ECG monitoring during human activities is crucial since many heart attacks occur when people are exercising, driving a car, operating a machine, etc. Unfortunately, existing ECG monitoring devices fail to timely detect abnormal ECG signals during activities due to the need for many cables or a sustained press on devices (e.g., smartwatches). This paper introduces ZeroEcg, a wireless, battery-free, lightweight, electronic-skin-like tag integrated with commodity RFIDs, which can continuously track a user's ECG during activities. By exploring and leveraging the RFID MOSFET switch, which is traditionally used for backscatter modulation, we map the ECG signal to the RFID RSS and phase measurement. It opens a new RFID sensing approach for sensing any physical world variable that can be translated into voltage signals. We model and analyze the RFID MOSFET-based backscatter modulation principle, providing design guidance for other sensing tasks. Real-world results illustrate the effectiveness of ZeroEcg on ECG sensing.

Battery-free sensing has revolutionized IoT applications, but current solutions relying on signal variations between transmitted and backscattered signals remain vulnerable to environmental dynamics and deployment variations. This paper promotes a paradigm shift: inferring targets through antenna impedance variations instead of signal fluctuations, thereby eliminating the impact of unpredictable wireless communication. We demonstrate the effectiveness of this paradigm by reimplementing three existing applications: RIO [1], Keystub [2], and RF-EATS [3]. Compared to original signal-based implementations, our approach shows significant improvements in accuracy and robustness across diverse environments. Furthermore, By integrating antenna engineering with advanced materials science, we also transform antennas into innovative sensors for pressure, temperature, and UV light sensing. This interdisciplinary methodology pushes the boundaries of battery-free sensing, opening new avenues for IoT applications.

We present MetaAI, a novel wireless computing paradigm that integrates neural network computation directly into wireless signal propagation. Unlike traditional approaches that treat wireless channels as mere data conduits, MetaAI transforms them into active computing elements through programmable metasurfaces, enabling concurrent data transmission and neural network processing. By leveraging the inherent linearity of both wireless propagation and neural networks, our design resolves the fundamental mismatch between sequential wireless transmission and parallel neural computation, while supporting efficient multi-sensor late-stage data fusion. We implemented MetaAI using metasurfaces at both dual-band (2.4/5 GHz) and single-band (3.5 GHz) frequencies. Extensive experiments demonstrate robust performance across diverse classification tasks, achieving 82.8% average accuracy (up to 89.8%) even with a simple linear architecture. Multi-sensor fusion further improves accuracy by up to 27.06%. MetaAI represents a fundamental shift in Edge AI architecture, where wireless infrastructure becomes an integral part of the computing pipeline.

We present METAATTACK, the first approach to leverage acoustic metamaterials for inaudible attacks for voice control systems. Compared to the state-of-the-art inaudible attacks requiring complex and large speaker setups, METAATTACK achieves a longer attacking range and higher accuracy using a compact, portable device small enough to be put into a carry bag. These improvements in portability and stealth have led to the practical applicability of inaudible attacks and their adaptation to a wider range of scenarios. We demonstrate how the recent advancement in metamaterials can be utilized to design a voice attack system with carefully selected implementation parameters and commercial off-the-shelf components. We showcase that METAATTACK can be used to launch inaudible attacks for representative voice-controlled personal assistants, including Siri, Alexa, Google Assistant, XiaoAI, and Xiaoyi. The average success rate of all assistants is 76%, with a range of 8.85 m.

Mobile devices have increasingly integrated with numerous deep learning-based visual applications, such as object classification and recognition models. While these models perform well in controlled environments, their effectiveness declines in real-world environment due to out-of-distribution (OOD) data not seen during training. Existing methods for detecting OOD data often compromise normal data recognition and require extensive training on unattainable OOD data. To address these issues, we propose$\mathtt {POD}$, a framework designed to enhance mobile visual applications by providing high-precision OOD detection without affecting original model performance. In the offline phase,$\mathtt {POD}$generates OOD detectors from any classification model by analyzing model's neuron responses to various data types. In the online phase, it continuously adjusts decision boundaries by integrating results from both the original model and the detector. Evaluated on two public datasets and one self-collected dataset across various popular classification models,$\mathtt {POD}$significantly improves OOD detection performance while maintaining the accuracy of original models.

Wi-Fi technology has emerged as a promising solution for contact-free sensing owing to the pervasiveness of Wi-Fi signals in indoor environments. However, Wi-Fi sensing faces several fundamental issues, including limited sensing range and unstable orientation-dependent sensing performance, hindering the widespread adoption of Wi-Fi sensing in real-life scenarios. In this paper, we propose RISensing, a novel system that leverages Reconfigurable Intelligent Surfaces (RIS) to address these two fundamental issues of Wi-Fi sensing and bring Wi-Fi sensing one step closer to real-world adoption. Unlike prior Wi-Fi sensing works which typically rely on a single target reflection signal to capture the target movement, RISensing utilizes two target reflection signals, i.e., the direct target reflection signal and RIS-based target reflection signal, to boost the sensing capability. RISensing characterizes the RIS-based target reflection signal, and constructively combines it with the direct target reflection. We evaluate the sensing performance of RISensing in various environments, including corridor, office and lab. Extensive experiments demonstrate RISensing can improve the sensing range of Wi-Fi from 4 m to 23 m, and effectively mitigate the orientation-dependent issue.

Contact-free finger gesture recognition unlocks plenty of applications in smart Human-Computer Interaction (HCI). However, existing solutions either require users to wear sensors on their fingers or use continuously monitored cameras, raising concerns regarding user comfort and privacy. In this paper, we propose mmFinger, an accurate and robust mmWave-based finger gesture recognition system that can extend the range of available custom commands. The core idea is that mmFinger leverages the finger tapping pattern as a basic gesture and encodes different number combinations of the basic gesture like Morse code. To enable reliable recognition across different locations and for various users, we carefully design a robust feature Dop-profile to effectively characterize finger movements. Furthermore, by leveraging the multi-views provided by multiple antennas of radar, we develop an adaptive weighted feature fusion network to enhance the system's robustness. Finally, we devise a novel sequence prediction network to enable the system to recognize new gestures without retraining. Comprehensive experiments demonstrate that mmFinger can achieve an average recognition accuracy of 92% for 36 predefined gestures and 88% for 5 new user-defined commands, and is robust against finger location and user diversity.

Pesticide overuse poses significant risks to human health and environmental integrity. Addressing the limitations of existing approaches, which struggle with the diversity of pesticide compounds, portability issues, and environmental sensitivity, this paper introduces Hornbill. A wireless and battery-free electrochemical bio-tag that integrates the advantages of NFC technology with electrochemical biosensors for portable, precise, and touchless multi-pesticide detection. The basic idea of Hornbill is comparing the distinct electrochemical responses between a pair of biological receptors and different pesticides to construct a unique set of feature fingerprints to make multi-pesticide sensing feasible. To incorporate this idea within small NFC tags, we reengineer the electrochemical sensor, spanning the antenna to the voltage regulator. Additionally, to improve the system's sensitivity and environmental robustness, we carefully design the electrodes by combining microelectrode technology and materials science. Experiments with 9 different pesticides show that Hornbill achieves a mean accuracy of 93% in different concentration environments and its sensitivity and robustness surpass that of commercial electrochemical sensors.

The majority of existing Wi-Fi backscatter systems transmit tag data at rates lower than 250 kbps, as the tag data is modulated at OFDM symbol level, allowing for demodulation using commercial Wi-Fi receivers. However, it is necessary to modulate tag data at OFDM sample level to satisfy the requirements for higher throughput. A comprehensive theoretical analysis and experimental investigation conducted in this paper demonstrates that demodulating sample-level modulated tag data using commercial Wi-Fi receivers is unattainable due to excessive computational overhead and demodulation errors. This is because the significant tag information dispersion, loss, and shuffling are caused by Wi-Fi physical layer operations. We conclude that the optimal position for demodulation is the time-domain IQ samples, which do not undergo any Wi-Fi physical layer operations and preserve the intact, ordered, and undispersed information of tag-modulated data, thereby minimizing complexity and maximizing accuracy.

Gas sensing plays a key role in detecting explosive/toxic gases and monitoring environmental pollution. Existing approaches usually require expensive hardware or high maintenance cost, and are thus ill-suited for large-scale long-term deployment. In this paper, we propose Gastag, a gas sensing paradigm based on passive tags. The heart of Gastag design is embedding a small piece of gas-sensitive material to a cheap RFID tag. When gas concentration varies, the conductivity of gas-sensitive materials changes, impacting the impedance of the tag and accordingly the received signal. To increase the sensing sensitivity and gas concentration range capable of sensing, we carefully select multiple materials and synthesize a new material that exhibits high sensitivity and high surface-to-weight ratio. To enable a long working range, we redesigned the tag antenna and carefully determined the location to place the gas-sensitive material in order to achieve impedance matching. Comprehensive experiments demonstrate the effectiveness of the proposed system. Gastag can achieve a median error of 6.7 ppm for CH4 concentration measurements, 12.6 ppm for CO2 concentration measurements, and 3 ppm for CO concentration measurements, outperforming a lot of commodity gas sensors on the market. The working range is successfully increased to 8.5 m, enabling the coverage of many tags with a single reader, laying the foundation for large-scale deployment.

