to deal with data isolation problem, (ii) develop a STG-SRIM model and a series of information union transformers to capture the hybrid semantic embedding of dynamic seller features and industry-specific risk factor. We conduct a comprehensive evaluation of our method using datasets from major Chinese e-commerce and logistics platforms. Experimental results demonstrate that our method outperforms state-of-the-art baselines in various metrics.

Deep long-tailed recognition (DTLR) has attracted much attention due to its close touch with realistic scenarios. Recent advances have focused on re-balancing across various aspects, e.g., sampling strategy, loss re-weighting, logit adjustment, and input/parameter perturbation, to name a few. However, few studies have considered dynamic re-balancing to address intrinsic optimization conflicts. In this paper, we first empirically argue that the optimizations of mainstream DLTR methods are still dominated by some categories (e.g., major) due to a fixed re-balancing strategy. Thus, they fail to deal with gradient conflicts among categories, which naturally deduces the motivation for reaching Pareto optimal solutions. Unfortunately, a naive integration of multi-objective optimization (MOO) with DLTR methods is not applicable due to the gap between multi-task learning (MTL) and DLTR, and can in turn lead to class-specific feature degradation. Thus, we provide effective alternatives by decoupling MOO-based MTL from the temporal rather than structure perspective, and enhancing it via optimizing variability collapse loss motivated by the derived MOO-based DLTR generalization bound. Moreover, we resort to anticipating worst-case optimization with theoretical insights to further ensure convergence. We build a Pareto deep long-tailed recognition method termed PLOT upon the proposed MOO framework. Extensive evaluations demonstrate that our method not only generally improves mainstream pipelines, but also achieves an augmented version to realize state-of-the-art performance across multiple benchmarks.

Two-way backscatter is essential to general-purpose backscatter communication as it provides rich interaction to support diverse applications on commercial devices. However, existing Bluetooth backscatter systems suffer from unstable uplinks due to poor carrier-identification capability and inefficient downlinks caused by packet-length modulation. This paper proposes EffBlue, an efficient two-way backscatter design for commercial Bluetooth devices. EffBlue employs a simple edge backscatter server that alleviates the computational burden on the tag and helps build efficient uplinks and downlinks. Specifically, efficient uplinks are designed by introducing an accurate synchronization scheme, which can effectively eliminate the use of non-compliant packets as carriers. To break the limitation of packet-level modulation, we design a new symbollevel WiFi-ASK downlink where the edge sends ASK-like WiFi signals and the tag can decode such signals using a simple envelope detector. We prototype the edge server using commodity WiFi and Bluetooth chips and build two-way backscatter tags with FPGAs. Experimental results show that EffBlue can identify the target excitations with more than 99% precision. Meanwhile, its WiFi-ASK downlink can achieve up to 124 kbps, which is 25x better than FreeRider.

Ambient LTE backscatter is a promising way to enable ubiquitous wireless communication with ultra-low power and cost. However, modulation in previous LTE backscatter systems relies heavily on the original data (content) of the signals. They either demodulate tag data using an additional receiver to provide the content of the excitation or modulate on a few predefined reference signals in random ambient LTE traffic.This paper presents CABLTE, a content-agnostic backscatter system that efficiently utilizes uncontrolled LTE PHY resources for backscatter communication using a single receiver. Our system is superior to prior work in two aspects: 1) Using one receiver to obtain tag data makes CABLTE more practical in real-world applications, and 2) Efficient modulation on LTE PYH resources improves the data rate of backscatter communication. To obtain the tag data without knowing the ambient content, we design a checksum-based codeword translation method. We also propose a customized channel estimation scheme and a signal identification component in the backscatter system to ensure our accurate modulation and demodulation. Extensive experiments show that our CABLTE provides maximum tag throughput of 22 kbps, which is 3.67x higher than the content-agnostic system CAB and even 1.38x higher than the content-based system SyncLTE.

Logistics advertising is an emerging task in online-to-offline logistics systems, where logistics companies expand parcel shipping services to new users through advertisements on shopping websites. Compared to existing online e-commerce advertising, logistics advertising has two significant new characteristics: (i) the complex factors in logistics advertising considering both users' offline logistics preference and online purchasing profiles; and (ii) data sparsity and mutual relations among multiple steps due to longer advertising conversion processes. To address these challenges, we design MCAC, a Multi-task Conditional Attention network-based logistics advertising Conversion prediction framework, which consists of (i) an offline shipping preference extraction model to extract the user's offline logistics preference from historical shipping records, and (ii) a multi-task conditional attention-based conversion rate prediction module to model mutual relations among multiple steps in logistics advertising conversion processes. We evaluate and deploy MCAC on one of the largest e-commerce platforms in China for logistics advertising. Extensive offline experiments show that our method outperforms state-of-the-art baselines in various metrics. Moreover, the conversion rate prediction results of large-scale online A/B testing show that MCAC achieves a 15.22% improvement compared to existing industrial practices, which demonstrates the effectiveness of the proposed framework.

In many scenarios, we need to identify an object and then locate it within high precision (centimeter or millimeter level). RFIDs have played a significant role in this field. While many state-of-the-art systems have shown good performance, they require expensive hardware or extra time. Based on a previous work, GLAC, we present PoM, a 3D localization system within millimeter-level precision using only COTS RFID devices. Inspired by the same idea, PoM also draws power from mobility, and makes two key technical improvements. First, to the best of our knowledge, PoM is the first localization system that simultaneously adopts Synthetic Aperture Radar (SAR) and Inverse Synthetic Aperture Radar (ISAR) method. In particular, we employ antenna motion to construct SAR and tag's mobility to construct ISAR. Second, we take actual application scenarios into consideration and apply an extra mechanism so that PoM can gain better performance in special situations. Our simulation experiments show that, in high-speed scenarios and other challenging real-world applications, PoM achieves better performance than the original GLAC system.

Existing IoT systems suffer from restricted communication distances, high deployment costs, and frequent battery replacements, making them ineffective for managing healthcare data. This paper presents Prometheus, a self-powered wristband for reporting personal health status over long distances and intelligently managing healthcare data. Prometheus backscatters ambient BLE and ZigBee signals for low-power communication while incorporating a multi-source energy harvester to convert ambient RF, light, and heat into electricity. It also features a biochemical sensor array for monitoring sweat biochemical markers. Prototyped on a flexible PCB, Prometheus demonstrates impressive efficiency, consuming only 5.8 mW for sweat sensing, with BLE and ZigBee transmission energies significantly lower than standard electrochemical workstations and commercial alternatives. Our experiments show consistent signal quality at distances up to 20 meters. In summary, Prometheus emerges as a convenient, efficient, and self-powered wristband, promising to provide ubiquitous healthcare data management in our lives.

Utilizing communication signals for indoor human activity recognition (HAS) is an important component of integrated sensing and communication (ISAC). The current majority HAS solutions adopt a single sensing strategy and only work in a simple environment. In this paper, we propose a new HAS method named WiSMLF that can flexibly select multiple sensing strategies and then use multi-level feature fusion for sensing. We first use the high frequency energy (HFE) method to categorize human activities into two types: static activities (SAs) and moving activities (MAs). Subsequently, for SAs, we adopt a joint localization and activity recognition sensing strategy, and use a multi-level feature fusion network based on visual geometry group (VGG). For MAs, we adopt a joint activity recognition and moving distance estimation sensing strategy, and use a multi-level feature fusion network based on long short-term memory (LSTM). The experimental results show that WiSMLF outperforms the existing methods especially in complex environments, and can obtain 92% higher accuracy in location, activity recognition, and distance estimation.

AI-aided Financial Regulation (AIFR) is a practical and significant task, but current solutions have yet to be optimized with customized model designs. Given the privacy concerns surrounding financial data, we aim to employ Neural Architecture Search (NAS) to help non-expert end-users automatically design architectures. The genetic algorithm-based NAS stands out due to its relatively low hardware requirements and robust theoretical foundation. However, constrained by limited data, the model would undergo architecture search on a general regulatory dataset while being deployed on private one owned by each organization. The data distribution of the private dataset may vary from that of public datasets, giving rise to the challenge of data domain shift. To alleviate this problem, we propose a novel fitness evaluation method. When scoring the fitness, we take into account both the architecture’s validation accuracy and its potential for generalization by the metric of loss landscape. In addition, we improve the training paradigm for evaluation, utilizing a prototype-based training paradigm based on embedding distances for classification, allowing for rapid domain adaptation and improve performance on the distribution-shift data. We further introduce GA-TextCNN, a GA-based NAS framework specifically designed for text recognition, enhancing its suitability for text data within AIFR tasks. To demonstrate the effectiveness of our approach, we collect two related datasets and evaluate our method on it. The extensive experiments demonstrate that our method significantly improves baseline models and is effective in solving the AIFR problem.

Different from intermittent ISM signals like Bluetooth and WiFi, LTE signals are continuous in time and more pervasive in space, which makes them suitable carriers for ambient backscatter systems. However, due to the continuous LTE traffic and complex frame structures, existing ambient backscatter systems, such as HitchHike and LScatter, cannot reliably backscatter LTE signals in a standard-compatible way. We observe that the primary cause of their failures is that tags cannot accurately synchronize with LTE excitations. To address this issue, we propose SyncLTE, an LTE backscatter system that achieves high-accuracy synchronization and standard-compatible backscatter communication. The key novelty is a new tag design that uses the periodicity of LTE signals for synchronization and provides a customized single-symbol modulation scheme for LTE carriers. Our design is prototyped using FPGAs, SDR LTE eNBs and UEs. Comprehensive experiments have been done in various scenarios, including LoS, NLoS, indoor and outdoor. Results show that SyncLTE is 22.4x and 7.4x better than LScatter and Multiscatter in terms of the 80th percentiles of synchronization errors. Also, SyncLTE can deliver throughputs of up to 200 bps using BPSK and 400 bps using QPSK while other systems suffer from failures.

In the past decade, backscatter communications have drawn significant attention as they are an ultra-low-power solution to transmit IoT sensor data, including video and audio. However, most of the state-of-the-art backscatter systems that are fully compatible with commodity radios suffer from poor synchronization accuracy and low throughput, being unable to support various multimedia sensors. In this paper, we propose Bitalign, a Bluetooth backscatter system that can make use of uncontrolled ambient signals as excitations and deliver high throughput for multimedia streaming applications. To do so, we identify several backscatter bottlenecks and employ a set of techniques to considerably boost backscatter throughput. In particular, we introduce an identification-based synchronization method that can efficiently distinguish various ambient signals and accurately decide where to modulate. We further propose a matching-based synchronization method with higher synchronization accuracy. In addition, we propose header reconstruction to make Bitalign truly compatible with commercial Bluetooth devices. Finally, we implement a tag prototype using FPGAs and conduct extensive experiments. Results show that the minimum bit error rate of Bitalign is 0.5%, which is 60 times better than that of FreeRider, a state-of-the-art Bluetooth system that features uncontrolled excitors. The maximal theoretical throughput of Bitalign is 1.98 Mbps.

Bidirectional links are widely adopted in many active radios to provide efficient data exchanges, e.g., WiFi and Bluetooth. However, they are underexplored for general-purpose backscatter communications. The downlinks of state-of-the-art Bluetooth backscatter systems are based on packet length modulation, which is inefficient and unreliable. In this paper, we propose BiBlue, a bidirectional Bluetooth backscatter system that uses an edge server to enable fast and reliable downlinks. Specifically, our edge server acts like a bridge that can translate commercial Bluetooth signals to Amplitude Shift Keying (ASK) signals. In addition, we design a reliable edge-to-tag link that adaptively decodes ASK signal under variance. Finally, we propose a low-power uplink design that enables connection-based bidirectional communication with commodity devices. We prototype BiBlue using FPGA and Software Defined Radio (SDR). Extensive experimental results demonstrate that BiBlue achieves more than 62x downlink throughput gains over FreeRider and meanwhile supports an 80 cm communication range of edge-to-tag with BER at around 1%.

Recently, the continuous evolution of deep learning has opened up promising avenues to groundbreaking advancements in wireless sensing systems, which significantly enhance the practical applications of WiFi-based Human Activity Recognition (HAR) systems. However, despite these strides, such systems remain susceptible to adversarial attacks. This article unveils the vulnerability of existing WiFi-based HAR systems to common adversaries, revealing their insufficient robustness. While the intuitive approach is to employ adversarial training to fortify the models, our investigation exposes inherent deficiencies in the current approach. Specifically, we confirm that the strength of perturbations directly influences training outcomes. Moreover, even when confined within a specified perturbation radius, the perturbation strength exhibits variability within a prescribed range, potentially giving rise to “extreme” samples that could compromise training results. To address this challenge, we propose a two-stage Selective Adversarial Training (SAT) approach that integrates model confidence calibration and sample selection. Specifically, we start with calibrating the model and then selectively choose samples from all adversarial examples based on the calibrated confidence outputs that align with the desired criteria for adversarial training. This sample-wise perturbation intensity control effectively prevents the inclusion of inappropriate samples in training, a capability lacking in previous domain-wise perturbation control. Our experiments demonstrate that the proposed fine-grained training method, SAT, is both straightforward and effective in augmenting adversarial training results.