While Reconfigurable Intelligent Surfaces (RIS) have been shown to enhance OFDM communication performance, this paper unveils a potential security concern arising from widespread RIS deployment. Malicious actors could exploit vulnerabilities to hijack or deploy rogue RIS, transforming them from communication boosters into attackers. We present a novel attack that disrupts the critical orthogonality property of OFDM subcarriers, severely degrading communication performance. This attack is achieved by manipulating the RIS to generate frequency-shifted reflections/harmonics of the original OFDM signal. We also propose algorithms to simultaneously beamform the multiple RIS-generated frequency-shifted reflections towards selected targets. Extensive experiments conducted in indoor, outdoor, 3D, and office settings demonstrate that Hydra can achieve a 90% throughput reduction in targeted attack scenarios and a 43% throughput reduction in indiscriminate attack scenarios. Furthermore, we validated the effectiveness of our attacks on both the 802.11 protocol and the 5G NR protocol.

Acoustic metasurface was recently proposed to enhance the performance of acoustic communication and sensing. While promising, there are two issues hindering the adoption of acoustic metasurface for real-life usage. The first issue is that configurable metasurface is still expensive and unscalable. The second issue is that it is difficult for an acoustic metasurface to work in a large frequency range. In this paper, we present a wideband and configurable acoustic metasurface for the first time. We show that with a large number of metasurface elements, a cheap and simple two-state element design can achieve performance very close to that achieved by expensive continuous-state elements. We also fine-tune the geometric parameter of the element structure to support similar phase changes across a large frequency range, laying the foundation to enable wideband acoustic metasurface. Extensive experiments show that our system can achieve an average signal strength improvement of 7.5 dB and 10.5 dB in LoS and NLoS scenarios respectively with the help of a metasurface with a size of 17.6 × 17.6 cm. Two representative sensing applications (i.e., respiration sensing and gesture recognition) and one communication case study are employed to show the effectiveness of the metasurface.

Wi-Fi signal-based person identification has become a hot research topic due to the widespread deployment of Wi-Fi devices and the fact that these approaches are noncontact, passive, and privacy-preserving. While the existing related methods and systems have achieved good performance for person identification, they also encounter many significant challenges in practical applications. Due to the propagation properties of Wi-Fi signals, the signal at the receiver will change significantly when the user’s appearance changes. This makes single-appearance trained models unusable for cross-appearance recognition tasks. To address this challenge, we propose a deep learning-based framework for appearance-independent identification using Wi-Fi signals (WiAi-ID), the core of which lies in the fact that the domain discriminator and feature extractor are trained together in an adversarial manner, thus forcing the model to extract identity-inherent features independent of human appearance, and introduces a multiscale CNN adaptation module to capture time-span-based features. We collected Wi-Fi signal data of pedestrians with different appearances. The experimental results show that WiAi-ID can effectively eliminate the impact on identification due to pedestrian appearance variations and accordingly outperforms the current state-of-the-art video and wireless signal-based recognition methods.

Federated learning (FL) is an emerging distributed and privacy-preserving machine learning framework. However, the performance of traditional FL methods is seriously impaired by the real-world data, which appear to be non-IID. The recent clustered federated learning (CFL) methods eliminate the impact of non-IID data by grouping clients with similar data distribution into the same cluster. Unfortunately, existing CFL methods heavily rely on the pre-setting of the cluster number, failing to achieve adaptive client clustering. We also experimentally observe that imbalanced data largely degrade their correctness of client clustering. In this paper, we present a novel CFL method without manual intervention, named AutoCFL, which can eliminate both effects of non-IID and imbalanced data simultaneously. To deal with imbalanced data, the local training adjustment strategy adaptively adjusts the number of local training epochs for each client. To further improve the clustering correctness and adaptability, the weighted voting-based client clustering strategy automatically groups each client into an appropriate cluster. Extensive experiments are conducted to evaluate the design of AutoCFL with three popular datasets under various data settings. Experimental results demonstrate that AutoCFL outperforms state-of-the-art methods, e.g., on average improving model accuracy by 9.24%, while reducing communication costs by 4.67 in an adaptive manner.

Wi-Fi CSI-based gait recognition is a non-intrusive passive biometric identification technology that has garnered significant attention in the fields of security and smart furniture due to its user-friendly nature. However, in practical application scenarios, gait recognition systems face the challenge of reliably identifying subjects across different scenes or states. To overcome this challenge, this paper proposes DCS-Gait, a domain adaptation solution for cross-scene and cross-state gait recognition based on Wi-Fi CSI. DCS-Gait leverages a novel data distribution measurement called Cross-Attention Metric to align the class-level data distribution differences, enabling the model to learn invariant features across scenes and states. To address the issue of data annotation, we employ a pre-training method to obtain pseudo labels for the dataset. Additionally, a combined matching filtering technique is utilized to generate high-quality pseudo labels for unrecognized data, which can be further employed for supervised model training. We evaluated the effectiveness of DCS-Gait on a large test set consisting of 34 subjects, 2 scenes, and 3 different states, and the results demonstrate significant improvements over the state-of-the-art baselines in both cross-scene and cross-state gait recognition tasks. DCS-Gait provides a promising and reliable solution for accurate cross-scene and cross-state gait recognition in real-world settings.

Clustered federated learning (CFL) has proved to be an effective way to alleviate the non-IID (not independently and identically distributed) data challenge, which severely restricts the wider application of federated learning. However, existing approaches either lack adaptability,i.e., they require an additional number of clusters as a guide when clustering, or lack effectiveness in terms of communication. In this paper, we explore the differences in the ability of different layers in a model to represent non-IID data, and propose a hierarchical CFL approach, namedHiCFL, which considers both adaptivity and communication efficiency. The improvement of communication efficiency is due to our proposed novel concept of model stability, which characterizes the variation of model weights during training. Based on model stability,HiCFLcan find the proper time to bi-partition the clusters of mobile devices in a hierarchical manner more quickly. We conduct extensive experiments based on popular datasets with various non-IID data settings. The results show thatHiCFLachieves excellent performance effectiveness and efficiency. Compared to state-of-the-art approaches,HiCFLcan improve the model accuracy by$2.0\% \sim 9.0\%$, while reducing the communication overheads by$27.3\% \sim 80.6\%$.

The quality of a learning-based Wi-Fi sensing system is bounded by the quantity and quality of training data. However, obtaining sufficient and high-quality data across different domains is difficult due to extensive user involvement. We present CARING, a federated-learning-based framework to support collaborative and cross-domain Wi-Fi sensing. A key challenge of CARING is to allow the effective exchange and learning of knowledge across local models that are derived from heterogeneous data sources with uneven data distributions. We overcome this challenge by first extracting the activity-related representation to train local models. The shared global model aggregates received local model parameters and sends them back to individual devices for fine-tuning locally in the deployed environment. By leveraging the crowdsourced knowledge, CARING allows local models to quickly adapt to domain changes using just a few samples seen at test time. We demonstrate the benefit of CARING by applying it to activity recognition across three public datasets collected from 5 environments, 7 deployments, 31 users, and 29 activities. Experimental results show that CARING is highly effective and robust, improving the alternative approach for using single-sourced training data by up to 47%, giving an accuracy of over 80% (up to 100%) for various cross-domain scenarios.

As the most natural and convenient way to communicate with people, speech is always preferred in Human-Computer Interactions. However, voice-based interaction still has several limitations. It raises privacy concerns in some circumstances and the accuracy severely degrades in noisy environments. To address these limitations, silent speech recognition (SSR) has been proposed, which leverages the inaudible information (e.g., lip movements and throat vibration) to recognize the speech. In this paper, we present EarSSR, an earphone-based silent speech recognition system to enable interaction with human and device without a need for vocalization. The key insight is that when people are speaking, their ear canals exhibit unique deformation patterns and the corresponding deformation patterns are related to words/letters even without any vocalization. We utilize the built-in microphone and speaker of an earphone to capture the ear canal deformation. Ultrasound signals are emitted and the reflected signals are analyzed to extract the signal features corresponding to speech-induced ear canal deformation for silent speech recognition. We design a two-channel hierarchical convolutional neural network to achieve fine-grained letter/word recognition. Our extensive experiments show that EarSSR can achieve an accuracy of 82% for single alphabetic letter recognition and an accuracy of 93% for word recognition.

Backscatter tag based sensing has received a lot of attention recently due to the battery-free, low-cost, and widespread use of backscatter tags, e.g., RFIDs. Despite that, they suffer from an extensive, costly, and time-consuming redesign effort when there are changes in application requirements, such as changes in sensing targets or working frequency bands. This paper introduces a reconfigurable sensing framework, which enables us to easily reconfigure the design parameters of chipless backscatter tags for sensing different targets or working with different frequency bands, without the need for onerous design effort. To realize this vision, we capture the relationship between the application requirements and the sensing tag’s design parameters. This relationship enables us to fast and efficiently reconfigure/change an existing sensing tag design to meet new application requirements. Real-world experiments show that, by using our reconfigurable framework to flexibly redesign a tag’s parameters, the sensing tag achieves more than 92.1% accuracy for sensing four different applications and working on four different frequency bands.

This paper presents the design and implementation of Fusang, a low-barrier system that brings accurate and robust 3D object recognition to Commercial-Off-The-Shelf mmWave devices. The basic idea of Fusang is leveraging the large bandwidth of mmWave Radars to capture a unique set of fine-grained reflected responses generated by object shapes. Moreover, Fusang constructs two novel graph-structured features to robustly represent the reflected responses of the signal in the frequency domain and IQ domain, and carefully designs a neural network to accurately recognize objects even in different multipath scenarios. We have implemented a prototype of Fusang on a commodity mmWave Radar device. Our experiments with 24 different objects show that Fusang achieves a mean accuracy of 97% in different multipath environments. The code, dataset, and trained models of Fusang can be obtained at https://github.com/OpenNISLab/Pro-Fusang.

Sweat contains a wealth of physiologically relevant information and has been used to detect underlying diseases or the sub-health state. However, existing sweat sensors suffer from high energy consumption due to the need for energy-hungry components (i.e., ADC and DAC) and active radio front-ends, making them unable to support continuous and long-term monitoring.

Scene graph is a structured representation of a scene that can clearly express the objects, attributes, and relationships between objects in the scene. As computer vision technology continues to develop, people are no longer satisfied with simply detecting and recognizing objects in images; instead, people look forward to a higher level of understanding and reasoning about visual scenes. For example, given an image, we want to not only detect and recognize objects in the image, but also understand the relationship between objects (visual relationship detection), and generate a text description (image captioning) based on the image content. Alternatively, we might want the machine to tell us what the little girl in the image is doing (Visual Question Answering (VQA)), or even remove the dog from the image and find similar images (image editing and retrieval), etc. These tasks require a higher level of understanding and reasoning for image vision tasks. The scene graph is just such a powerful tool for scene understanding. Therefore, scene graphs have attracted the attention of a large number of researchers, and related research is often cross-modal, complex, and rapidly developing. However, no relatively systematic survey of scene graphs exists at present. To this end, this survey conducts a comprehensive investigation of the current scene graph research. More specifically, we first summarize the general definition of the scene graph, then conducte a comprehensive and systematic discussion on the generation method of the scene graph (SGG) and the SGG with the aid of prior knowledge. We then investigate the main applications of scene graphs and summarize the most commonly used datasets. Finally, we provide some insights into the future development of scene graphs.

Object detection (OD) is a crucial computer vision task that has seen the development of many algorithms and models over the years. While the performance of current OD models has improved, they have also become more complex, making them impractical for industry applications due to their large parameter size. To tackle this problem, knowledge distillation (KD) technology was proposed in 2015 for image classification and subsequently extended to other visual tasks due to its ability to transfer knowledge learned by complex teacher models to lightweight student models. This paper presents a comprehensive survey of KD-based OD models developed in recent years, with the aim of providing researchers with an overview of recent progress in the field. We conduct an in-depth analysis of existing works, highlighting their advantages and limitations, and explore future research directions to inspire the design of models for related tasks. We summarize the basic principles of designing KD-based OD models, describe related KD-based OD tasks, including performance improvements for lightweight models, catastrophic forgetting in incremental OD, small object detection, and weakly/semi-supervised OD. We also analyze novel distillation techniques, i.e. different types of distillation loss, feature interaction between teacher and student models, etc. Additionally, we provide an overview of the extended applications of KD-based OD models on specific datasets, such as remote sensing images and 3D point cloud datasets. We compare and analyze the performance of different models on several common datasets and discuss promising directions for solving specific OD problems.

Contactless wireless sensing without attaching a device to the target has achieved promising progress in recent years. However, one severe limitation is the small sensing range. This paper presents Widesee to realize wide-area sensing with only one transceiver pair. Widesee utilizes the LoRa signal to achieve a larger range of sensing and further incorporates drone’s mobility to broaden the sensing area. Widesee presents solutions across software and hardware to overcome two aspects of challenges for wide-range contactless sensing: (i) the interference brought by device mobility and LoRa’s high sensitivity; and (ii) the ambiguous target information such as location when employing just a single pair of transceivers for sensing. We have developed a working prototype of Widesee for human target detection and localization that are especially useful in emergency scenarios such as rescue search, and evaluated Widesee with both controlled experiments and the field study in a high-rise building. Extensive experiments demonstrate the great potential of Widesee for wide-area contactless sensing with a single LoRa transceiver pair hosted on a drone.

Liquid identification plays an essential role in our daily lives. However, existing RF sensing approaches still require dedicated hardware such as RFID readers and UWB transceivers, which are not readily available to most users. In this article, we propose Akte-Liquid, which leverages the speaker on smartphones to transmit acoustic signals, and the microphone on smartphones to receive reflected signals to identify liquid types and analyze the liquid concentration. Our work arises from the acoustic intrinsic impedance property of liquids, in that different liquids have different intrinsic impedance, causing reflected acoustic signals of liquids to differ. Then, we discover that the amplitude-frequency feature of reflected signals may be utilized to represent the liquid feature. With this insight, we propose new mechanisms to eliminate the interference caused by hardware and multi-path propagation effects to extract the liquid features. In addition, we design a new Siamese network-based structure with a specific training sample selection mechanism to reconstruct the extracted feature to container-irrelevant features. Our experimental evaluations demonstrate that Akte-Liquid is able to distinguish 20 types of liquids at a higher accuracy, and to identify food additives and measure protein concentration in the artificial urine with a 92.3% accuracy under 1 mg/100 mL as well.

Gesture detection based on radio frequency signals has gained increasing popularity in recent years due to several benefits it has brought, such as eliminating the need to carry additional devices and providing better privacy. In traditional methods, significant breakthroughs have been made to improve recognition accuracy and scene robustness, but the limited computing power of edge devices (the first-level equipment to receive signals) and the requirement of fast response for detection have not been adequately addressed. In this article, we propose a lightweight Wi-Fi gesture recognition system, referred to as WiFine, which is designed and implemented for deployment on low-end edge devices without the use of any additional high-performance services in the process. Toward these goals, we first design algorithms for phase difference selection and amplitude enhancement, respectively, to tackle the problem of data drift caused by user change. Then, we design a cross-dimension fusion method to extract features of finer granularity from information of different dimensions, thus solving the precision problem of feature granularity. Finally, we design a lightweight neural network architecture by leveraging redundancy to reduce computational cost while ensuring satisfactory recognition accuracy. Extensive experimental results show that the proposed system achieves fast recognition of various actions with an accuracy up to 96.03% in 0.19 seconds.

Backscatter tag based sensing has received a lot of attention recently due to the battery-free, low-cost and widespread use of backscatter tags, e.g., RFIDs. Despite that, they suffer from an ex-tensive, costly, and time-consuming redesign effort when there are changes in application requirements, such as changes in sensing targets or working frequency bands. This paper introduces a reconfigurable sensing framework, which enables us to easily reconfigure the design parameters of chipless backscatter tags for sensing different targets or working with differ-ent frequency bands, without the need of onerous design effort. To realize this vision, we capture the relationship between the application requirements and the sensing tag's design parameters. This relationship enables us to fast and efficiently reconfigure/change an existing sensing tag design for meeting new application requirements. Real-world experiments show that, by using our reconfig-urable framework to flexibly redesign a tag's parameters, the sensing tag achieves more than 92.1 % accuracy for sensing four different applications and working on four different frequency bands.

Phased array beamforming has been extensively explored as a physical layer primitive to improve the secrecy capacity of wireless communication links. However, existing solutions are incompatible with low-profile IoT devices due to cost, power and form factor constraints. More importantly, they are vulnerable to eavesdroppers with a high-sensitivity receiver. This paper presents Protego, which offloads the security protection to a metasurface comprised of a large number of 1-bit programmable unit-cells (i.e., phase shifters). Protego builds on a novel observation that, due to phase quantization effect, not all the unit-cells contribute equally to beamforming. By judiciously flipping the phase shift of certain unit-cells, Protego can generate artificial phase noise to obfuscate the signals towards potential eavesdroppers, while preserving the signal integrity and beamforming gain towards the legitimate receiver. A hardware prototype along with extensive experiments has validated the feasibility and effectiveness of Protego.

As the most important organs of sense, human eyes perceive 80% information from our surroundings. Eyeball movement is closely related to our brain health condition. Eyeball movement and eye blink are also widely used as an efficient human-computer interaction scheme for paralyzed individuals to communicate with others. Traditional methods mainly use intrusive EOG sensors or cameras to capture eye activity information. In this work, we propose a system named SmartLens to achieve eye activity sensing using zero-power contact lens. To make it happen, we develop dedicated antenna design which can be fitted in an extremely small space and still work efficiently to reach a working distance more than 1 m. To accurately track eye movements in the presence of strong self-interference, we employ another tag to track the user's head movement and cancel it out to support sensing a walking or moving user. Comprehensive experiments demonstrate the effectiveness of the proposed system. At a distance of 1.4 m, the proposed system can achieve an average accuracy of detecting the basic eye movement and blink at 89.63% and 82%, respectively.