The use of controlled excitation makes pervasive backscatter communication difficult to achieve and the redundant modulation severely limits the performance of the system. We present a novel WiFi backscatter system that can take uncontrolled OFDM WiFi signals as excitations and efficiently embed tag data at the single-symbol rate. Specifically, we are the first to discover the fundamental reason why the previous systems have to rely on multi-symbol modulation, which makes it possible to demodulate tag data on the single-symbol level. Further, we design deinterleaving-twins decoding that can reuse any uncontrolled WiFi signals as carriers to backscatter tag data. Moreover, we present how to robustly handle high-order excitations, including different demapping rules for diverse excitations and three different bit-translation methods for decoding. To verify the effectiveness of our proposal, we prototype our solution using various FPGAs and SDRs. Comprehensive evaluations show that our solution’s maximum throughput is 3.92x and 1.97x better than FreeRider and MOXcatter. In addition, with 16QAM excitations, the decoding BERs of majority voting are around 5%, which is 10x better than subsequence matching and jaccard similarity methods. Meanwhile, the throughput of deinterleaving-level demodulation is 2x better than payload-level demodulation with 16QAM ambient traffic.

This paper presents GLAC, the first 3D localization system that enables millimeter-level object manipulation for robotics using only COTS RFID devices. The key insight of GLAC is that mobility reduces ambiguity (One-to-many mapping relationship between phase and distance) and thus improves accuracy. Unlike state-of-the-art systems that require extra time or hardware to boost performance, it draws the power of modeling mobility in a delicate way. In particular, we build a novel framework for real-time tracking using the Hidden Markov Model (HMM). In our framework, multiple Kalman filters are designed to take a single phase observation for updating mobility states, and a fast inference algorithm is proposed to efficiently process an exponentially large number of candidate trajectories. We prototype GLAC with only UHF tags and a commercial reader of four antennas. Comprehensive experiments show that the median position accuracies of x/y/z dimensions are within 1 cm for both LoS and NLoS cases. The median position accuracy for slow-moving targets is 0.41 cm, which is 2.2$\times$, 17.3$\times$, and 14.9$\times$better than TurboTrack, Tagoram, and RF-IDraw, respectively. Also, its median velocity accuracy is at least 20$\times$better than all three competitors for fast-moving targets. Besides accuracy, it achieves more than 4$\times$localization time gains over state-of-the-art systems.

Since backscatter communication has the advantage of low power, it is promising to be widely used in IoT applications. For a space-time stream backscatter, we envision it can efficiently leverage ready-to-use multiform space-time streams in the environment. Previous works have struggled to achieve this because they use various tag-data modulations on various types of ambient space-time stream signals. What’s worse is that it is challenging to identify the properties of a space-time stream in a passive way. To this end, we present STScatter, a WiFi backscatter that is universally applicable for ambient space-time streams. Besides, STScatter achieves efficient tag-data modulation at the symbol-level, allowing space-time stream backscatter to reach high tag-data throughput with multiform excitation. Additionally, unlike previous systems that decoded tag data from both the original ambient and backscattered signal, STScatter can decode ambient data and backscattered data simultaneously from the backscattered signal. STScatter was prototyped using commercial FPGAs and SDRs. Our extensive experiments show that the STScatter is transmitter-agnostic and universal with different ambient space-time streams. STScatter can achieve a goodput of up to 455.9 kbps in single-stream excitation and 460.0 kbps in multi-stream excitation, which is 4.2x and 157.5x better than MOXcatter. Besides, STScatter successfully uses signals from various types of WiFi hardware for excitation and achieves an average tag-data goodput of 293.62 kbps with real-life file-downloading traffic generated by a household WiFi router.

WiFi backscatter has attracted intensive attention because the large population of WiFi radios can provide plenty of excitation signals. However, WiFi backscatter communication has imposed unwanted constraints on either exciters or receivers since its inception. In this paper, we present Chameleon, a native WiFi backscatter system, where WiFi tags can generate native WiFi packets using uncontrolled productive WiFi signals as carriers. Our tag-only design requires no particular excitation patterns and no changes in software/hardware on WiFi network interface cards (NICs). The key idea is for the Chameleon tag to demodulate the productive WiFi signal and backscatter it into a full-function packet using on-the-fly modulation. To align tag decoding and modulation with excitation symbols, we design a time synchronization and clock compensation scheme suitable for low-power tags. We prototype WiFi tags using ultra-low-power FPGAs and evaluate them in real-world scenarios where excitations are ambient traffic and backscatter receivers are a wide range of commercial off-the-shelf (COTS) NICs. Comprehensive field studies show that the maximal backscatter throughput of Chameleon is almost 1 Mbps, which is over$125\times $and$1000\times $higher than what WiTAG and FS-Backscatter tags could achieve, respectively. We also show that Chameleon can natively communicate with various COTS WiFi devices on Windows, iOS, and Android platforms. We believe that this design will enable ubiquitous WiFi connectivity for billions of IoT devices via widely available mobile gadgets and existing wireless infrastructure.

Recently, using deep learning to achieve WiFi-based human activity recognition (HAR) has drawn significant attention. While capable of achieving accurate identification in a single domain (i.e., training and testing in the same consistent WiFi environment), it would become extremely tough when WiFi environments change significantly. As such, domain adversarial neural networks-based approaches have been proposed to handle such diversities across domains, yet often found to share the same limitation in practice: the imbalance between high-capacity of feature extractors and data insufficiency of source domains. This article proposes i-Sample, an intermediate sample generation-based framework, striving to tackle this issue for WiFi-based HAR. i-Sample is mainly designed as two-stage training, where four data augmentation operations are proposed to train a coarse domain-invariant feature extractor in the first stage. In the second stage, we leverage the gradients of classification error to generate intermediate samples to refine the classifiers together with original samples, making i-Sample also capable to be integrated into most domain adversarial adaptation methods without neural network modification. We have implemented a prototype system to evaluate i-Sample, which shows that i-Sample can effectively augment the performance of nowadays mainstream domain adversarial adaptation models for WiFi-based HAR, especially when source domain data is insufficient.

Backscatter is an emerging ultra-low-power wireless communication technology for Internet-of-Things. However, as the widely used modulation scheme in OFDM systems, the combination of QAM and WiFi backscatter is not very satisfactory. To deploy QAM in the WiFi backscatter system, we propose PilotScatter, the first OFDM backscatter system supporting both 16-QAM modulation and non-redundant coding. These changes significantly improve the throughput of PilotScatter over previous backscatter systems. The key insight of PilotScatter is the use of pilot tones and differential demodulation. By modifying the phase and amplitude of pilot tones of the carrier signal, PilotScatter can modulate tag data on ambient WiFi with a 16-QAM scheme. At the receiver, PilotScatter uses a differential algorithm to demodulate tag data. It makes PilotScatter achieve 16-QAM demodulation without relying on ambient WiFi data and past symbols. We prototype PilotScatter using FPGAs, commodity radios, and USRPs. Comprehensive evaluations demonstrate that PilotScatter achieves up to 932.22 kbps throughput for WiFi 802.11g, and the backscatter communication range (Tag-to-Rx) is up to 19 m in Line-of-Sight (LoS) and 16 m in Non-Line-of-Sight (NLoS). Compared with the symbol-level backscatter research, PilotScatter has 7.49x and 3.78x goodput gains over MOXcatter and RapidRider, respectively.

Despite significant progress in recent years, few-shot learning (FSL) still faces two critical challenges. Firstly, most FSL solutions in the training phase rely on exploiting auxiliary tasks, while target tasks are underutilized. Secondly, current benchmarks sample numerous target tasks, each with only an N-way C-shot shot query set in the evaluation phase, which is not representative of real-world scenarios. To address these issues, we propose Guidepost, a target-oriented FSL method that can implicitly learn task similarities using a task-level learn-to-learn mechanism and then re-weight auxiliary tasks. Additionally, we introduce a new FSL benchmark that satisfies realistic needs and aligns with our target-oriented approach. Mainstream FSL methods struggle under this new experimental setting. Extensive experiments demonstrate that Guidepost outperforms two classical few-shot learners, i.e., MAML and ProtoNet, and one state-of-the-art few-shot learner, i.e., RENet, on several FSL image datasets. Furthermore, we implement Guidepost as a domain adaptor to achieve high accuracy wireless sensing on our collected WiFi-based human activity recognition dataset.

With the rapid advancement of smartphones and other mobile devices, an ever-increasing desire for wireless indoor localization has emerged. This technology is capable of determining the position of a user or device in an indoor setting and facilitating an array of captivating applications. Due to the low cost and wide availability of WiFi, WiFi-based indoor localization has received considerable attention and has become a prominent research focus in recent times. We have distilled that an ideal WiFi-based indoor localization system is anticipated to meet three criteria: high-accuracy, pervasiveness, and easy-deployment. Nevertheless, it is not a trivial task to satisfy all three criteria simultaneously. This document scrutinizes the key issues, basic models, and current methods for WiFi indoor localization with the objective of highlighting the underlying principles and challenges. Finally, this manuscript pinpoints the prospective research paths for WiFi indoor localization.

It's widely acknowledged that deep learning models with flatter minima in its loss landscape tend to generalize better. However, such property is under-explored in deep long-tailed recognition (DLTR), a practical problem where the model is required to generalize equally well across all classes when trained on highly imbalanced label distribution. In this paper, through empirical observations, we argue that sharp minima are in fact prevalent in deep long-tailed models, whereas naï ve integration of existing flattening operations into long-tailed learning algorithms brings little improvement. Instead, we propose an effective two-stage sharpness-aware optimization approach based on the decoupling paradigm in DLTR. In the first stage, both the feature extractor and classifier are trained under parameter perturbations at a class-conditioned scale, which is theoretically motivated by the characteristic radius of flat minima under the PAC-Bayesian framework. In the second stage, we generate adversarial features with class-balanced sampling to further robustify the classifier with the backbone frozen. Extensive experiments on multiple long-tailed visual recognition benchmarks show that, our proposed Class-Conditional Sharpness-Aware Minimization (CC-SAM), achieves competitive performance compared to the state-of-the-arts. Code is available at https://github.com/zzpustc/CC-SAM.

The boom of the Internet of Things has revolutionized people's lives, but it has also resulted in massive resource consumption and environmental pollution. Recently, Green IoT (GIoT) has become a worldwide consensus to address this issue. In this paper, we propose EEWScatter, an energy-efficient WiFi backscatter communication system to pursue the goal of GIoT. Unlike previous backscatter systems that solely focus on tags, our approach offers a comprehensive system-wide view on energy conservation. Specifically, we reuse ambient signals as carriers and utilize an ultra-low-power and battery-free design for tag nodes by backscatter. Further, we design a new CRC-based algorithm that enables the demodulation of both ambient and tag data by only a single receiver while using ambient carriers. Such a design eliminates system reliance on redundant transceivers with high power consumption. Results demonstrate that EEWScatter achieves the lowest overall system power consumption and saves at least half of the energy. What's more, the power consumption of our tag is only 1/1000 of that of active radio.

Ambient backscatter is one of the most promising wireless communication techniques that can accommodate the ever-growing scale of IoT devices and connections. The state-of-the-art ambient backscatter systems, however, have to employ two commodity receivers for tag-data demodulation, resulting in additional software and hardware overhead. Using off-the-shelf ambient WiFi signals and a single AP to achieve backscatter communication has seldom been studied before. In this paper, we introduce Twoferscatter, an innovative ambient backscatter system that uses a single commodity AP to decode tag data alone. We propose a nomination-verification decoding strategy to achieve tag-data decoding with the backscatter data. We also develope a block-level modulation method that extends our system to massive MIMO excitation. Our experiments demonstrate that Twoferscatter can successfully demodulate tag data with a single AP. It can deliver up to 5.62 kbps tag-data goodput and work at tag-Rx distances of up to 12$m$using an unmodified NIC receiver. We believe Twoferscatter has a bright future in passive IoT applications since it is the first ambient backscatter that uses ambient WiFi as excitation and requires only a single AP for demodulation.

Codeword translation is well-known for translating excitation signals into other codewords to provide productive ZigBee backscatter. But the limited throughput of conventional systems using info-rich codewords to transmit a single bit challenge their effectiveness to perform sensor-data transmission. In this paper, we introduce ChipScatter, a novel high-throughput modulation technology that translates excitation codewords into more controllable codewords through fine-grained chip-level modulation. The more controllable categories available, the more bits can be transmitted simultaneously. Evaluation results show that ChipScatter can increase the throughput of productive ZigBee backscatter by up to 8×.