Long-range backscatter communication has the potential to provide enough time and space for vehicles to detect traffic information, which is an attractive solution for Vehicle-to-Infrastructure (V2I) communication. However, existing backscatter studies either require the tag to be close to the carrier source (ambient backscatter) or have poor receiver sensitivity (RFID), making it challenging to satisfy the high range requirements of V2I communication. In this article, we develop AllSpark to investigate the feasibility of long-range communication enabling backscatter to be applied to V2I. Specifically, to increase RSS to compensate for the enormous dual-path loss in backscatter systems, we first redesign the tag’s radio frequency (RF) front-end to amplify the incident signals, and then we design high-gain directional antennas for the tag and reader. Second, we present EC-Net, an end-to-end demodulator that selectively enhances signal features (denoising) and is tuned to lower classification (demodulation) error to maximize demodulation accuracy. Furthermore, we adopt convolutional encoding for tag data and use the output probability of EC-Net to design an effective soft decoder to resist occasional interference and noise, improving communication robustness. Our prototype and experiments outdoors verify the effectiveness of AllSpark which can provide a communication range of 700 m. Even in a moving scene, it can achieve a communication range of 600 m, demonstrating that AllSpark has the potential to be applied for autonomous driving and low-flying drones to detect traffic conditions.

With the rapid development of the Internet of Things (IoT), we have seen a larger number of devices deployed with different wireless communication protocols (i.e., WiFi, ZigBee, Bluetooth). Working in the same place opens a new opportunity for these devices to communicate directly with each other, leveraging on Cross-technology Communication (CTC). However, since these devices operate in the same frequency band which results in the competition against each other for network resources, severe interfere may arise. In this paper, we explore pseudo-random sequence (PR sequence) to design a novel CTC protocol that enables low-cost direct communication between WiFi and ZigBee in noisy indoor environments. Pseudorandom sequence offers a unique statistical feature to accomplish both information transmission and synchronization between heterogeneous devices. We design a dynamic synchronous decoding strategy to handle interference coexisted among different wireless protocols. Our system does not require any modification of communication protocol and underlying hardware and firmware. We implement our system on commercial devices (Intel 5300 WiFi NIC and MicaZ CC2420), and conduct extensive experiments to evaluate the system performance in three typical scenarios. The experimental results show that the synchronization time of our approach is lower than 0.5 ms, and the accuracy is greater than 84% while the channel occupancy is as high as 50%.

Beamforming can improve the communication and sensing capabilities for a wide range of IoT applications. However, most existing IoT devices cannot perform beamforming due to form factor, energy, and cost constraints. This paper presents RFlens, a reconfigurable metasurface that empowers low-profile IoT devices with beamforming capabilities. The metasurface consists of many unit-cells, each acting as a phase shifter for signals going through it. By encoding the phase shifting values, RFlens can manipulate electromagnetic waves to "reshape" and resteer the beam pattern. We prototype RFlens for 5 GHz Wi-Fi signals. Extensive experiments demonstrate that RFlens can achieve a 4.6 dB median signal strength improvement (up to 9.3 dB) even with a relatively small 16 × 16 array of unit-cells. In addition, RFlens can effectively improve the secrecy capacity of IoT links and enable passive NLoS wireless sensing applications.

Wireless sensing builds upon machine learning shows encouraging results. However, adopting wireless sensing as a large-scale solution remains challenging as experiences from deployments have shown the performance of a machine-learned model to suffer when there are changes in the environment, e.g., when furniture is moved or when other objects are added or removed from the environment. We present Rise, a novel solution for enhancing the robustness and performance of learning-based wireless sensing techniques against such changes during a deployment. Rise combines probability and statistical assessments together with anomaly detection to identify samples that are likely to be misclassified and uses feedback on these samples to update a deployed wireless sensing model. We validate Rise through extensive empirical benchmarks by considering 11 representative sensing methods covering a broad range of wireless sensing tasks. Our results show that Rise can identify 92.3% of misclassifications on average. We showcase how Rise can be combined with incremental learning to help wireless sensing models retain their performance against dynamic changes in the operating environment to reduce the maintenance cost, paving the way for learning-based wireless sensing to become capable of supporting long-term monitoring in complex everyday environments.

Long range (LoRa) is an attractive low-power wide-area networks (LPWANs) technology for its features of low power, long range, and support for concurrent transmission. Our study reveals LoRa concurrent transmission suffer from the mismatch between the sender’s reception (RX) and gateway’s transmission (TX) window, which leads to the decline of goodput even the throughput is improved. Our experiment shows that goodput only accounts for two-fifths of the throughput in concurrent transmissions with 48 nodes at a duty cycle of 20%. This article presents a window match scheme named Cantor which improves the goodput of LoRa concurrent transmission by controlling the RX window size. Cantor does not require the frequent exchange of controlling information. Instead, it introduces a novel concurrent transmission model to estimate the downlink packet reception rate (PRR) with different network parameters, and a regression model is used to make the result more realistic. Then, we propose a simple optimization algorithm to select optimal RX window sizes in which nodes are able to receive acknowledgments. We implement and evaluate Cantor with commodity LoRa gateway and nodes, and conduct experiments in different scenarios. The experimental results show that Cantor increases the goodput by 70% and reduces energy consumption by 30% in LoRa concurrent transmissions with 48 nodes operate at a duty cycle of 20%.

Operating in unlicensed ISM bands, ZigBee devices often yield poor performance due to the interference from ever increasing wireless devices in the 2.4 GHz band. Our empirical results show that, a specific interference is likely to have different influence on different outbound links of a ZigBee sender, which indicates the chance of concurrent transmissions. Based on this insight, we propose Smoggy-Link, a practical protocol to exploit the potential concurrency for adaptive ZigBee transmissions under harsh interference. Smoggy-Link maintains an accurate link model to quantify and trace the relationship between interference and link qualities of the sender's outbound links. With such a link model, Smoggy-Link can translate low-cost interference information to the fine-grained spatiotemporal link state. The link information is further utilized for adaptive link selection and intelligent transmission schedule. We implement and evaluate a prototype of our approach with TinyOS and TelosB motes. The evaluation results show that Smoggy-Link has consistent improvements in both throughput and packet reception ratio under interference from various interferers.

Material identification and target imaging play an important role in many applications. This paper introduces TagScan, a system that can identify the material type and image the horizontal cut of a target simultaneously with cheap commodity Radio-Frequency IDentification (RFID) devices. The key intuition is that different materials and/or target sizes cause different amounts of phase and RSS (Received Signal Strength) changes, when radio frequency (RF) signal penetrates through the target. Multiple challenges need to be addressed before we can turn the idea into a functional system, including (i) indoor environments exhibit rich multipath which breaks the linear relationship between the phase change and the propagation distance inside a target; (ii) without knowing either material type or target size, trying to obtain these two information simultaneously is challenging; and (iii) stitching pieces of the propagation distances inside a target for an image estimate is non-trivial. We propose solutions to all the challenges and evaluate the system's performance in three different environments. TagScan is able to achieve higher than 94 percent material identification accuracies for 10 liquids and differentiates even very similar objects such as Coke and Pepsi. TagScan can accurately estimate the horizontal cut images of more than one target behind a wall.

Parallel backscatter is a promising technique for high throughput, low power communications. The existing approaches of parallel backscatter are based on a common assumption, i.e. the states of the collided signals are distinguishable from each other in either the time domain or the IQ (the In-phase and Quadrature) domain. We in this paper disclose the superclustering phenomenon, which invalidates that assumption and seriously affects the decoding performance. Then we propose an interstellar travelling model to capture the bursty Gaussian process of a collided signal. Based on this model, we design Hubble, a reliable signal processing approach to support parallel backscatter in the wild. Hubble addresses several technical challenges: (i) a novel scheme based on Pearson's Chi-Square test to extract the collided signals' combined states, (ii) a Markov driven method to capture the law of signal state transitions, and (iii) error correction schemes to guarantee the reliability of parallel decoding. Theoretically, Hubble is able to decode all the backscattered data, as long as the signals are detectable by the receiver. The experiment results demonstrate that the median throughput of Hubble is 11.7× higher than that of the state-of-the-art approach.

Many real-world applications involve data collected from different views and with high data dimensionality. Furthermore, multi-view data always has unavoidable noise. Clustering on this kind of high-dimensional and noisy multi-view data remains a challenge due to the curse of dimensionality and ineffective de-noising and integration of multiple views. Aiming at this problem, in this paper, we propose a Robust Self-weighted Multi-view Projection Clustering (RSwMPC) based on ℓ2,1-norm, which can simultaneously reduce dimensionality, suppress noise and learn local structure graph. Then the obtained optimal graph can be directly used for clustering while no further processing is required. In addition, a new method is introduced to automatically learn the optimal weight of each view with no need to generate additional parameters to adjust the weight. Extensive experimental results on different synthetic datasets and real-world datasets demonstrate that the proposed algorithm outperforms other state-of-the-art methods on clustering performance and robustness.