This paper presents TiScatter, a timespan-based WiFi backscatter system that provides high-throughput communication with a single COTS receiver used. It outperforms the prior works that tradeoff between considerable data rate and practical deployment. To improve the data rate, TiScatter introduces a symbol-level times-pan modulation method that encodes tag data into the timespan between two modulated WiFi codewords in two successive WiFi packets. For decoding, TiScatter for the first time employs the injective feature between the checksum and the modulated codeword positions, which enables the demodulation of both the tag and original WiFi data using only one COTS receiver. This makes TiScatter more practical. Furthermore, we design TiScatter+ that shows these advantages while providing an even higher throughput under 802.11b excitations. We prototype our design, and comprehensive evaluations demonstrate that TiScatter shows a throughput over 100× higher than prior single-receiver backscatter systems like FS-Backscatter. It even has a better BER and throughput than the prior double-receiver backscatter systems like MOXcatter. Specifically, TiScatter provides 1) 2× higher peak throughput than MOXcatter and 2) an order of magnitude lower BER than MOXcatter with the presence of substantial interferences. In addition, TiScatter+ can deliver a throughput 3× higher than TiScatter under 802.11b ambient excitations. Our evaluation also confirms that TiScatter is generic and applicable to excitations under diverse WiFi standards (e.g., 802.11b/g/n).

Generally, image acquisition and storage requires a considerable amount of energy. To address this issue, we present an innovative approach towards developing a battery-free WiFi camera. Our system uses FRAM and DMA to cache images quickly, select appropriate color modes, and make some optimizations to the code, allowing for the acquisition and storage of a grayscale image of 144 × 176 pixels using only 6mJ of energy in 96ms. Our proposed system provides a promising step towards the realization of a battery-free WiFi camera, enabling energy-efficient image acquisition and storage in applications such as environmental monitoring, surveillance, and IoT.

WiFi backscatter communication has required unwanted constraints on either excitations or receivers since its inception eight years ago. We present Chameleon, the first native WiFi backscatter system where WiFi tags can generate native WiFi packets using uncontrolled productive WiFi as carriers. Our tag-only solution requires no particular excitation patterns and no change for software/hardware on WiFi NICs. The key insight is that the Chameleon tag can demodulate productive WiFi and backscatter this arbitrary carrier into a full-function packet using on-the-fly modulation. We prototype WiFi tags using ultra-low-power FPGAs and evaluate them in real-world scenarios where excitations are ambient traffic and backscatter receivers are a range of COTS NICs. Comprehensive field studies show that the maximal backscatter throughput of Chameleon is almost 1 Mbps, which is over 125× and 1000× better than WiTAG and FS-Backscatter. Also, we show that Chameleon can natively communicate with various COTS WiFi devices on Windows, iOS, and Android platforms. We believe this native WiFi backscatter design will enable ubiquitous WiFi connectivity for billions of IoT devices via widely available mobile gadgets and existing wireless infrastructure.

The research on backscatter multiple access schemes has been an interesting topic. Existing schemes in the area of backscatter multiple access are not designed for specific applications, and leave out the contemplation of practical application requirements. In this paper, therefore, we present Ortho-CodeA, a backscatter multiple access scheme that targets indoor Internet of Things (IoT) applications (e.g., the smart home) and considers the practical application requirements. To implement Ortho-CodeA, we address several key challenges including imperfect time synchronization, high hardware complexity at receiver, and synchronizing multiple tags information in a low-power manner. We theoretically analyze the feasibility of our scheme and evaluate the performance of Ortho-CodeA through extensive simulations. The results show that Ortho-CodeA supports concurrent transmissions of up to 7 tags and achieves BER of 0 when SNR is greater or equal to 0 dB.

We present BumbleBee, a novel backscatter system that creates ZigBee transmissions over productive Bluetooth Low Energy (BLE) carriers. In contrast to prior content-aware or non-productive backscatter, BumbleBee overwrites tag information independently on any ambient BLE. The backscattered signal is dominated by the tag information and compliant with commodity ZigBee radios. Since BLE signals are widespread, BumbleBee enables the vision of pervasive ZigBee backscatter communication. We prototype BumbleBee using commodity BLE transmitters, an off-the-shelf FPGA, and commodity ZigBee receivers. Through extensive experiments and field studies, we show that BumbleBee works universally with ambient BLE and commodity receivers. Further, when the signal strength is -80 dBm, BumbleBee has a throughput of 218 kbps and 204 kbps in the line-of-sight (LOS) and nonline-of-sight (NLOS) scenarios, respectively. The throughput improvement is up to 3x compared with the advanced non-productive backscatter system, Interscatter [1], and$32\mathrm{x}$over the content-aware backscatter system, FreeRider [2]. The bit error ratio (BER) is below 1% when the tag-to-receiver distance is 20 meters. As the first ambient ZigBee backscatter system that works universally with commodity transceivers, we believe BumbleBee takes a crucial step towards pervasive ZigBee backscatter communication.

Traditional energy harvesters only work in a single frequency band and cannot make full use of the RF energy in the environment. This paper proposes an optimized multi-band RF energy harvesting system, which can efficiently work in different frequency bands. The structure of this harvester consists of a wideband receiving antenna linking multiple narrowband rectifier branches. Each branch is composed of a bandpass filter, an impedance matching network, and a multi-stage Dickson charge pump circuit. The output power of each branch is summarized by a diode summation network. On the basis of the existing multi-band RF energy harvester, we optimize the impedance matching network, the rectifier, and the diode summation network, which further improves the system performance. The results show that when the input power is 3 dBm, the power conversion efficiency of this system under the simulation conditions at 1 GHz reaches the maximum value of 78.3%. In addition, we also verify the possibility of this multi-band RF energy harvester to provide energy for backscatter communication in the 2.4 GHz ISM band. The simulation results show that the system can reach a power conversion efficiency of more than 50% in the three selected bands.

Over the past years, Human Activity Recognition (HAR) has shown its great value and has been further developed with the help of deep learning. However, existing HAR systems that use deep learning methods to achieve the ideal accuracy of recognition heavily rely on massive amounts of labeled training samples. Unfortunately, it requires considerable human effort and is unrealistic for real-life applications. In this paper, we propose a novel system, which combines active learning with WiFi-based HAR. The system is capable of building a good activities recognizer in HAR with a limited amount of labeled training samples. We thus call the system ALSensing. To the best of our knowledge, ALSensing is the first system to apply active learning to WiFi-based HAR. We implement ALSensing using commercial WiFi devices and evaluated it with realistic data in several different environments. Our experimental results show that ALSensing achieves 52.83% recognition accuracy using 3.7% training samples, 58.97% recognition accuracy using 15% training samples and the baseline predicted with the existing method achieves 62.19% recognition accuracy using 100% training samples. When the performance of ALSensing is similar to that of the baseline, the required labeled samples are much less than that of the baseline.

Clinical decision-making models have been developed to support therapeutic interventions based on medical data from either a single hospital or multiple hospitals. However, models based on multihospital data require collaboration among hospitals to integrate local data, which can result in information leakage and violate patient privacy. To address this challenge, we propose a novel approach that combines federated learning (FL) with inverse reinforcement learning (IRL) to create an efficient medical decision-making support tool while preserving patient privacy. Our approach uses an IRL algorithm with differential privacy to train a neural network-based agent on local data containing clinician trajectories, which learns a private treatment policy by observing patients’ conditions. Additionally, we integrate FL into the proposed algorithm to learn a global optimal action policy collaboratively among various smart intensive care units, overcoming data limitations at each hospital. We evaluate our approach using real-world medical data and demonstrate that it achieves superior performance in a distributed manner.

Over the past decade, backscatter nodes have received booming interest for many emerging mobile applications, such as sports analytics and interactive gaming. However, backscatter networks are not ready to provide a high-throughput and stable communication platform for billions of such mobile nodes due to two main factors in rate adaptation. First, the common mapping paradigm that chooses the optimal rate based on RSSIs is hardly adaptable to hardware diversity. Second, the current probing processes are not optimized for mobile scenarios due to inefficient probing trigger, inaccurate channel estimation, and unique self-interference. To address those issues, we propose MobiRate, a mobility-aware rate adaptation link-layer that fully exploits the mobility hints from PHY information to deliver a high-throughput link-layer for mobile backscatter networks. The key insight is that mobility-hints can greatly benefit link-layer design, including rate selection and channel probing. Specifically, we introduce a novel velocity-based loss-rate estimation module, a mobility-assisted probing trigger, a selective probing module, and a robust self-interference detection module, significantly saving probing time and improving probing accuracy. As MobiRate is fully compatible with the current standard, we prototype it using COTS RFID readers and commercial tags. Our extensive experiments demonstrate that MobiRate can successfully identify self-interference with detection accuracy over 90% for tags of different velocities. Moreover, it achieves up to 3.8x throughput gain over the state-of-the-art methods across a wide range of mobility, channel conditions, and tag types.

We present multiscatter, a novel battery-free backscatter design that can simultaneously work with multiple excitation signals for personal IoT sensors. Specifically, we show for the first time that the backscatter tag can identify various excitation signals in an ultra-low-power way, including WiFi, Bluetooth, and ZigBee. Further, we employ a new modulation approach, overlay modulation, that can leverage those excitation signals to convey tag data on top of productive data, which makes decoding both data possible with only a single personal radio. Moreover, we introduce a low-power listening scheme to improve energy efficiency. Since 2.4 GHz signals and personal radios are everywhere, multiscatter is readily deployable in our everyday IoT applications. We prototype multiscatter using an FPGA and various commodity radios. Extensive experiments show that for mixed 802.11b&n, Bluetooth and ZigBee signals, the average identification accuracy of four protocols is more than 93%. The maximal aggregate throughput of both productive and tag data is 278.4 kbps with a single Bluetooth radio. When the transmitter-to-tag distance is increased from 0.2 to 1.8 m, the maximal communication for BLE drops from 71 m to 29 m. And it can leverage excitation diversity to provide uninterrupted communication and greater throughput gains, whereas the single-protocol tag being idle when carrier signals are unavailable. With indoor office light as harvesting sources, the low-power listening scheme can support backscatter rate at 12 pkts/s.

The continuous and ubiquitous nature of LTE traffic makes it advantageous as a backscatter carrier. However, existing backscatter works ignore the importance of being LTE-like, resulting in incompatibility with the standard and inability to access LTE services. We observe that achieving LTE-like requires a service initiation mechanism, that is, paging and accurate downlink synchronization. To this end, we present LTElike, a novel backscatter design that can page tags to initiate LTE service requests. Specifically, we page our tag using the form of amplitude modulation, enabling the ultra-low-power tag to receive paging and wake up. Further, we design an accurate synchronization strategy combining detection and template matching based on correlation, which improves the accuracy to single symbol level. We prototype LTElike using FPGAs, SDR LTE eNBs and UEs. The evaluation results show that the paging identification rate is as high as 100%, and the mean synchronization error of LTElike is 17x better than the counterpart of LScatter.

Recently, backscatter technology has attracted much interest in wireless communication due to its novel low-cost and battery-free design. Since Bluetooth Low Energy (BLE) was born to be low energy consumption, great efforts have been made in BLE-based backscatter systems, like FreeRider, RBLE. In this paper, we present BonusBlue, a BLE backscatter system that enables high-goodput communication with commercial BLE devices. BonusBlue tag generates BLE packets by modulating data on excitation signals and set up a data connection with BLE receiver using a state machine, thus achieving a high-goodput communication link. We present this state machine design and build a prototype of our tag using an FPGA, and evaluate its performance with BLE devices. Our evaluation shows that the backscatter tag can build a robust data connection link with commodity BLE device in the guidance of our state machine and transmit tag data on this link. Experimental results show that our backscatter system can achieve a goodput of up to 16.9 kbps.

To piggyback information, the instantaneous-phase shift (IPS) modulation toggles discrete phases on ambient RF carriers, which is popular with advanced backscatter systems. However, IPS has poor spectrum efficiency. It produces serious spectrum sidelobes and prevents the formation of large networks. In this paper, we propose frequency-phase shift (FPS) modulation, a fine-grained RF switches toggling that modulates carriers with a continuous phase shift. The phase continuity suppresses spectrum sidelobes without disturbing the demodulation results. We first apply FPS to optimize a ZigBee backscatter tag. ZigBee signals, consisting of a non-single-tone header and a single-tone payload, are transmitted as RF carriers. The backscatter tag leverages FPS to modulate the single-tone for phase-continuity data transmission. Further, the tag recycles the non-single-tone header using sub-symbol codeword translation to improve carrier utilization. Through extensive experiments and field studies, we demonstrate that FPS enables ZigBee transmissions with a bandwidth of 2.4 MHz, which is 3x lower than that of Interscatter [1] and much closer to active radios. The system prototype consists of a microchip transmitter, a backscatter tag, and a commodity receiver. Specifically, when the transmitter-to-tag distance is within 5 centimeters, the system enables a goodput of 16.6 kbps at a channel capacity of 16.8 kbps, and the communication distance can be extended to 17 meters.

Throughput is one of the key performance indicators in backscatter communication systems. Existing works either have limited throughput or modify the transmitter or receiver to fit the backscatter tag, thereby causing incompatibility with commodity radios. In this paper, we present SubScatter, which realizes a high throughput and keeps compatibility at the same time. For high throughput, SubScatter uses one CCK-modulated 802.11b WiFi symbol to carry eight tag bits by manipulating the phase of the backscattered signal in eight-time slots of a symbol separately. For compatibility with commercial-off-the-shelf (COTS) radios, SubScatter leverages only the physical service data unit (PSDU) to recover the backscatter modulation in which the tag bits are conveyed. And still, to fit the sub-symbol backscatter modulation, SubScatter calculates the Hamming distance between the binary envelope provided by the synchronization circuit and a reference sequence in real-time for synchronization. Extensive experiments in our prototype have proven the effectiveness of SubScatter. SubScatter achieves a throughput of about 11× over state-of-the-art backscatter systems compatible with COTS radios. The Hamming-distance-based synchronization outperforms the design of merely detecting the change of signal power and helps reduce the bit error rate from over 10% to below 1%.

Incorporating domain adaptation is a promising solution to mitigate the domain shift problem of WiFi-based human activity recognition (HAR). The state-of-the-art solutions, however, do not fully exploit all the data, only focusing either on unlabeled samples or labeled samples in the target WiFi environment. Moreover, they largely fail to carefully consider the discrepancy between the source and target WiFi environments, making the adaptation of models to the target environment with few samples become much less effective. To cope with those issues, we propose a Target-Oriented Semi-Supervised (TOSS) domain adaptation method for WiFi-based HAR that can effectively leverage both labeled and unlabeled target samples. We further design a dynamic pseudo label strategy and an uncertainty-based selection method to learn the knowledge from both source and target environments. We implement TOSS with a typical meta learning model and conduct extensive evaluations. The results show that TOSS greatly outperforms state-of-the-art methods under comprehensive 1 on 1 and multi-source one-shot domain adaptation experiments across multiple real-world scenarios.

We propose EAScatter, the first excitor-aware Bluetooth backscatter system with stable performance for different Bluetooth excitors. We first point out that the backscatter tag should fit the guard interval for different Bluetooth excitors. EAScatter uses a connection-based identification method, which can identify excitor during Bluetooth connection according to different shortest high level lengths. Thus, it can select a specific optimal guard interval for each excitor. Moreover, we introduce Plus-one modulation, which can further improve the goodput. We built a prototype of EAScatter and evaluated it with extensive experiments. EAScatter can achieve up to 98% identification accuracy. Using TI CC1352 as an excitor, it has a 25x PRR gain and a 28x goodput gain over the state-of-the-art RBLE.

We present CAB, a content-agnostic backscatter system that can demodulate both tag and ambient data from ambient backscattered WiFi alone. In contrast to prior ambient backscatter systems that use ambient data (content) as tag-data carriers, we focus on zero-subcarriers, which are invariant and independent for any ambient OFDM WiFi. The idea of using zero-subcarriers to convey tag data is simple and elegant. Not only does it for the first time remove the dependency of tag-data demodulation on ambient data, but it also significantly improves the practicality of ambient backscatter.

Existing backscatter systems using commodity radios all have a limited throughput, which prevents their usage in high-rate scenarios. We present SubScatter, which realizes a high throughput and keeps commodity-radio compatibility at the same time. Specifically, SubScatter boosts throughput for 11× by using 802.11b WiFi as excitation (11/8×) and designing sub-symbol modulation (8×).

Backscatter communication is the core technology supporting the passive internet of things (IoT). Existing space-time stream backscatter systems are non-universal and inefficient due to aspects like tag-data modulation being heavily related to the ambient excitation’s stream number, low throughput with multi-stream excitation, and the need for the original signal for tag-data demodulation.We introduce STScatter, a universal space-time stream backscatter. STScatter includes the following unique features that set it apart from other systems: For starters, STScatter’s tag-data modulation is universal for excitation with any number of space-time streams. Furthermore, STScatter achieves efficient tag-data modulation at the symbol-level, allowing space-time stream backscatter to reach high tag-data throughput. Additionally, unlike previous systems that decoded tag data from both the original ambient and backscattered signal, STScatter can decode ambient data and backscatter data simultaneously from the backscattered signal.STScatter is prototyped using commercial FPGAs and SDRs. Our extensive experiments show that the STScatter is universal with different space-time streams excitation, including WiFi 4/5. STScatter can achieve a throughput of up to 499.95 kbps in single-stream excitation and 499.94 kbps in multi-stream excitation, which is 4.10x and 171.10x better than MOXcatter.

The Internet of Things advancements has enabled smart city scenarios worldwide. A tight nit communication, on the one hand, is required for device management and monitoring, but on the other hand, it has raised risks in information exchange in cross-domain scenarios. Keeping in view the above issues, this article considered well-known cross-domain access control protocols, i.e., Shibboleth, xDAuth, and OAuth to ensure user’s security. And, the existing work focused specifically on their implementations without verifying if they are effective for the security scenarios. We aim to verify their claim of fulfilling the claimed security requirements or not. We did a Cyber Attack Analysis to highlight the possible security attacks in smart city scenarios. Then, the protocols are verified against those issues to critically analyze if they are the best fit for the said scenario. The Z3-solver and Satisfiability Modulo Theories Library have been in use for protocol verification purposes. Results reveal that the protocol models are functioning correctly in terms of confidentiality, availability, and integrity. Verifying these protocols in terms of other security properties is part of our future work.

Human activity recognition (HAR) has attracted significant attention during recent years due to its critical role in a wide range of applications. Among existing recognition algorithms, most of them utilize domain adversarial neural networks, such as DANN [6], to achieve recognition between diverse domains. However, these methods try to fully align the feature distributions while each domain has specific characteristics, which leads to different decision boundaries and substantially degrades the recognition accuracy. In this paper, we propose a fine-grained method called Wi-Adaptor to tackle these problems. Wi-Adaptor utilizes two classifiers to match distributions of source and target samples by considering the decision boundaries. In order to detect target samples that are far from the support of the source, we train the classifiers to maximize the discrepancy between their outputs and train the feature generator to generate target features that minimize the discrepancy. Our experiments show that Wi-Adaptor outperforms other traditional domain adversarial adaptation models and show robustness as we limit the source samples. Especially in the case of reducing the source samples to a half, Wi-Adaptor achieves more than 30% accuracy gain in different domain adaptation experiments.

Due to extremely low power consumption, backscatter is promising to become general-purpose communication for Internet of Things (IoT) networks. The state-of-the-art BLE backscatter system RBLE can work fully with commodity Bluetooth Low Energy (BLE) devices, however, the communication reliability of RBLE is still limited by its modulation scheme based on Binary Frequency Shift Keying (BFSK). To address this issue, we refer to the modulation of active BLE radios to improve the modulation scheme of the backscatter tag. We present PBLE, a practical BLE backscatter communication system that modulates using phase shift. We conduct experiments to verify the feasibility and efficacy. Evaluation results demonstrate the feasibility and performance gains over RBLE in terms of bit error rate (BER). With PBLE, it is possible for us to reliably read the sensor data of tags using daily BLE devices.

This paper presents RapidRider, the first WiFi backscatter system that takes uncontrolled OFDM WiFi signals, e.g., 802.11a/g/n, as excitations and efficiently embeds tag data at the single-symbol rate. Such design brings us closer to the dream of pervasive backscatter communication since uncontrolled WiFi signals are everywhere. Specifically, we show that RapidRider can demodulate tag data for each OFDM symbol while previous systems rely on multi-symbol demodulation. Further, we design deinterleaving-twins decoding that enables RapidRider to use any uncontrolled WiFi signals as carriers. We prototype RapidRider using FPGAs, commodity radios, and USRPs. Comprehensive evaluations show that RapidRider's maximum throughput is 3.92x and 1.97x better than FreeRider and MOXcatter. To accommodate cases where there is only one receiver available, we design RapidRider+ that can take productive data and tag data on the same packet. Results demonstrate that it can achieve an aggregated goodput of productive and tag data around 1 Mbps on average.

Modern acoustic wearables with microphone arrays are promising to offer rich experience (e.g., 360° sound and acoustic imaging) to consumers. Realtime multi-track audio streaming with precise synchronization however poses significant challenges to the existing wireless microphone array designs that depend on complex digital synchronization as well as bulky and power-hungry hardware.

The communication reliability of state-of-the-art Bluetooth Low Energy (BLE) backscatter systems is fundamentally limited by their modulation schemes because the Binary Frequency Shift Keying (BFSK) modulation of the tag does not exactly match commodity BLE receivers designed for Gauss Frequency Shift Keying (GFSK) modulated signals with high bandwidth efficiency. Gaussian pulse shaping is a missing piece in state-of-the-art BLE backscatter systems. Inspired by active BLE and applying calculus, we present IBLE, a BLE backscatter communication system that achieves full compatibility with commodity BLE devices. IBLE leverages the fact that phase shift is the integral of frequency over time to build a reliable physical layer for BLE backscatter. IBLE uses instantaneous phase shift (IPS) modulation, GFSK modulation, and optional FEC coding to improve the reliability of BLE backscatter communication to the commodity level. We prototype IBLE using various commodity BLE devices and a customized tag with FPGA. Empirical results demonstrate that IBLE achieves PERs of 0.04% and 0.68% when the uplink distances are 2 m and 14 m respectively, which are 280x and 70x lower than the PERs of the state-of-the-art system RBLE. On the premise of meeting the BER requirements of the BLE specification, the uplink range of IBLE is 20 m. Since BLE devices are everywhere, IBLE is readily deployable in our everyday IoT applications.

This paper presents GLAC, the first 3D localization system that enables millimeter-level object manipulation for robotics using only COTS RFID devices. The key insight of GLAC is that mobility reduces ambiguity and thus improves accuracy. Unlike state-of-the-art systems that require extra time or hardware to boost performance, it draws the power of modeling mobility in a delicate way. In particular, we build a novel framework for real-time tracking using the Hidden Markov Model (HMM). In our framework, multiple Kalman filters are designed to take a single phase observation for updating mobility states, and a fast inference algorithm is proposed to efficiently process an exponentially large number of candidate trajectories. We prototype GLAC with only UHF tags and a commercial reader of four antennas. Comprehensive experiments show that the median position accuracies of x/y/z dimensions are within 1 cm for both LoS and NLoS cases. The median position accuracy for slow-moving targets is 0.41 cm, which is 2.2×, 17.3×, and 14.9× better than TurboTrack, Tagoram, and RF-IDraw, respectively. Also, its median velocity accuracy is at least 20 better than all three competitors for fast-moving targets. Besides accuracy, it achieves more than 4× localization time gains over state-of-the-art systems.

We present Apollo, a self-powered wireless biochemical system that is wearable and can continuously monitor personal health states. It has three key enablers. First, we design an ultra-low-power Bluetooth backscatter module, which can effectively take BLE signals as excitations and transmit sensor data at high data rates to smartphones. Further, at the core of self-power management is a novel multi-source harvester that can transform RF, light, and thermal energies into electrical energy. Moreover, we introduce a biochemical sensor array that can accurately measure sweat metabolites and electrolytes.We prototype Apollo on a compactly integrated flexible PCB using all commercial off-the-shelf components. Through extensive experiments and field studies, we show that it is able to measure the glucose level in sweat within 5% error rate while consuming 5.8 mW for sensing and 720 pJ/bit for wireless transmissions. This translates to over 1700x lower power than standard electrochemical workstations, and 26x lower power than existing commercial BLE chipsets. We believe Apollo marks an important step towards the dream of ubiquitous healthcare for everyone as it provides a highly convenient, self-managed, and fully integrated way for health monitoring.

Wireless tracing technologies have seen great potentials in many applications, including drawing and writing, gesture-based commanding, and gaming. Many state-of-the-art systems have recently been proposed along this line. However, none of them can strike a balance among hardware complexity, time delay, and accuracy in real-world scenarios. In this paper, we propose RF-Pen, a practical and complete RFID tracking system that achieves centimeter-level real-time tracing with 4 antennas. To do so, RF-Pen mainly employs two key designs, namely selective hologram and hybrid voting. Our selective hologram places antennas in large separation, which not only expedites the tracking process by producing a handful of good-quality candidate points but also maximizes tracing resolution. Nevertheless, a big challenge is ambiguity. To address this, we introduce hybrid voting that effectively integrates RSSI and phase measurements to evaluate the likelihood of all candidate points. This way, a precise initial position and fine-resolution tracing beams are located. We implement RF-Pen using off-the-shelf readers and tags and compare it against state-of-the-art systems. Results show that with a single reader of 4 antennas, RF-pen achieves a median trajectory error of 2.15 cm and a median position error of 12.8 cm, which are$3.7\times$and$4.1\times$better than RF-IDraw, respectively.

In-car human activity recognition opens a new opportunity toward intelligent driving behavior detection and touchless human-car interaction. Among the many sensing technologies (e.g., using cameras and wearable sensors), radio frequency identification (RFID) exhibits unique advantages given its low cost, easy deployment, and less privacy concerns. Existing RFID-based solutions for activity recognition are mostly confined to working in stable indoor spaces. The inside space of a car however is much more compact and complex, not to mention the fast-changing driving conditions. All these introduce non-negligible noises that pollute the activity-related information, and the existence of various car models in the market further complicates the problem. In this article, we for the first time closely examine the distinct factors that affect the RFID-based in-car activity recognition. We present RF-CAR, a novel RFID-based tag-free solution that well adapts to different in-car environments. RF-CAR smartly filters the domain-specific features in RF signals and retains activity-related features to the maximum extent. It then integrates a deep learning architecture and an advanced multi-adversarial domain adaptation network for training and prediction. With only one-time pre-training, RF-CAR can adapt to new data domains such as new driving conditions, car models, and human subjects for robust activity recognition. We also demonstrate that it is readily deployable in cars with commercial off-the-shelf (COTS) RFID devices. Our extensive experiments suggest that RF-CAR achieves an overall recognition accuracy of around 95 percent, which significantly outperforms the state-of-the-art solutions.