Modern distributed applications are moving toward a microservice architecture, in which each service is developed and managed independently, and new features and updates are delivered continuously. A guiding principle of microservice architecture is that it is vital to anticipate and mitigate a variety of hardware and software failures. In order to test the fault handling capabilities of microservices automatically, this paper presents IntelliFT, a guided resilience testing technique for microservice based applications, which aims to expose the defects in the fault-handling logic effectively within a fixed time limit. The characteristic of IntelliFT is that it leverages existing integration tests of the applications under test to explore the fault space, and decides whether injected faults can lead to severe failures by designing fitness-guided search technique. Our experimental results on a medium-size microservice benchmark system show that the proposed technique is effective, improves the state-of-the-art, and can quickly expose bugs in the recovery logic.

Machine learning models, particularly reinforcement learning (RL), have demonstrated great potential in optimizing video streaming applications. However, the state-of-the-art solutions are limited to an "offline learning" paradigm, i.e., the RL models are trained in simulators and then are operated in real networks. As a result, they inevitably suffer from the simulation-to-reality gap, showing far less satisfactory performance under real conditions compared with simulated environment. In this work, we close the gap by proposing OnRL, an online RL framework for real-time mobile video telephony. OnRL puts many individual RL agents directly into the video telephony system, which make video bitrate decisions in real-time and evolve their models over time. OnRL then aggregates these agents to form a high-level RL model that can help each individual to react to unseen network conditions. Moreover, OnRL incorporates novel mechanisms to handle the adverse impacts of inherent video traffic dynamics, and to eliminate risks of quality degradation caused by the RL model's exploration attempts. We implement OnRL on a mainstream operational video telephony system, Alibaba Taobao-live. In a month-long evaluation with 543 hours of video sessions from 151 real-world mobile users, OnRL outperforms the prior algorithms significantly, reducing video stalling rate by 14.22% while maintaining similar video quality.

Monitoring the vibration characteristics of a machine or structure provides valuable information of its health condition and this information can be used to detect problems in their incipient stage. Recently, researchers employ RFID signals for vibration sensing. However, they mainly focus on vibration frequency estimation and still face difficulties in accurately sensing the other important characteristic of vibration which is vibration amplitude in the scale of sub-millimeter. In this paper, we introduce TagSMM, a contactless RFID-based vibration sensing system which can measure vibration amplitude in sub-millimeter resolution. TagSMM employs the signal propagation theory to deeply understand how the signal phase varies with vibration and proposes a coupling-based method to amplify the vibration-induced phase change to achieve sub-millimeter level amplitude sensing for the first time. We design and implement TagSMM with commodity RFID hardware. Our experiments show that TagSMM can detect a 0.5 mm vibration, 10 times better than the state-of-the-arts. Our field studies show TagSMM can sense a drone's abnormal vibration and can also effectively detect a small 0.2 cm screw loose in a motor at a 100% accuracy.

Record-and-replay tools are important for quality assurance of Web applications by capturing user case scenarios and executing them automatically when needed. However, the tests generated by existing techniques are brittle, and often lead to test breakages as the dynamic behavior and frequent updates of modern Web applications. In this paper, we propose WebRR, a self-replay enhanced robust record-and-replay technique for Web applications testing. The novelty of WebRR is that, it introduces a new self-replay mechanism in the recording phase, which checks the captured event from the record module online, and generates multiple locators (including DOM locators, visual locator and proximity locators) automatically, to improve the robustness of generated test cases. During the replay, it combines multiple locators and new local workflow repair technique to repair test breakages, and can improve the resilience of generated tests to frequent updates of the applications. We applied our approach to 3 enterprise Web applications, which are deployed in a large power grid company of China. The experimental results show that WebRR is effective, and substantially improve the robustness of end-to-end web tests that are generated using record-and-replay technique.

In best-effort packet networks, a buffer is commonly used to eliminate network jitter and enable smooth video playback at the expense of additional delay (i.e. jitterdelay). In this work, we examine how jitter buffer performs in the Web Real-Time Communications (WebRTC), which is the de-facto standard used in interactive multimedia applications. We collect a dataset from a live video streaming service provider, which adopts WebRTC. After an in-depth analysis of the dataset, we find that jitter buffer can dynamically adjust the jitterdelay but is too conservative, resulting in a very slow decline of jitterdelay. To address the issue, we analyze the control logic of the jitter buffer and find that the reason lies in the use of a fixed reduction factor, known as psi (ψ). We propose an enhanced jitter buffer adaptation mechanism called JTB-ψ, which dynamically adjusts ψ according to frame size and frame duration, to reasonably speed up the decline of jitterdelay. Practical testbed experiments show that JTB-ψ achieves a 41.5% lower jitterdelay and improves receiving frame rate, quantization parameter (QP) and sending bit-rate under different network conditions, compared to the fixed-ψ approach.

Sitting posture has a close relationship with our health, and keeping a healthy sitting posture is critical to each of us. Poor sitting postures often inevitably increase the risk of modern health musculoskeletal disorders. Previous works either used a camera to record the image or attached wearable sensors on the human body to recognize sitting postures. However, video-based approaches may face privacy issues while the wearable sensor-based approaches may cause uncomfortable to the user. In this work, we propose SitR, the first sitting posture recognition system using RF signals alone, which neither compromises the privacy nor requires wearing various sensors on the human body. We demonstrate that SitR can successfully recognize seven habitual sitting postures with just three lightweight and low-cost RFID tags pasted to the user's back. Our design exploits the correlation between the phase change of RFID tags and the sitting postures. By extracting effective features of the measured phase sequences and employing appropriate machine learning algorithm, SitR can achieve robust and high performance. We evaluate the performance of SitR with ten volunteers in two different scenarios. Extensive experiments show SitR can recognize seven sitting postures with a high accuracy across different scenarios and various conditions. SitR can further detect the abnormal respiration, stand up, and sit down and provide sitting posture history for sedentary people.

Backscatter communications have two major antenna configurations: the bistatic configuration, in which the reader employs two different sets of antennas to transmit and receive, and the monostatic configuration, in which the reader employs one set of antennas for both transmitting and receiving. In this paper, we provide a comprehensive study on the MIMO techniques for the M × L monostatic channel. Particularly we study on the joint design of the query and the coding matrices. We show that the maximum achievable diversity order of the monostatic channel is the diversity order achieved by the block-lever unitary query and orthogonal space-time block code (BUTQ-OSTBC) design pair, which is ML 2 , exactly the half of the diversity order of the conventional MIMO channel. Then we show that uniform query, the simplest query approach, cannot achieve the maximum achievable diversity order in the monostatic channel. We generalize BUTQ-OSTBC to the general augmenting approach, and show that unitary matrix is optimal in terms of SER performance among all possible query matrices, when OSTBC is employed. The above results further indicate that in the backscatter channel, additional diversity can be obtained by varying the query signals over time slots within the channel coherent time, which is quite different from the results from the conventional MIMO channels. We verify our results by Monte Carlo simulations.

Text-based CAPTCHAs remains a popular scheme for distinguishing between a legitimate human user and an automated program. This article presents a novel genetic text captcha solver based on the generative adversarial network. As a departure from prior text captcha solvers that require a labor-intensive and time-consuming process to construct, our scheme needs significantly fewer real captchas but yields better performance in solving captchas. Our approach works by first learning a synthesizer to automatically generate synthetic captchas to construct a base solver. It then improves and fine-tunes the base solver using a small number of labeled real captchas. As a result, our attack requires only a small set of manually labeled captchas, which reduces the cost of launching an attack on a captcha scheme. We evaluate our scheme by applying it to 33 captcha schemes, of which 11 are currently used by 32 of the top-50 popular websites. Experimental results demonstrate that our scheme significantly outperforms four prior captcha solvers and can solve captcha schemes where others fail. As a countermeasure, we propose to add imperceptible perturbations onto a captcha image. We demonstrate that our countermeasure can greatly reduce the success rate of the attack.

In the real world, the duality of high-dimensional data is widespread. The coclustering method has been widely used because they can exploit the co-occurring structure between samples and features. In fact, most of the existing coclustering methods cluster the graphs in the original data matrix. However, these methods fail to output an affinity graph with an explicit cluster structure and still call for the postprocessing step to obtain the final clustering results. In addition, these methods are difficult to find a good projection direction to complete the clustering task on high-dimensional data. In this article, we modify the flexible manifold embedding theory and embed it into the bipartite spectral graph partition. Then, we propose a new method called structured optimal graph-based clustering with flexible embedding (SOGFE). The SOGFE method can learn an affinity graph with an optimal and explicit clustering structure and does not require any postprocessing step. Additionally, the SOGFE method can learn a suitable projection direction to map high-dimensional data to a low-dimensional subspace. We perform extensive experiments on two synthetic data sets and seven benchmark data sets. The experimental results verify the superiority, robustness, and good projection direction selection ability of our proposed method.

Drowsiness detection is critical to driver safety, considering thousands of deaths caused by drowsy driving annually. Professional equipment is capable of providing high detection accuracy, but the high cost limits their applications in practice. The use of mobile devices such as smart watches and smart phones holds the promise of providing a more convenient, practical, non-invasive method for drowsiness detection. In this article, we propose a real-time driver drowsiness detection system based on mobile devices, referred to as dWatch, which combines physiological measurements with motion states of a driver to achieve high detection accuracy and low power consumption. Specifically, based on heart rate measurements, we design different methods for calculating heart rate variability (HRV) and sensing yawn actions, respectively, which are combined with steering wheel motion features extracted from motion sensors for drowsiness detection. We also design a driving posture detection algorithm to control the operation of the heart rate sensor to reduce system power consumption. Extensive experimental results show that the proposed system achieves a detection accuracy up to 97.1% and reduces energy consumption by 33%.

Google congestion control (GCC) is the de-facto standard for web real-time communications (WebRTC) applications and has been implemented in mainstream browsers including Chrome and Firefox. While GCC is designed to achieve high video bit-rate and low latency simultaneously, we find that GCC's performance is far from satisfactory particularly under good network conditions. In particular, we collect a GCC trace dataset with over 1.18 million sessions from a major crowd-sourced live video streaming service provider. We perform in-depth analytics using the dataset, which shows that the sending video bit-rate unnecessarily experiences frequent rollbacks caused by minor fluctuation of transmission delay. To address the issue, we propose a mechanism called GCC-β, which can distinguish normal network fluctuation from real network congestion, and then adaptively sends appropriate bitrates. We implement GCC-β in the WebRTC framework and evaluate its performance using test-bed experiments. The results show that GCC-β is able to avoid up to 90% unnecessary bitrate rollbacks.

Target material identification is playing an important role in our everyday life. Traditional camera and video-based methods bring in severe privacy concerns. In the last few years, while RF signals have been exploited for indoor localization, gesture recognition and motion tracking, very little attention has been paid in material identification. This paper introduces WiMi, a device-free target material identification system, implemented on ubiquitous and cheap commercial off-the-shelf (COTS) Wi-Fi devices. The intuition is that different materials produce different amounts of phase and amplitude changes when a target appears on the line-of-sight (LoS) of a radio frequency (RF) link. However, due to multipath and hardware imperfection, the measured phase and amplitude of the channel state information (CSI) are very noisy. We thus present novel CSI pre-processing schemes to address the multipath and hardware noise issues before they can be used for accurate material sensing. We also design a new material feature which is only related to the material type and is independent of the target size. Comprehensive real-life experiments demonstrate that WiMi can achieve fine-grained material identification with cheap commodity Wi-Fi devices. WiMi can identify 10 commonly seen liquids at an overall accuracy higher than 95% with strong multipath indoors. Even for very similar items such as Pepsi and Coke, WiMi can still differentiate them at a high accuracy.

Despite the pervasive use of real-time video telephony services, the users' quality of experience (QoE) remains unsatisfactory, especially over the mobile Internet. Previous work studied the problem via controlled experiments, while a systematic and in-depth investigation in the wild is still missing. To bridge the gap, we conduct a large-scale measurement campaign on \appname, an operational mobile video telephony service. Our measurement logs fine-grained performance metrics over 1 million video call sessions. Our analysis shows that the application-layer video codec and transport-layer protocols remain highly uncoordinated, which represents one major reason for the low QoE. We thus propose \name, a machine learning based framework to resolve the issue. Instead of blindly following the transport layer's estimation of network capacity, \name reviews historical logs of both layers, and extracts high-level features of codec/network dynamics, based on which it determines the highest bitrates for forthcoming video frames without incurring congestion. To attain the ability, we train \name with the aforementioned massive data traces using a custom-designed imitation learning algorithm, which enables \name to learn from past experience. We have implemented and incorporated \name into \appname. Our experiments show that \name outperforms state-of-the-art solutions, improving video quality while reducing stalling time by multi-folds under various practical scenarios.

Despite the pervasive use of real-time video telephony services, their quality of experience (QoE) remains unsatisfactory, especially over the mobile Internet. We conduct a large-scale measurement campaign on \appname, an operational mobile video telephony service. Our analysis shows that the application-layer video codec and transport-layer protocols remain highly uncoordinated, which represents one major reason for the low QoE. We thus propose \name, a machine learning based framework to resolve the issue. We train \name with the massive data traces from the measurement campaign using a custom-designed imitation learning algorithm, which enables \name to learn from past experience following an expert's iterative demonstration/supervision. We have implemented and incorporated \name into the \appname. Our experiments show that \name outperforms state-of-the-art solutions, improving video quality while reducing stalling time by multi-folds under various practical scenarios.

Contactless wireless sensing without attaching a device to the target has achieved promising progress in recent years. However, one severe limitation is the small sensing range. This paper presents WideSee to realize wide-area sensing with only one transceiver pair. WideSee utilizes the LoRa signal to achieve a larger range of sensing and further incorporates drone's mobility to broaden the sensing area. WideSee presents solutions across software and hardware to overcome two aspects of challenges for wide-range contactless sensing: (i) the interference brought by the device mobility and LoRa's high sensitivity; and (ii) the ambiguous target information such as location when employing just a single pair of transceivers. We have developed a working prototype of WideSee for human target detection and localization that are especially useful in emergency scenarios such as rescue search, and evaluated WideSee with both controlled experiments and the field study in a high-rise building. Extensive experiments demonstrate the great potential of WideSee for wide-area contactless sensing with a single LoRa transceiver pair hosted on a drone.

Contactless wireless sensing without attaching a device to the target has achieved promising progress in recent years. However, one severe limitation in this field is the limited sensing range. This paper presents WideSee to realize wide-area sensing with only one transceiver pair. WideSee utilizes the LoRa signal to achieve a larger range of sensing and further incorporates drone's mobility to broaden the sensing area. We have developed a working prototype of WideSee for human target detection and localization that are especially useful in emergency scenarios like rescue and terrorist search. We also evaluated WideSee with field study in a high-rise building, which demonstrates the great potential of WideSee for supporting wide-area contactless sensing applications with a single LoRa transceiver pair hosted on a drone.

Material sensing is an essential ingredient for many IoT applications. While hyperspectral camera, infrared, X-Ray, and Radar provide potential solutions for material identification, high cost is the major concern limiting their applications. In this paper, we explore the capability of employing RF signals for fine-grained material sensing with commodity RFID device. The key reason for our system to work is that the tag antenna's impedance is changed when it is close or attached to a target. The amount of impedance change is dependent on the target's material type, thus enabling us to utilize the impedance-related phase change available at commodity RFID devices for material sensing. Several key challenges are addressed before we turn the idea into a functional system: (i) the random tag-reader distance causes an additional unknown phase change on top of the phase change caused by the target material; (ii) the tag rotations cause phase shifts and (iii) for conductive liquid, there exists liquid reflection which interferes with the impedance-caused phase change. We address these challenges with novel solutions. Comprehensive experiments show high identification accuracies even for very similar materials such as Pepsi and Coke.

With parallel decoding for backscatter communication, tags are allowed to transmit concurrently and more efficiently. Existing parallel decoding mechanisms, however, assume that signals of the tags are highly stable and, hence, may not perform optimally in the naturally dynamic backscatter systems. This paper introduces FlipTracer, a practical system that achieves highly reliable parallel decoding even in hostile channel conditions. FlipTracer is designed with a key insight; although the collided signal is time-varying and irregular, transitions between signals' combined states follow highly stable probabilities, which offers important clues for identifying the collided signals and provides us with an opportunity to decode the collided signals without relying on stable signals. Motivated by this observation, we propose a graphical model, called one-flip-graph (OFG), to capture the transition pattern of collided signals, and design a reliable approach to construct the OFG in a manner robust to the diversity in backscatter systems. Then, FlipTracer can resolve the collided signals by tracking the OFG. We have implemented FlipTracer and evaluated its performance with extensive experiments across a wide variety of scenarios. Our experimental results have shown that FlipTracer achieves a maximum aggregated throughput that approaches 2 Mb/s, which is 6× higher than the state of the art.

Mobile sensing is a promising sensing paradigm in the era of Internet of Things (IoT) that utilizes mobile device sensors to collect sensory data about sensing targets and further applies learning techniques to recognize the sensed targets to correct classes or categories. Due to the recent great success of deep learning, an emerging trend is to adopt deep learning in this recognition process, while we find an overlooked yet crucial issue to be solved in this paper - The size of deep learning models should be sufficiently large for reliably classifying various types of recognition targets, while the achieved processing delay may fail to satisfy the stringent latency requirement from applications. If we blindly shrink the deep learning model for acceleration, the performance cannot be guaranteed. To cope with this challenge, this paper presents a compact deep neural network architecture, namely cDeepArch. The key idea of the cDeepArch design is to decompose the entire recognition task into two lightweight sub-problems: context recognition and the context-oriented target recognitions. This decomposition essentially utilizes the adequate storage to trade for the CPU and memory resource consumptions during execution. In addition, we further formulate the execution latency for decomposed deep learning models and propose a set of enhancement techniques, so that system performance and resource consumption can be quantitatively balanced. We implement a cDeepArch prototype system and conduct extensive experiments. The result shows that cDeepArch achieves excellent recognition performance and the execution latency is also lightweight.