Recently backscatter communication with commodity radios has received significant attention since specialized hardware is no longer needed. The state-of-the-art BLE backscatter system, FreeRider, realizes ultra-low-power BLE backscatter communication entirely using commodity devices. It, however, suffers from several key reliability issues, including unreliable two-step modulation, productive-data dependency, and lack of interference countermeasures. To address these problems, we propose RBLE, a robust BLE backscatter system that works with an excitation BLE device and a single BLE receiver. First, it uses BLE signals with partial single tones as excitations, making single-bit modulation much more robust. Then it designs dynamic channel configuration that enables channel hopping to avoid interfered channels. Moreover, it presents BLE packet regeneration that uses adaptive encoding to further enhance reliability for various channel conditions. The prototype is implemented using TI BLE radios, iPhones, Android phones, and customized tags with FPGAs. Empirical results demonstrate that RBLE achieves more than 17x uplink goodput gains over FreeRider under indoor LoS, NLoS, and outdoor environments. We also show that RBLE can realize uplink ranges of up to 25 m for indoors and 56 m for outdoors.

In a large-scale wireless sensor network, hundreds and thousands of sensors sample and forward data back to the sink periodically. In two real outdoor deployments GreenOrbs and CitySee, we observe that some bottleneck nodes strongly impact other nodes’ data collection and thus degrade the whole network performance. To figure out the importance of a node in the process of data collection, system manager is required to understand interactive behaviors among the parent and child nodes. So we present a management tool BOND (BOttleneck Node Detector), which explains the concept of Node Dependence to characterize how much a node relies on each of its parent nodes, and also models the routing process as a Hidden Markov Model and then uses a machine learning approach to learn the state transition probabilities in this model. Moreover, BOND can predict the network dataflow if some nodes are added or removed to avoid data loss and flow congestion in network redeployment. We implement BOND on real hardware and deploy it in an outdoor network system. The extensive experiments show that Node Dependence indeed help to explore the hidden bottleneck nodes in the network, and BOND infers the Node Dependence with an average accuracy of more than 85%.

We present multiscatter, a novel backscatter design that can simultaneously work with multiple excitation signals for personal IoT sensors. Specifically, we show for the first time that the backscatter tag can identify various excitation signals in an ultra-low-power way, including WiFi, Bluetooth, and ZigBee. Further, we employ a new modulation approach, overlay modulation, that can leverage those excitation signals to convey tag data on top of productive data, which makes decoding both data possible with only a single personal radio. Since 2.4 GHz signals and personal radios are everywhere, multiscatter is readily deployable in our everyday IoT applications.

Backscatter communication with ambient excitations receives great attention recently as it provides a practical battery-free way to convey various IoT data. However, state-of the-art solutions are of low data rates and thus cannot serve highbandwidth applications, e.g., live streaming. This paper presents Hermit Crab, the first WiFi-backscatter system that achieves high-throughput communication for video streaming. The key contribution is a differential decoding algorithm using pilot phase. By doing so, it supports single symbol encoding, which is much faster than multi-symbol encoding of previous systems. In addition, Hermit Crab can recover the production data and tag data at the same time. Through extensive experiments, we show that it achieves throughputs of up to 960 Kbps with 802.1lg ambient signals, which is 7. 6x better than the state-of-the-art system. We also demonstrate that with such good throughputs, it can stably support live streaming of 480p videos at 30 fps.

In most indoor localization systems deployed on commodity WiFi infrastructure, channel state information (CSI) data is usually transmitted over multiple subcarriers of different frequencies. An observation is that there exists a certain subcarrier that can best estimate the location of the target. Based on it, we propose MobiFi to leverage deep learning to automatically select the best subcarrier. MobiFi mainly consists of two steps: First, a lightweight end-to-end Convolution Neural Network (CNN) is taken as the backbone network to extract features and do classification while avoiding serious overfitting. After selecting the best subcarrier by the first two steps, MobiFi calculates the AoA estimation and corresponding location estimation in the same way as SpotFi. Since the backbone network is lightweight, MobiFi can realize near real-time on mobile devices with guaranteed localization performance. Extensive experiments show that MobiFi is comparable to SpotFi; both methods achieve a median AoA estimation error of 8.6° and median location estimation error of 1. 5m in an indoor office scenario. At the same time, MobiFi which consumes less than 0. 21s and 1. 7s on Personal Computer (PC) and mobile devices respectively is 5 times faster than SpotFi. Particularly, because MobiFi enables real-time localization on mobile devices, it provides an economical solution for some cases where a central server is replaced by a mobile device.

Backscatter offers a novel low-cost and low-energy solution for tags to communicate with existing wireless devices. The latest Bluetooth standards (i.e., Bluetooth 4. x and Bluetooth 5) use the Bluetooth Low Energy technology, which is the mainstream of the current Bluetooth market. In this paper, we present multi-channel backscatter with BLE, a BLE backscatter communication system that achieves compatibility with standard BLE devices. Tag information in backscatter communication can be directly obtained from backscatter packets with a single BLE receiver. We present the first multichannel backscatter tag design and build a prototype of our tag using an FPGA and evaluate it with BLE devices. Our evaluation shows that the tag can backscatter BLE signals in a channel-hopping manner, which can be decoded by standard BLE devices. Results demonstrate that our backscatter system can achieve a goodput of up to 2.8 kbps.

Recently backscatter communication with commodity radios has received significant attention since specialized hardware is no longer needed. The state-of-the-art BLE backscatter system, FreeRider, realizes ultra-low-power BLE backscatter communication entirely using commodity devices. It, however, suffers from several key reliability issues, including unreliable two-step modulation, productive-data dependency, and lack of interference countermeasures. To address these problems, we propose RBLE, a reliable BLE backscatter system that works with a single commodity receiver. It first introduces direct frequency shift modulation with the single tone generated by an excitation BLE device, making robust single-bit modulation possible. Then it designs dynamic channel configuration that enables channel hopping to avoid interfered channels. Moreover, it presents BLE packet regeneration that uses adaptive encoding to further enhance reliability for various channel conditions. The prototype is implemented using TI BLE radios and customized tags with FPGAs. Empirical results demonstrate that RBLE achieves more than 17x uplink goodput gains over FreeRider under indoor LoS, NLoS, and outdoor environments. We also show that RBLE can realize uplink ranges of up to 25 m for indoors and 56 m for outdoors.

This paper presents DeTrack, a tracking system that can continuously trace WiFi objects at decimeter-level in real-time. To enable this, we make three main proposals. The first one is a super-resolution localization scheme that combines compressed sensing and expectation-maximization algorithms to iteratively resolve multi-path, which realizes better resolution compared against traditional MUSIC. The second one is a customized particle filter that takes advantage of WiFi signals and the geometric nature of AOA estimates to properly update location states and particle weights. Finally, an SVD-based multipacket fusion is employed to reinforce the signal space and improve tracking efficiency at the same time. A prototype is built using only commercial WiFi NICs. Extensive experiments demonstrate that DeTrack achieves an 80thpercentile localization accuracy of 0.9 meters and a median latency of around 90 milliseconds. As a result, DeTrack is looking to benefit a wide range of applications, e.g., indoor navigation, intelligent logistics, and smart cities.

This paper presents a systematic study on adaptive streaming over MPTCP. We start from realworld experiments with Dynamic Adaptive Streaming over HTTP (DASH) and analysis on its performance over MPTCP. We show that DASH can greatly benefit from the improved aggregated throughput by MPTCP; yet the inter-path throughput difference and the intra-path throughput fluctuation have noticeable (negative) impact, too. Without a proper design of path selection and adaptation in MPTCP, they can easily confuse the adaptation logic of DASH, resulting in low bitrates or frequent rebuffering even if high-bandwidth paths are available. We present MPTCP+, an extended multipath TCP solution to offer high quality and smooth playback for adaptive HTTP streaming. MPTCP+ incorporates a path use decision algorithm that smartly disables/enables a path to minimize the inter-path difference, and a novel congestion control algorithm that smooths congestion window evolution with multiple paths. We have implemented MPTCP+ in the MPTCP Linux kernel, with minimum change on the server-side MPTCP module only. It is fully compatible with the existing MPTCP clients and requires no change on the upper-layer protocols, too. Our experiments suggest that MPTCP+ increases the quality of experience (QoE) of DASH by up to 50%.

Backscatter communication, in which data is conveyed through reflecting excitation signals, has been advocated as a promising green technology for Internet of Things (IoT). Existing backscatter solutions however are mostly centralized, relying on a single excitation source, typically within one hop. Though recent works have demonstrated the viability of multi-hop backscatter, the excitation signal remains centralized, which attenuates quickly and fundamentally limits the communication scope. For long-range and high-quality communication, distributed excitations are expected and also naturally available as ambient signals (WiFi, BLE, cellular, FM, light, sound, etc.), albeit not being explored for boosting nearby tags for relaying.

With rapid development of radio frequency identification (RFID) technology, ever-increasing research effort has been dedicated to devising various RFID-enabled services. The missing tag identification, which is to identify all missing tags, is one of the most important services in many Internet-of-Things applications such as inventory management. Prior work on missing tag detection all rely on hash functions implemented at individual tags. However, in reality hash functions are not supported by commercial off-the-shelf (COTS) RFID tags. To bridge this gap between theory and practice, this paper is devoted to detecting missing tags with COTS Gen2 devices. We first introduce a point-to-multipoint protocol, named P2M that works in an analog frame slotted Aloha paradigm to interrogate tags and collect their electronic product codes (EPCs). A missing tag will be found if its EPC is not present in the collected ones. To reduce time cost of P2M resulted from tag response collisions, we further present a collision-free point-to-point protocol, named P2P that selectively specifies a tag to reply with its EPC in each slot. If the EPC is not received, this tag is regarded to be missing. We develop two bitmask selection methods to enable the selective query while reducing communication overhead. We implement P2M and P2P with COTS RFID devices and evaluate their performance under diverse settings.

Ever-increasing research efforts have been dedicated to radio frequency identification (RFID) systems, such as finding top-k, elephant groups, and missing-tag detection. While group labeling, which is how to tell tags their associated group data, is the common prerequisite in many RFID applications, its efficiency is not well optimized due to the transmission of useless data with only one seed used. In this paper, we introduce a unified protocol called GLMS which employs multiple seeds to construct a composite indicator vector (CIV), reducing the useless transmission. Technically, to address Seed Assignment Problem (SAP) arising during building CIV, we develop an approximation algorithm (AA) with a competitive ratio 0.632 by globally searching for the seed contributing to the most useful slot. We then further design two simplified algorithms through local searching, namely c-search-I and its enhanced version c-search-II, reducing the complexity by one order of magnitude while achieving comparable performance. We conduct extensive simulations to demonstrate the superiority of our approaches.

Recently backscatter networks have received booming interest because, they offer a battery-free communication paradigm using propagation radio waves as opposed to active radios in traditional sensor networks while providing comparable sensing functionalities, ranging from light and temperature sensors to recent microphones and cameras. While sensing data on backscatter nodes has been seen on a clear path to increasing in both volume and variety, backscatter communication is not well prepared and optimized for transferring such continuous and high-volume data. To bridge this gap, we propose a high-throughput rate adaptation scheme for backscatter networks by exploring the unique characteristics of backscatter links and the design space of the ISO 18000-6C (C1G2) protocol. Our key insight is that while prior work has left the downlink unattended, we observe that the quality of downlink is affected significantly by multipath fading and thus can degrade the uplink and overall throughput considerably. Therefore, we introduce a novel rate mapping algorithm that chooses the best rate for both the downlink and uplink. Also, we design an efficient channel estimation method fully compatible with the C1G2 protocol and a reliable probing trigger, substantially saving probing overhead. To combat interference, we further design an interference detector using clusters and lightweight countermeasures to make rate adaptation more robust. Our scheme is prototyped using commercial RFID readers and tags. The results show that we can achieve up to 2.6× throughput gain over state-of-the-art approaches across various mobility, channel, network-size, and interference conditions.

Multipath TCP (MPTCP) has recently been suggested as a promising transport protocol to boost the utilization of underlaying datacenter networks, yet it also increases the host CPU power consumption. It remains unclear whether datacenters can indeed benefit from using MPTCP from the perspective of energy efficiency. Through realworld measurement of MPTCP, we show that the energy efficiency of MPTCP is largely related to the flow completion time and the existence of link-sharing subflows. In particular, we find that the link-sharing subflows in MPTCP will significantly elevate the CPUs' power consumption on hosts. To make the matter worse, it will also reduce the transmission efficiency for both throughput-sensitive long flows and latency-sensitive short flows. To address such a problem, we present MPTCP-D, an energy-efficient enhancement of MPTCP in datacenter networks. MPTCP-D incorporates a novel congestion control algorithm that improves energy efficiency by minimizing the flow completion time. It also has a build-in subflow elimination mechanism that precludes link-sharing subflows from increasing the host CPU power consumption. We implement MPTCP-D in the Linux kernel, analyze the parameter selection in the algorithm and study its performance through packet-level simulation and on Amazon EC2. Our results show that, without degrading the performance of the long flow throughput and the short flow completion time, MPTCP-D reduces the long flow energy consumption by up to 72% compared to DCTCP for data transfers, and reduces the short flow power consumption by up to 46% compared to MPTCP with link-sharing subflows.