Wireless sensor networks (WSNs) have been used for many long-term monitoring applications with the strip topology that is ubiquitous in the real-world deployment, such as pipeline monitoring, water quality monitoring, vehicle monitoring, and Great Wall monitoring. The efficiency of routing strategy has been playing a key role in serving such monitoring applications. In this article, we first present a robust geographic opportunistic routing (GOR) approach—LIght Propagation Selection (LIPS)—that can provide a short path with low energy consumption, communication overhead, and packet loss. To overcome the complication caused by the multi-turning point structure, we propose the virtual Plane mirror (VPM) algorithm, inspired by the light propagation, which is to map the strip topology into the straight one logically. We then select partial neighbors as the candidates to avoid blindly involving all next-hop neighbors and ensure the data transmission along the correct direction. Two implementation problems of VPM—transmission spread angle and the communication range—are thoroughly analyzed based on the percolation theory. Based on the preceding candidate selection algorithms, we propose a GOR algorithm in the strip topology network. By theoretical analysis and extensive simulation, we illustrate the validity and higher transmission performance of LIPS in strip WSNs. In addition, we have proved that the length of the path in LIPS is two times the length of the shortest path via geometrical analysis. Simulation results show that the transmission success rate of our approach is 26.37% higher than the state-of-the-art approach, and the communication overhead and energy consumption rate are 33.11% and 40.23% lower, respectively.

Despite several attacks have been proposed, text-based CAPTCHAs are still being widely used as a security mechanism. One of the reasons for the pervasive use of text captchas is that many of the prior attacks are scheme-specific and require a labor-intensive and time-consuming process to construct. This means that a change in the captcha security features like a noisier background can simply invalid an earlier attack. This paper presents a generic, yet effective text captcha solver based on the generative adversarial network. Unlike prior machine-learning-based approaches that need a large volume of manually-labeled real captchas to learn an effective solver, our approach requires significantly fewer real captchas but yields much better performance. This is achieved by first learning a captcha synthesizer to automatically generate synthetic captchas to learn a base solver, and then fine-tuning the base solver on a small set of real captchas using transfer learning. We evaluate our approach by applying it to 33 captcha schemes, including 11 schemes that are currently being used by 32 of the top-50 popular websites including Microsoft, Wikipedia, eBay and Google. Our approach is the most capable attack on text captchas seen to date. It outperforms four state-of-the-art text-captcha solvers by not only delivering a significant higher accuracy on all testing schemes, but also successfully attacking schemes where others have zero chance. We show that our approach is highly efficient as it can solve a captcha within 0.05 second using a desktop GPU. We demonstrate that our attack is generally applicable because it can bypass the advanced security features employed by most modern text captcha schemes. We hope the results of our work can encourage the community to revisit the design and practical use of text captchas.

Multi-view clustering has played a vital role in real-world applications. It aims to cluster the data points into different groups by exploring complementary information of multi-view. A major challenge of this problem is how to learn the explicit cluster structure with multiple views when there is considerable noise. To solve this challenging problem, we propose a novel Robust Auto-weighted Multi-view Clustering (RAMC), which aims to learn an optimal graph with exactly k connected components, where k is the number of clusters. ℓ1-norm is employed for robustness of the proposed algorithm. We have validated this in the later experiment. The new graph learned by the proposed model approximates the original graphs of each individual view but maintains an explicit cluster structure. With this optimal graph, we can immediately achieve the clustering results without any further post-processing. We conduct extensive experiments to confirm the superiority and robustness of the proposed algorithm.

The lines in topographic maps are difficult to be separated from each other because of their confusing colors. To solve this problem, we propose a novel line separation method using their regional color and spatial information. Firstly, we divide the lines into lots of circular regions with a certain diameter, and consider these regions as the basic processing units. Then based on a new concept of regional color confusion, we classify all the divided circular regions into two kinds of regions by whether the color is pure or mixed. Further, for pure color regions, a fuzzy clustering algorithm with Gaussian kernel can be used to cluster them into different lines based on their color information. Meanwhile, we determine the memberships of the mixed color regions according to their spatial relations with the clustered pure color regions. The concept of regional color confusion is proposed to reduce the influences of the confusing colors to line separation, and the spatial relations are utilized to solve the problems of the membership determination of the mixed color regions. The experimental results demonstrate that our method can achieve higher accuracy compare with other two state-of-the-art methods, which provides a novel idea for line element segmentation from scanned topographic maps.

Chinese calligraphy is a popular, highly esteemed art form in the Chinese cultural sphere and worldwide. Ink brushes are the traditional writing tool for Chinese calligraphy and the subtle nuances of brush movements have a great impact on the aesthetics of the written characters. However, mastering the brush movement is a challenging task for many calligraphy learners as it requires many years’ practice and expert supervision. This paper presents a novel approach to help Chinese calligraphy learners to quantify the quality of brush movements without expert involvement. Our approach extracts the brush trajectories from a video stream; it then compares them with example templates of reputed calligraphers to produce a score for the writing quality. We achieve this by first developing a novel neural network to extract the spatial and temporal movement features from the video stream. We then employ methods developed in the computer vision and signal processing domains to track the brush movement trajectory and calculate the score. We conducted extensive experiments and user studies to evaluate our approach. Experimental results show that our approach is highly accurate in identifying brush movements, yielding an average accuracy of 90%, and the generated score is within 3% of errors when compared to the one given by human experts.

Parallel backscatter is a promising technique for high throughput, low power communications. The existing approaches of parallel backscatter are based on a common assumption, i.e. the states of the collided signals are distinguishable from each other in either the time domain or the IQ (the In-phase and Quadrature) domain. We in this paper disclose the superclustering phenomenon, which invalidates that assumption and seriously affects the decoding performance. Then we propose an interstellar travelling model to capture the bursty Gaussian process of a collided signal. Based on this model, we design Hubble, a reliable signal processing approach to support parallel backscatter in the wild. Hubble addresses several technical challenges: (i) a novel scheme based on Pearson's Chi-Square test to extract the collided signals' combined states, (ii) a Markov driven method to capture the law of signal state transitions, and (iii) error correction schemes to guarantee the reliability of parallel decoding. Theoretically, Hubble is able to decode all the backscattered data, as long as the signals are detectable by the receiver. The experiment results demonstrate that the median throughput of Hubble is $11.7\times$ higher than that of the state-of-the-art approach.

RFID-based positioning is emerging as a promising solution for inventory management in places like warehouses and libraries. However, existing solutions either are too sensitive to the environmental noise, or require deploying a large number of reference tags which incur expensive deployment cost and increase the chance of data collisions. This paper presents CSRP, a novel RFID based positioning system, which is highly accurate and robust to environmental noise, but relies on much less reference tags compared with the state-of-the-art. CSRP achieves this by employing an noise-resilient RFID fingerprint scheme and a compressive sensing based algorithm that can recover the target tag's position using a small number of signal measurements. This work provides a set of new analysis, algorithms and heuristics to guide the deployment of reference tags and to optimize the computational overhead. We evaluate CSRP in a deployment site with 270 commercial RFID tags. Experimental results show that CSRP can correctly identify 84.7% of the test items, achieving an accuracy that is comparable to the state-of-the-art, using an order of magnitude less reference tags.

Mobile sensing is a promising sensing paradigm that utilizes mobile device sensors to collect sensory data about sensing targets and further applies learning techniques to recognize the sensed targets to correct classes or categories. Due to the recent great success of deep learning, an emerging trend is to adopt deep learning in this recognition process, while we find an overlooked yet crucial issue to be solved in this paper - The size of deep learning models should be sufficiently large for reliably classifying various types of recognition targets, while the achieved processing delay may fail to satisfy the stringent latency requirement from applications. If we blindly shrink the deep learning model for acceleration, the performance cannot be guaranteed. To cope with this challenge, this paper presents a compact deep neural network architecture, namely cDeepArch. The key idea of the cDeepArch design is to decompose the entire recognition task into two lightweight sub-problems: context recognition and the context-oriented target recognitions. This decomposition essentially utilizes the adequate storage to trade for the CPU and memory resource consumptions during execution. In addition, we further formulate the execution latency for decomposed deep learning models and propose a set of enhancement techniques, so that system performance and resource consumption can be quantitatively balanced. We implement a cDeepArch prototype system and conduct extensive experiments. The result shows that cDeepArch achieves excellent recognition performance and the execution latency is also lightweight.

Target material identification is playing an important role in our everyday life. This paper introduces a device-free target material identification system, implemented on ubiquitous and cheap commercial off-the-shelf (COTS) Wi-Fi devices. The intuition is that different materials produce different amounts of phase and amplitude changes when a target appears on the line-of-sight (LoS) of a radio frequency (RF) link. However, due to multipath and hardware imperfection, the measured phase and amplitude of the channel state information (CSI) are very noisy. We thus present novel CSI pre-processing schemes to address the multipath and hardware noise issues before they can be used for accurate material sensing. Comprehensive real-life experiments demonstrate that we can identify 10 commonly seen liquids at an overall accuracy higher than 95% with strong multipath indoors.