In-car human activity recognition is playing a critical role in detecting distracted driving and improving human-car interaction. Among multiple sensing technologies, WiFi-based in-car activity recognition exhibits unique advantages since it does not rely on visible light, avoids privacy leaks and is cost-efficient with integrated WiFi signals in cars. Existing WiFi-based recognition systems mostly focus on the relatively stable indoor space, which only yield reasonably good performance in limited situations. Based on our field studies, the in-car activity recognition, however, is much more complicated suffering from more impact factors. First, the external moving objects and the surrounding WiFi signals can cause various disturbances to the in-car activity sensing. Second, considering the compact in-car space, different car models can also lead to different multipath distortions. Moreover, different people can also perform activities in different shapes. Such extraneous information related to specific driving conditions, car models and human subjects is implicitly contained for training and prediction, inevitably leading to poor recognition performance for new environment and people.

The deeply penetrated WiFi signals not only provide fundamental communications for the massive Internet of Things devices but also enable cognitive sensing ability in many other applications, such as human activity recognition. State-of-the-art WiFi-based device-free systems leverage the correlations between signal changes and body movements for human activity recognition. They have demonstrated reasonably good recognition results with a properly placed transceiver pair, or, in other words, when the human body is within a certain sweet zone. Unfortunately, the sweet zone is not ubiquitous. When the person moves out of the area and enters a dead zone, or even just the orientation changes, the recognition accuracy can quickly decay. In this paper, we closely examine such spatial diversity in WiFi-based human activity recognition. We identify the dead zones and their key influential factors, and accordingly present WiSDAR, a WiFi-based spatial diversity-aware device-free activity recognition system. WiSDAR overshadows the dead zones yet with only one physical WiFi sender and receiver. The key innovation is extending the multiple antennas of modern WiFi devices to construct multiple separated antenna pairs for activity observing. Profiling activity features from multiple spatial dimensions can be more complicated and offer much richer information for further recognition. To this end, we propose a deep learning-based framework that integrates the hidden features from both temporal and spatial dimensions, achieving highly accurate and reliable recognition results. WiSDAR is fully compatible with commercial off-the-shelf WiFi devices, and we have implemented it on the commonly available Intel WiFi 5300 cards. Our real-world experiments demonstrate that it recognizes human activities with a stable accuracy of around 96%.

With the multi-antenna design of WiFi interfaces, phased array has become a promising mechanism for accurate WiFi localization. State-of-the-art WiFi-based solutions using Angle-of-Arrival (AoA), however, face a number of critical challenges. First, their localization accuracy degrades dramatically due to low Signal-to-Noise Ratio (SNR) and incoherent processing. Second, they tend to produce outliers when the available number of packets is low. Moreover, the prior phase calibration schemes are not multipath robust and accurate enough. All of the above degrade the robustness of localization systems. In this paper, we present ROArray, a RObust Array based system that accurately localizes a target even with low SNRs. The key insight of ROArray is to use sparse recovery and coherent processing across all available domains, including time, frequency, and spatial domains. Specifically, in the spatial domain, ROArray can produce sharp AoA spectrums by parameterizing the steering vector based on a sparse grid. Then, to expand into the frequency domain, it jointly estimates the Time-of-Arrival (ToAs) and AoAs of all the paths using multi-subcarrier OFDM measurements. Furthermore, through a novel multi-packet fusion scheme, ROArray is enabled to perform coherent estimation over multiple packets. Such coherent processing not only increases the virtual aperture size, which enlarges the number of maximum resolvable paths but also improves the system robustness to noise. In addition, ROArray includes an online phase calibration technique that can eliminate random phase offsets while keeping communication uninterrupted. Our implementation using off-the-shelf WiFi cards demonstrates that, with low SNRs, ROArray significantly outperforms state-of-the-art solutions in terms of localization accuracy; when medium or high SNRs are present, it achieves comparable accuracy.

Tag search, which is to find a particular set of tags in a radio frequency identification (RFID) system, is a key service in such important Internet-of-Things applications as inventory management. When the system scale is large with a massive number of tags, deterministic search can be prohibitively expensive, and probabilistic search has been advocated, seeking a balance between reliability and time efficiency. Given a failure probability$\frac {1}{\mathcal {O}(K)}$, where$K$is the number of tags, state-of-the-art solutions have achieved a time cost of$\mathcal {O}(K \log K)$through multi-round hashing and verification. Further improvement, however, faces a critical bottleneck of repetitively verifying each individual target tag in each round. In this paper, we present an efficient tree-based tag search (TTS) that approaches$\mathcal {O}(K)$through batched verification. The key novelty of TTS is to smartly hash multiple tags into each internal tree node and adaptively control the node degrees. It conducts bottom–up search to verify tags group by group with the number of groups decreasing rapidly. Furthermore, we design an enhanced tag search scheme, referred to as TTS+, to overcome the negative impact of asymmetric tag set sizes on time efficiency of TTS. TTS+ first rules out partial ineligible tags with a filtering vector and feeds the shrunk tag sets into TTS. We derive the optimal hash code length and node degrees in TTS to accommodate hash collisions and the optimal filtering vector size to minimize the time cost of TTS+. The superiority of TTS and TTS+ over the state-of-the-art solution is demonstrated through both theoretical analysis and extensive simulations. Specifically, as reliability demand on scales, the time efficiency of TTS+ reaches nearly 2 times at most that of TTS.

RFID-based human activity identification has become a key component in today's Internet-of-Things applications. State-of-the-art solutions mostly focus on the simple scenario with a single person in the open space. Extension to the more realistic realworld scenarios with multiple persons however is non-trivial. Given the much richer interactions among them, the backscattered signals will inevitably mixed, obscuring the information of individual activities. This is further complicated with multi-path in a common indoor environment. In this paper, we however argue that, though often considered harmful, the rich interactions combined with multi-path indeed offer more observable data. After careful processing the raw signals, critical information about the activities can be unveiled through modern learning tools. We present M2AI, which for the first time accommodates both multi-path and multi-object for activity identification. M2AI incorporates a phase calibration mechanism to automatically eliminate the frequency hopping offsets, and a novel decoupling mechanism for the periodogram and pseduospectrum in the raw signal mixture. The refined data are then fed into an advanced deep-learning engine that integrates a Convolutional Neural Network and a Long Short Term Memory network, which examines both spatial and temporal information in realtime for activity identification. Our M2AI is readily deployable using off-the-shelf RFID readers. We have implemented an M2AI prototype with Impinj UHF passive tags and a Speedway R420 reader. Experiments with multiple objects in a multipath-rich indoor environments report an activity identification accuracy of 97%, a significant gain (27%) over state-of-art solutions.

Multihop wireless networking is a key enabling technology for interconnecting a vast number of IoT devices. Measurement is fundamental to various network operations including management, diagnostics, and optimization. Out-of-band measurement approaches use external sniffers to monitor the network traffic passively, and they provide detailed information about the network. However, existing approaches do not carefully consider lossy and correlated links which are common in low-power wireless networks, resulting in unsatisfactory packet capture ratio and low measurement quality. In this paper, we present NetVision, a practical out-of-band measurement system with special consideration for sniffer deployment. By explicitly considering link quality and link correlation, we are able to achieve a high measurement quality while minimizing the deployment cost. We formulate the sniffer deployment problem as an optimization problem and propose efficient algorithms for solving this problem. We further design a set of instructions and APIs to simplify a variety of common measurement tasks. We implement NetVision on the TinyOS/TelosB platform and evaluate its performance extensively both in simulation and an indoor testbed with 80 TelosB nodes. Results show that NetVision is accurate, generic, and robust. Three typical case studies demonstrate that NetVision can facilitate various measurement and debugging tasks.

Recent years have seen various acoustic applications on mobile devices, e.g. range finding, gesture recognition, and device-to-device data transport, which use near-ultrasound signals at frequencies around 18-24 kHz. Due to the fixed low sound sample rate and hardware limitation, the highest detectable sound frequency on commercial-off-the-shelf (COTS) mobile devices is capped at 24 kHz, presenting a daunting barrier that prevents high-frequency ultrasounds from benefiting acoustic applications. To bridge this gap, we present iChemo, a technology that enables COTS mobile devices to sense high-frequency ultrasound signals. Specifically, we demonstrate how to detect the power spectral density (PSD) of a high-frequency ultrasound signal by customizing the coprime sampling algorithm on COTS devices. Through our prototype and evaluation on extensive mobile devices, we demonstrate that iChemo can sense the PSD of ultrasound at frequency of 60 kHz, which is over twice of the current sensible frequency threshold.

This paper presents WordRecorder, an efficient and accurate handwriting recognition system that identifies words using acoustic signals generated by pens and paper, thus enabling ubiquitous handwriting recognition. To achieve this, we carefully craft a new deep-learning based acoustic sensing framework with three major components, i.e., segmentation, classification, and word suggestion. First, we design a dual-window approach to segment the raw acoustic signal into a series of words and letters by exploiting subtle acoustic signal features of handwriting. Then we integrate a set of simple yet effective signal processing techniques to further refine raw acoustic signals into normalized spectrograms which are suitable for deep-learning classification. After that, we customize a deep neural network that is suitable for smart devices. Finally, we incorporate a word suggestion module to enhance the recognition performance. Our framework achieves both computation efficiency and desirable classification accuracy simultaneously. We prototype our design using off-the-shelf smartwatches and conduct extensive evaluations. Our results demonstrate that WordRecorder robustly archives 81% accuracy rate for trained users, and 75% for users without training, across a range of different environment, users, and writing habits.

In the past few years, various backscatter nodes have been invented for many emerging mobile applications, such as sports analytics, interactive gaming, and mobile healthcare. Backscatter networks are expected to provide a high-throughput and stable communication platform for those interconnected mobile nodes. Yet, through experiments, we find state-of-the-art rate adaptation methods for backscatter networks share a fundamental limitation of accommodating the hardware diversity of nodes because the common mapping paradigm that chooses the optimal rate based on the radio signal strength indicator (RSSI) or the like is hardly adaptable to hardware-dependent RSSIs. To address this issue, we propose MobiRate (Mobility-aware Rate adaptation) that fully exploits the mobility hints from PHY information and the characteristics of backscatter systems. The key insight is that mobility-hints, like velocity and position, can greatly benefit rate selection and channel probing. Specifically, we introduce a novel velocity-based loss rate estimation method that dynamically re-weighs packets based on time and mobility. In addition, we design a mobility-assisted probing trigger and a new selective-probing mechanism, significantly saving probing time. As MobiRate is fully compatible with the current standard, it is prototyped using a COTS RFID reader and a variety of commercial tags. Our extensive experiments demonstrate that MobiRate achieves up to 3.8x throughput gain over the state-of-the-art methods across a wide range of mobility, channel conditions, and tag types.

Searching for a particular group of tags in an RFID system is a key service in such important Internet-of-Things applications as inventory management. When the system scale is large with a massive number of tags, deterministic search can be prohibitively expensive, and probabilistic search has been advocated, seeking a balance between reliability and time efficiency. Given a failure probability [1/(O(K))], where K is the number of tags, state-of-the-art solutions have achieved a time cost of O(K log K) through multi-round hashing and verification. Further improvement however faces a critical bottleneck of repetitively verifying each individual target tag in each round. In this paper, we present a novel Tree-based Tag Search (TTS) that approaches O (K) through batched verification. TTS smartly hashes multiple tags into each internal tree node and adaptively controls the node degrees. It conducts bottom-up search to verify tags group by group with the number of groups decreasing rapidly. We derive the optimal hash code length and node degrees to accommodate hash collisions, and demonstrate the superiority of TTS through both theoretical analysis and extensive simulations. In particular, we show that, with increasing reliability demand and system size, TTS achieves an even higher performance gain, making it a highly scalable solution.

Many efforts have been made on sensing ultrasound with commercial-off-the-shelf (COTS) mobile devices in the recent literature. Yet due to the limited sound sample rate, current COTS mobile devices can not directly capture any sound at the frequency over 24 kHz. This issue prevents true ultrasound, of which the frequency is typically over 40 kHz, from benefiting the existing sound sensing applications. In this work, we show that by subtly customizing the sampling process, we can make COTS mobile devices hear the true ultrasound that is typically beyond their capability to fully capture. Particularly, we present a system that enable COTS mobile devices to sense the power spectrum density (PSD) of true ultrasounds, of which the frequency can be as high as 60 kHz.

With rapid development of radio frequency identification (RFID) technology, ever-increasing research effort has been dedicated to devising various RFID-enabled services. The key tag monitoring, which is to detect anomaly of key tags, is one of the most important services in such important Internet-of-Things applications as inventory management. Yet prior work assumes that all tags are armed with hashing functionality and a reader would report channel states in every slot, which is not supported by commercial off-the-shelf (COTS) RFID tags and readers. To bridge this gap, this paper is devoted to enabling key tag monitoring service with COTS devices. In particular, we introduce two anomaly monitoring protocols to detect whether there is any key tag absent from the system. The first protocol employs Q-query that works in an analog frame slotted Aloha paradigm to interrogate tags and collect tag IDs. An anomaly event will be found if at least one key tag ID is not present in the collected ones. To reduce time cost of the first protocol resulted from tag collisions, we present a collision-free method that uses select-query to specify a key tag to reply in each slot. Once there is no response in a slot, the specified key tag is regarded as a missing tag. We conduct experiments to evaluate two protocols.

Backscatter communication offers a cost- and energy-efficient means for IoT sensor data exchange. The IoT vision for ubiquitous interconnection, in practice, demands multi-hop connectivity for robust and scalable sensor networks, as well as compatibility with such prevailing wireless technologies as WiFi. Today's backscatter solutions however typically follow a single-hop paradigm, i.e., tags do not relay for each other. This paper presents X-Tandem, a multi-hop backscatter system that works with commodity WiFi devices. For the first time, we demonstrate that sensing tags can not only work as relays for each other but also modulate their sensing data into a single backscatter packet, which remains a legit WiFi packet that can be decoded with any commercial WiFi NICs. We discuss the design details of X-Tandem and have built a prototype with FPGAs and off-the-shelf WiFi devices. The prototype demonstrates a two-hop implementation, achieving a throughput up to 200 bps with tag-to-tag distances up to 0.4 m and communication ranges up to 8 m. Compared to single-hop solutions, X-Tandem can improve backscatter throughput by more than 10x in challenging indoor environments with obstacles.

Backscatter WiFi offers a novel low-cost and low-energy solution for RFID tags to communicate with existing WiFi devices. State-of-the-art backscatter WiFi solutions have seldom explored advanced features in the latest WiFi standards, in particular, spatial multiplexing, which has been the cornerstone for 802.11n and beyond. In this paper, we present MOXcatter, a WiFi backscatter communication system that works with spatial streams using commodity radios, while keeping the ongoing data communication unaffected. In MOXcatter, a backscatter tag can embed its sensing data on ambient spatial-stream packets, and both the sensing data and the original packets can be decoded by commodity WiFi devices. We have built a MOXcatter prototype with FPGAs and commodity WiFi devices. The experiments show that MOXcatter achieves up to 50 Kbps throughput for a single stream and up to 1 Kbps for double streams with a communication range (tag-to-RX) up to 14 m. We discuss the tradeoffs therein and possible enhancements, and also showcase the applicability of our design through a sensor communication system.

With the multi-antenna design of WiFi interfaces, phased array has become a promising mechanism for accurate WiFi localization. State-of-the-art WiFi-based solutions using AoA (Angle-of-Arrival), however, face a number of critical challenges. First, their localization accuracy degrades dramatically when the Signal-to-Noise Ratio (SNR) becomes low. Second, they do not fully utilize coherent processing across all available domains. In this paper, we present ROArray, a Robust Array based system that accurately localizes a target even with low SNRs. In the spatial domain, ROArray can produce sharp AoA spectrums by parameterizing the steering vector based on a sparse grid. Then, to expand into the frequency domain, it jointly estimates the ToAs (Time-of-Arrival) and AoAs of all the paths using multi-subcarrier OFDM measurements. Furthermore, through multi-packet fusion, ROArray is enabled to perform coherent estimation across the spatial, frequency, and time domains. Such coherent processing not only increases the virtual aperture size, which enlarges the number of maximum resolvable paths, but also improves the system robustness to noise. Our implementation using off-the-shelf WiFi cards demonstrates that, with low SNRs, ROArray significantly outperforms state-of-the-art solutions in terms of localization accuracy; when medium or high SNRs are present, it achieves comparable accuracy.

Recently backscatter networks have received booming interest because, they offer a battery-free communication paradigm using propagation radio waves as opposed to active radios in traditional sensor networks while providing comparable sensing functionalities, ranging from light and temperature sensors to recent microphones and cameras. While sensing data on backscatter nodes has been seen on a clear path to increase in both volume and variety, backscatter communication is not well prepared and optimized for transferring such continuous and high-volume data. To bridge this gap, we propose a high-throughput rate adaptation scheme for backscatter networks by exploring the unique characteristics of backscatter links and the design space of the ISO 18000-6C (C1G2) protocol. Our key insight is that while prior work has left the downlink unattended, we observe that the quality of downlink is affected significantly by multipath fading and thus can degrade the uplink and overall throughput considerably. Therefore, we introduce a novel rate mapping algorithm that chooses the best rate for both the downlink and uplink. Also, we design an efficient channel estimation method fully compatible with the C1G2 protocol and a reliable probing trigger, substantially saving probing overhead. Our scheme is prototyped using a COTS RFID reader and tags. The results show that we achieve up to 2.5x throughput gain over state-of-the-art approaches across various mobility, channel, and network-size conditions.

The state-of-the-art cloud computing platforms are facing challenges, such as the high volume of crowdsourced data traffic and highly computational demands, involved in typical deep learning applications. More recently, Edge Computing has been recently proposed as an effective way to reduce the resource consumption. In this paper, we propose an edge learning framework by introducing the concept of edge computing and demonstrate the superiority of our framework on reducing the network traffic and running time.

There has been explosive growth in Mobile Personal Livestreaming (MPL) market since 2016. MPL services are booming not only because they introduce the popular live content by spontaneous and personalized broadcasters, but also because they are deliberately designed to be the innovative social networking service (SNS) platforms. The latter is a very important aspect that distinguishes MPL from the traditional livestreaming services. In this paper, we study the social networking of a large scale MPL service “Inke” (with more than 200 million registered users, 15 million daily active users) in China. By analyzing the dataset we crawl and the features of Inke app, we show that the social media stickiness of Inke comes from three aspects: the follower-followee model, the virtual-gift-based incentive mechanism, and the multi-perspective interactivity between broadcasters and viewers. First, Inke introduces the follower-followee model rather than the traditional broadcaster-viewer model, and every user in Inke can be a broadcaster. This makes MPL have some different patterns from both the traditional livestreaming services and SNS platforms. Second, Inke use virtual gift giving and user ranking as its incentive mechanism. Our measurement results show that this mechanism can indeed enhance user stickiness. Furthermore, Inke incorporates a variety of features during broadcasting to strengthen interactivity. The insight we gain in this paper has important implications for both existing and future designs.

Unmanned aerial vehicles (UAVs) are experiencing a major increase in popularity in both consumer and industrial markets as prices fall and the technology matures. No longer are drones limited to military purposes as manufacturers begin to mass produce civilian models, ushering in a new era of transportation technology. While consumer drones are still in their infancy stage, there is little in the way of rules and regulations regarding privacy issues of these new devices. In this paper, we design an energy efficient off-the-shelf hardware system capable of detecting and selectively disabling video feeds of WiFi based consumer drones if they enter a defended area.

Radio Frequency Identification (RFID) is an emerging technology for electronic labeling of objects for the purpose of automatically identifying, categorizing, locating, and tracking the objects. But in their current form RFID systems are susceptible to cloning attacks that seriously threaten RFID applications but are hard to prevent. Existing protocols aimed at detecting whether there are cloning attacks in single-reader RFID systems. In this paper, we investigate the cloning attacks identification in the multireader scenario and first propose a time-efficient protocol, called the time-efficient Cloning Attacks Identification Protocol (CAIP) to identify all cloned tags in multireaders RFID systems. We evaluate the performance of CAIP through extensive simulations. The results show that CAIP can identify all the cloned tags in large-scale RFID systems fairly fast with required accuracy.

Many radio frequency identification (RFID) applications, such as virtual shopping cart and tag-assisted gaming, involve sensing and recognizing tag mobility. However, existing RFID localization methods are mostly designed for static or slowly moving targets (less than 0.3m/sec). More importantly, we observe that prior methods suffer from serious performance degradation for detecting real-world moving tags in typical indoor environments with multipath interference. In this article, we present i 2 tag, an intelligent mobility-aware activity identification system for RFID tags in multipath-rich environments (e.g., indoors). i 2 tag employs a supervised learning framework based on our novel fine-grain mobility provile, which can quantify different levels of mobility. Unlike previous methods that mostly rely on phase measurement, i 2 tag takes into account various measurements, including RSSI variance, packet loss rate, and our novel relative phase--based fingerprint. Additionally, we design a multidimensional dynamic time warping--based algorithm to robustly detect mobility and the associated activities. We show that i 2 tag is readily deployable using off-the-shelf RFID devices. A prototype has been implemented using a ThingMagic reader and standard-compatible tags. Experimental results demonstrate its superiority in mobility detection and activity identification in various indoor environments.

Cardinality estimation is one of the fundamental problems in large-scale radio frequency identification systems. While many efforts have been made to achieve faster approximate counting, the accuracy of estimates itself has not received enough attention. Specifically, most state-of-the-art schemes share a two-phase paradigm implicitly or explicitly, which needs a rough estimate first and then refines it to a final estimate meeting the desired accuracy; we observe that the final estimate can largely deviate from the expectation due to the skewed rough estimate, i.e., the accuracy of final estimates is not rigorously bounded. This negative impact is hidden because former solutions either assume perfect rough estimates or rough estimates that can be produced by uniform random data or perfect hash functions that can turn any data into uniform random data. Unfortunately, both of them are hard to meet in practice. To address the above issues, we propose a novel scheme, namely, “rigorous and practical cardinality (RPC)” estimation. RPC adopts the two-phase paradigm, in which the rough estimate is derived in the first phase using pairwise-independent hashing. In the second phase, we employ t-wise-independent hashing to reinforce the rough estimate to meet arbitrary accuracy requirements. We validate the effectiveness and performance of RPC through theoretical analysis and extensive simulations. The results show that the RPC can meet the desired accuracy all the time with diverse practical settings while previous designs fail with non-uniform data.

One of the most important applications of radio frequency identification (RFID) technology is to detect unknown tags brought by new tagged items, misplacement, or counterfeit tags. While unknown tag identification is able to pinpoint all the unknown tags, probabilistic unknown tag detection is preferred in large-scale RFID systems that need to be frequently checked up, e.g., real-time inventory monitoring. Nevertheless, most of the previous solutions are neither efficient nor reliable. The communication efficiency of former schemes is not well optimized due to the transmission of unhelpful data. Furthermore, they do not consider characteristics of unreliable wireless channels in RFID systems. In this paper, we propose a fast and reliable method for probabilistic unknown tag detection, white paper (WP) protocol. The key novelty of WP is to build a new data structure of composite message that consists of all the informative data from several independent detection synopses; thus it excludes useless data from communication. Furthermore, we employ packet loss differentiation and adaptive channel hopping techniques to combat unreliable backscatter channels. We implement a prototype system using USRP software-defined radio and WISP tags to show the feasibility of this design. We also conduct extensive simulations and comparisons to show that WP outperforms previous methods. Compared with the state-of-the-art protocols, WP achieves more than 2× performance gain in terms of time-efficiency when all the channels are assumed free of errors and the number of tags is 10000, and achieves up to 12× success probability gain when the burstiness is more than 80%.

Backscatter communication networks receive much attention recently due to the small size and low power of backscatter nodes. As backscatter communication is often influenced by the dynamic wireless channel quality, rate adaptation becomes necessary. Most existing approaches share a common drawback: they do not distinguish channel qualities from different nodes or sub-channels. Consequently, the transmission rate may be improperly selected, resulting in low network throughput. Through extensive experimental studies, we observe that channel diversity plays a significant role in rate selection. Therefore, there are opportunities of exploiting channel diversity for better rate adaptation, improving network throughput. In this paper, we propose a Channel-Aware Rate Adaptation framework (CARA) for backscatter communication networks. By employing a lightweight channel probing scheme, we are able to obtain fine-grained channel information that enables accurate channel estimation. We further design a novel channel selection algorithm, benefiting as many backscatter nodes as possible. On each selected channel, CARA chooses data rate with respect to the node that has the best channel condition. We implement CARA on commercial readers and the experiment results show that CARA achieves up to 4× goodput gain compared with state-of-the-art rate adaptation scheme.

One of the most important applications of Radio Frequency Identification (RFID) technology is to detect unknown tags brought by new tagged items moved in, misplacement, or counterfeit tags. While unknown tag identification is able to pinpoint all the unknown tags, probabilistic unknown tag detection is preferred in large-scale RFID systems that need to be frequently checked up, e.g., real-time inventory monitoring. Nonetheless, we find that the efficiency of most previous works is not well optimized due to the transmission of unhelpful data. In this paper, we propose a fast and reliable method for probabilistic unknown tag detection, White Paper (WP) protocol. The key novelty of WP is to build a composite message data structure that consists of all the informative data from several independent detection synopses, i.e., excluding the useless data from communication. Hence, this design allows us to optimize the detection and communication efficiency at the same time. In particular, the compact detection synopsis is designed and tuned to minimize the failure probability for detection and the detection message is compositely constructed to reduce the transmission overhead, achieving the optimal detection and communication efficiency, respectively. We implement a prototype system using USRP software-defined radio and WISP tags to show the feasibility of this design. We also conduct extensive simulations and comparisons to show that WP achieves more than 2x performance gain compared to the state-of-the-art protocols.

Counting the number of tags is a fundamental issue and has a wide range of applications in RFID systems. Most existing protocols, however, only apply to the scenario where a single reader counts the number of tags covered by its radio, or at most the union of tags covered by multiple readers. They are unable to achieve more complex counting objectives, i.e., counting the number of tags in a composite set expression such as (S1∪ S2) - (S3∩ S4). This type of counting has realistic significance as it provides more diversity than existing counting scenario, and can be applied in various applications. We formally introduce the RFID composite counting problem, which aims at counting the tags in an arbitrary set expression and obtain its strong lower bounds on the communication cost. We then propose a generic Composite Counting Framework (CCF) that provides estimates for any set expression with desired accuracy. The communication cost of CCF is proved to be within a small factor from the optimal. We build a prototype system for CCF using USRP software defined radio and Intel WISP computational tags. Also, extensive simulations are conducted to evaluate the performance of CCF. The experimental results show that CCF is generic, accurate and time-efficient.

Radio Frequency Identification (RFID) technology plays an important role in supply chain logistics and inventory control. In these applications, a series of scanning operations at different locations are often needed to cover the entire inventory (tags). In such continuous scanning scenario, adjacent scans inevitably read overlapping tags multiple times. Most existing methods suffer from low scanning efficiency when the overlap is small, since they do not distinguish the size of overlap which is an important factor of scanning performance. In this paper, we analytically unveil the fundamental relationship between the performance of continuous scanning and the size of overlap, deriving a critical threshold for the selection of scanning strategy. Further, we design an accurate estimator to approximate the overlap. Combining the estimate and a compact data structure, an adaptive scanning scheme is introduced to achieve low communication time. Through detailed analysis and extensive simulations, we demonstrate that the proposed scheme significantly outperforms previous approach in total scanning time.

Many algorithms have been introduced to deterministically authenticate Radio Frequency Identification (RFID) tags, while little work has been done to address scalability issue in batch authentications. Deterministic approaches verify tags one by one, and the communication overhead and time cost grow linearly with increasing size of tags. We design a fast and scalable counterfeits estimation scheme, INformative Counting (INC), which achieves sublinear authentication time and communication cost in batch verifications. The key novelty of INC builds on an FM-Sketch variant authentication synopsis that can capture key counting information using only sublinear space. With the help of this well-designed data structure, INC is able to provide authentication results with accurate estimates of the number of counterfeiting tags and genuine tags, while previous batch authentication methods merely provide 0/1 results indicating the existence of counterfeits. We conduct detailed theoretical analysis and extensive experiments to examine this design and the results show that INC significantly outperforms previous work in terms of effectiveness and efficiency.

Sensor networks are widely used in various domains like the intelligent transportation systems. Users issue queries to sensors and collect sensing data. Due to the low quality sensing devices or random link failures, sensor data are often noisy. In order to increase the reliability of the query results, continuous queries are often employed. In this work we focus on continuous holistic queries like Median. Existing approaches are mainly designed for non-holistic queries like Average. However, it is not trivial to answer holistic ones due to their non-decomposable property. We first propose two schemes based on the data correlation between different rounds, with one for getting the exact answers and the other one for deriving the approximate results. We then combine the two proposed schemes into a hybrid approach, which is adaptive to the data changing speed. We evaluate this design through extensive simulations. The results show that our approach significantly reduces the traffic cost compared with previous works while maintaining the same accuracy.

Network diagnosis is crucial in managing a wireless sensor network (WSN) since many network-related faults, such as node and link failures, can easily happen. Diagnosis tools usually consist of two key components, information collection and root-cause deduction, while in most cases information collection process is independent with root-cause deduction. This results in either redundant information which might pose high communication burden on WSNs, or incomplete information for root-cause inference that leads false judgments. To address the issue, we propose DID, a directional diagnosis approach, in which the diagnosis information acquirement is guided by the fault inference process. Through several rounds of incremental information probing and fault reasoning, root causes of the network abnormalities with high credibility are deduced. We employ a node tracing scheme to reconstruct the topical topology of faulty regions and build the inference model accordingly. We implement the DID approach in our forest monitoring sensor network system, GreenOrbs. Experimental results validate the scalability and effectiveness of this design.

Efficient tag identification is fundamentally required in large-scale RFID systems. Tag signal collision degrades identification efficiency as tag IDs involved in collision cannot be decoded. The situation becomes even worse in large-scale RFID systems when tag cardinality booms. Existing anti-collision protocols focus on either reducing collision probability or adopting spread spectrum techniques. Unfortunately, the former approach cannot resolve collision radically and the latter one occupies extra bandwidth resources. To address these issues, we propose to resolve tag collision using orthogonal Walsh code, in which tags map their IDs to a group of Walsh codes and transmit them sequentially. The reader can retrieve tag IDs by inverse mapping even under collision circumstances. We further design a new efficient tag identification protocol, Wonder, which reduces identification time without spreading the bandwidth. We conduct extensive simulations to examine its effectiveness and the results show that our protocol significantly improves identification efficiency over previous anti-collision protocols.

Counting the number of RFID tags is a fundamental issue and has a wide range of applications in RFID systems. Most existing protocols, however, only apply to the scenario where a single reader counts the number of tags covered by its radio, or at most the union of tags covered by multiple readers. They are unable to achieve more complex counting objectives, i.e., counting the number of tags in a composite set expression such as (S_1 big cup S_2) - (S_3 big cap S_4). This type of counting has realistic significance since it provides more diversity than existing counting scenario, and can be applied in various applications. In this paper, we formally introduce the RFID composite counting problem, which aims at counting the tags in arbitrary set expression. We obtain strong lower bounds on the communication cost of composite counting. We then propose a generic Composite Counting Framework (CCF) that provides estimates for any set expression with desired accuracy. The communication cost of CCF is proved to be within a small factor from the optimal. We build a prototype system for CCF using USRP software defined radio and Intel WISP computational tags. Also, extensive simulations are conducted to evaluate the performance of CCF. The experimental results show that CCF is generic, accurate and time-efficient.

In a large-scale wireless sensor network, thousands of sensor nodes periodically generate and forward data back to the sink. In our recent outdoor deployment, we observe that some bottleneck nodes can greatly determine other nodes' data collection ratio, and thus affect the whole network performance. To figure out the importance of a node in data collection, the manager needs to understand the interactive behaviors among the parent and child nodes. To address this issue, we present a management tool BOND (Bottleneck Node Detector). We introduce the concept of Node Dependence to characterize how much a node relies on each of its parent nodes. BOND models the routing process as a Hidden Markov Model, and uses a machine learning approach to learn the state transition probabilities in this model based on the observed traces. BOND utilizes Node Dependence to explore the hidden bottleneck nodes in the network. Moreover, we can predict how adding or removing the sensor nodes would impact the data flow, thus avoid data loss and flow congestion in redeployment. We implement our tool on real hardware and deploy it in an outdoor system. Our extensive experiments show that BOND infers the Node Dependence with an average accuracy of more than 85%.

One important issue of RFID applications is to estimate the cardinality of large-scale RFID tags in the interested region. From a practical perspective, we require: (i) the estimate can be arbitrarily accurate, and (ii) its time cost should be scalable with the tags size, regardless of the tags distribution. Existing solutions, however, either assume the use of hash functions with ideal random properties, or impose unacceptable computation/storage overhead for tags. More importantly, those approaches only give asymptotic results and fail to provide rigorous bounds for the rate of convergence. In this paper, we propose a new scheme, Arbitrarily Accurate Approximation (A3), to reliably estimate the number of tags with any desired accuracy. In particular, for a given requirement of (ε,δ), we show that A3achieves O((log log n+ε-2) log δ-1) time efficiency. Results show that A3significantly outperforms previous designs under various distributions of tags.

Radio Frequency Identification (RFID) technology plays an important role in supply chain logistics and inventory control. In these applications, a series of scanning operations at different locations are often needed to cover the entire inventory (tags). In such continuous scanning scenario, adjacent scans inevitably read overlapping tags multiple times. Most existing methods suffer from low scanning efficiency when the overlap is small, since they do not distinguish the size of overlap which is an important factor of scanning performance. In this paper, we analytically unveil the fundamental relationship between the performance of continuous scanning and the size of overlap, deriving a critical threshold for the selection of scanning strategy. Further, we design an accurate estimator to approximate the overlap. Combining the estimate and a compact data structure, an adaptive scanning scheme is introduced to achieve low communication time. Through detailed analysis and extensive simulations, we demonstrate that the proposed scheme significantly outperforms previous approach in total scanning time.

Radio Frequency Identification technology (RFID) is widely used in many applications, such as asset monitoring, e-passport and electronic payment, and is becoming one of the most effective solutions in cyber physical system. Since the identification alone does not provide any guarantee that tag corresponds to genuine identity, authentication of tag information is needed in most RFID systems. Meanwhile, as the number of tags is rapidly growing in recent years, per-tag based methods suffer from severely low efficiency and thus give way to probabilistic batch authentication. Most previous methods, however, share a common drawback from statistical perspective: they fail to explore correlation information, i.e., they do not comprehensively utilize all the information in authentication data structures. In addition, those schemes are not scalable well when multiple tag sets need to be verified simultaneously. In this paper, we propose a fast and efficient batch authentication scheme, Wise Counting (WIC), for large-scale RFID systems. We are the first to formally introduce the general batch authentication problem with multiple tag sets and give counterfeits estimation scheme with high efficiency. By employing a novel hierarchical authentication structure, we show that WIC is able to fast and efficiently authenticate both a single tag set and multiple tag sets in an easy, intuitive way. Through detailed theoretical analysis and extensive simulations, we validate the design of WIC and demonstrate its large superiority over state-of-the art approaches.

Existing approaches to diagnosing sensor networks are generally sink based, which rely on actively pulling state information from sensor nodes so as to conduct centralized analysis. First, sink-based tools incur huge communication overhead to the traffic-sensitive sensor networks. Second, due to the unreliable wireless communications, sink often obtains incomplete and suspicious information, leading to inaccurate judgments. Even worse, it is always more difficult to obtain state information from problematic or critical regions. To address the given issues, we present a novel self-diagnosis approach, which encourages each single sensor to join the fault decision process. We design a series of fault detectors through which multiple nodes can cooperate with each other in a diagnosis task. Fault detectors encode the diagnosis process to state transitions. Each sensor can participate in the diagnosis by transiting the detector's current state to a new state based on local evidences and then passing the detector to other nodes. Having sufficient evidences, the fault detector achieves the Accept state and outputs a final diagnosis report. We examine the performance of our self-diagnosis tool called TinyD2 on a 100-node indoor testbed and conduct field studies in the GreenOrbs system, which is an operational sensor network with 330 nodes outdoor.

Much research has been conducted to securely outsource multiple parties' data aggregation to an untrusted aggregator without disclosing each individual's privately owned data, or to enable multiple parties to jointly aggregate their data while preserving privacy. However, those works either require secure pair-wise communication channels or suffer from high complexity. In this paper, we consider how an external aggregator or multiple parties can learn some algebraic statistics (e.g., sum, product) over participants' privately owned data while preserving the data privacy. We assume all channels are subject to eavesdropping attacks, and all the communications throughout the aggregation are open to others. We propose several protocols that successfully guarantee data privacy under this weak assumption while limiting both the communication and computation complexity of each participant to a small constant.

Many algorithms have been introduced to deterministically authenticate Radio Frequency Identification (RFID) tags, while little work has been done to address the scalability issue in batch authentications. Deterministic approaches verify them one by one, and the communication overhead and time cost grow linearly with increasing size of tags. We design a fine-grained batch authentication scheme, INformative Counting (INC), which achieves sublinear authentication time and communication cost in batch verifications. INC also provides authentication results with accurate estimates of the number of counterfeiting tags and genuine tags, while previous batch authentication methods merely provide 0/1 results indicating the existence of counterfeits. We conduct detailed theoretical analysis and extensive experiments to examine this design and the results show that INC significantly outperforms previous work in terms of effectiveness and efficiency.

Driven by rising application demands, the scale of Wireless Sensor Networks has grow rapidly in recent years. Thus, the traditional network management pattern, in which the sink is responsible for managing the entire network, reveals many problems in terms of both performance and efficiency. We propose an energy-efficient scheme to improve the performance of online network management and diagnosis services based on the quality of interactive communications in large-scale sensor networks. By abstracting the uncertain network model from our deployed sensor systems, we define the manageable nodes. Under different practical constraints, we then design algorithms in which multiple management centers can work in cooperation to cover as many manageable nodes as many as possible. Using the data obtained from our urban sensing sensor network system, CitySee with 494 nodes, we conduct trace-driven simulations. The experimental results verify the feasibility and effectiveness of this design.