This paper presents PLoRa, an ambient backscatter design that enables long-range wireless connectivity for batteryless IoT devices. PLoRa takes ambient LoRa transmissions as the excitation signals, conveys data by modulating an excitation signal into a new standard LoRa "chirp" signal, and shifts this new signal to a different LoRa channel to be received at a gateway faraway. PLoRa achieves this by a holistic RF front-end hardware and software design, including a low-power packet detection circuit, a blind chirp modulation algorithm and a low-power energy management circuit. To form a complete ambient LoRa backscatter network, we integrate a light-weight backscatter signal decoding algorithm with a MAC-layer protocol that work together to make coexistence of PLoRa tags and active LoRa nodes possible in the network. We prototype PLoRa on a four-layer printed circuit board, and test it in various outdoor and indoor environments. Our experimental results demonstrate that our prototype PCB PLoRa tag can backscatter an ambient LoRa transmission sent from a nearby LoRa node (20 cm away) to a gateway up to 1.1 km away, and deliver 284 bytes data every 24 minutes indoors, or every 17 minutes outdoors. We also simulate a 28-nm low-power FPGA based prototype whose digital baseband processor achieves 220 μW power consumption.

Code virtualization built upon virtual machine (VM) technologies is emerging as a viable method for implementing code obfuscation to protect programs against unauthorized analysis. State-of-the-art VM-based protection approaches use a fixed scheduling structure where the program always follows a single, deterministic execution path for the same input. Such approaches, however, are vulnerable in certain scenarios where the attacker can reuse knowledge extracted from previously seen software to crack applications protected with the same obfuscation scheme. This paper presents Dsvmp, a novel VM-based code obfuscation approach for software protection. Dsvmp brings together two techniques to provide stronger code protection than prior VM-based approaches. Firstly, it uses a dynamic instruction scheduler to randomly direct the program to execute different paths without violating the correctness across different runs. By randomly choosing the program execution path, the application exposes diverse behavior, making it much more difficult for an attacker to reuse the knowledge collected from previous runs or similar applications to launch an attack. Secondly, it employs multiple VMs to further obfuscate the mapping from VM opcode to native machine instructions, so that the same opcode could be mapped to different native instructions at runtime, making code analysis even harder. We have implemented Dsvmp in a prototype system and evaluated it using a set of widely used applications. Experimental results show that Dsvmp provides stronger protection with comparable runtime overhead and code size, when it is compared to two commercial VM-based code obfuscation tools.

Pattern lock is widely used for identification and authentication on Android devices. This article presents a novel video-based side channel attack that can reconstruct Android locking patterns from video footage filmed using a smartphone. As a departure from previous attacks on pattern lock, this new attack does not require the camera to capture any content displayed on the screen. Instead, it employs a computer vision algorithm to track the fingertip movement trajectory to infer the pattern. Using the geometry information extracted from the tracked fingertip motions, the method can accurately infer a small number of (often one) candidate patterns to be tested by an attacker. We conduct extensive experiments to evaluate our approach using 120 unique patterns collected from 215 independent users. Experimental results show that the proposed attack can reconstruct over 95% of the patterns in five attempts. We discovered that, in contrast to most people’s belief, complex patterns do not offer stronger protection under our attacking scenarios. This is demonstrated by the fact that we are able to break all but one complex patterns (with a 97.5% success rate) as opposed to 60% of the simple patterns in the first attempt. We demonstrate that this video-side channel is a serious concern for not only graphical locking patterns but also PIN-based passwords, as algorithms and analysis developed from the attack can be easily adapted to target PIN-based passwords. As a countermeasure, we propose to change the way the Android locking pattern is constructed and used. We show that our proposal can successfully defeat this video-based attack. We hope the results of this article can encourage the community to revisit the design and practical use of Android pattern lock.

Smartphone theft is a non-negligible problem that causes serious concerns on personal property and privacy. The existing solutions to this problem either provide only functions like retrieving a phone, or require dedicated hardware to detect thefts. How to protect smartphones from being stolen at all times is still an open problem. In this paper, we propose iGuard, a real-time anti-theft system for smartphones. iGuard utilizes only the inertial sensing data from the smartphone. The basic idea behind iGuard is to distinguish different people holding a smartphone, by identifying the order of the motions during the `take-out' behavior andhoweach motion is performed. For this purpose, we design a motion segmentation algorithm to detect the transition between two motions from the noisy sensing data. We then leverage the distinct feature contained in each sub-segment of a motion to estimate the probability that the motion is performed by the smartphone owner himself/herself. Based on such pre-processed data, we propose a Markov based model to track the behavior of a smartphone user. According to this model, iGuard instantly alarms once the tracked data deviate from the smartphone owner's usual habit. We implement iGuard on Android and evaluate its performance in real environments. The experimental results show that iGuard is accurate and robust in various scenarios.

Device-free localization of objects not equipped with RF radios is playing a critical role in many applications. This paper presents LIFS, a Low human-effort, device-free localization system with fine-grained subcarrier information, which can localize a target accurately without offline training. The basic idea is simple: channel state information (CSI) is sensitive to a target's location and thus the target can be localized by modelling the CSI measurements of multiple wireless links. However, due to rich multipath indoors, CSI can not be easily modelled. To deal with this challenge, our key observation is that even in a rich multipath environment, not all subcarriers are affected equally by multipath reflections. Our CSI pre-processing scheme tries to identify the subcarriers not affected by multipath. Thus, CSI on the “clean” subcarriers can still be utilized for accurate localization. Without the need of knowing the majority transceivers' locations, LiFS achieves a median accuracy of 0.5 m and 1.1 m in line-of-sight (LoS) and non-line-of-sight (NLoS) scenarios, respectively, outperforming the state-of-the-art systems.

While most existing indoor localization techniques are device-based, many emerging applications such as intruder detection and elderly monitoring drive the needs of device-free localization, in which the target can be localized without any device attached. Among the diverse techniques, received signal strength (RSS) fingerprint-based methods are popular because of the wide availability of RSS readings in most commodity hardware. However, current fingerprint-based systems suffer from high human labor cost to update the fingerprint database and low accuracy due to the large degree of RSS variations. In this paper, we propose a fingerprint-based device-free localization system named iUpdater to significantly reduce the labor cost and increase the accuracy. We present a novel self-augmented regularized singular value decomposition (RSVD) method integrating the sparse attribute with unique properties of the fingerprint database. iUpdater is able to accurately update the whole database with RSS measurements at a small number of reference locations, thus reducing the human labor cost. Furthermore, iUpdater observes that although the RSS readings vary a lot, the RSS differences between both the neighboring locations and adjacent wireless links are relatively stable. This unique observation is applied to overcome the short-term RSS variations to improve the localization accuracy. Extensive experiments in three different environments over 3 months demonstrate the effectiveness and robustness of iUpdater.

Apps repackaged through reverse engineering pose a significant security threat to the Android smart phone ecosystem. Previous solutions have mostly focused on the detection and identification of repackaged apps. Nevertheless, current app anti-repackaging services can only protect applications at a coarse level and get a significant performance overhead. These approaches can neither meet the performance requirements of Android nor achieve fine-grained protection against cumulative attack 1 at the same time. Specifically, these solutions rely on a fix-structure detecting engine and then will execute the same path at different times, which lead to the entire protection performs poorly when faced with dynamic cumulative attack, which is typical in real-world attack. This paper introduces AppIS, a reinforced anti-repackaging immune system, that is robust to app-repackaging attack scenarios. Unlike prior work, which mostly focuses on simple protection only from just one respect, our design exploits an interlocking guarding net with time diversity for the tamper-proofing of Android applications. The intuition underlying our design is that a dynamic and static combining method can provide a multi-level protection for the codes, core algorithm and sensitive data. We analyze and classify the existing threats on Android platform and furthermore abstract then model the repackaging attack scenarios. We then adopt a random controller used by the dispatcher to randomly construct guarding net with different structure every time. We have built a prototype of our design using Java Native Interface cross-layer calling mechanism for performance requirement. Results from a deployment of AppIS on several kinds of popular apps demonstrate that the new design can prevent our apps from cumulative attack without extra performance cost.

The augmented reality(AR) technology can enrich a person actual life, and is attracting more and more attention. However, the tradition methods have some problems such as high time delay, high deployment cost and low accuracy. These problems greatly hinder the AR technology to ubiquitous applications. In this paper, we design a AR system deployed on mobile devices, named SpeAR, which leveraging the feature matching algorithm to fast recognize the object with high accuracy. In SpeAR, we employ the depth camera embedded in the mobile device to obtain the distance between mobile device and object to use as a feature of this object image. For the accurately matching, the flutter-free algorithm will be designed to extract more accurate image feature. For the fast matching, the shrunken SURF(sSURF) is proposed to match images combining the distance feature. We have implemented a prototype system to evaluate the actual performance. The experiment results show that our solution achieves an 141ms delay in object recognition in this system.

Smartphone theft is a non-negligible problem that causes serious concerns on personal property, privacy, and public security. The existing solutions to this problem either provide only functions like retrieving a phone, or require dedicated hardware to detect thefts. How to protect smartphones from being stolen at all times is still an open problem. In this paper, we propose iGuard, a real-time anti-theft system for smartphones. iGuard utilizes only the inertial sensing data from the smartphone. The basic idea behind iGuard is to distinguish different people holding a smartphone, by identifying the order of the motions during the `take-out' behavior and how each motion is performed. For this purpose, we design a motion segmentation algorithm to detect the transition between two motions from the noisy sensing data. We then leverage the distinct feature contained in each sub-segment of a motion, instead of the entire motion, to estimate the probability that the motion is performed by the smartphone owner himself/herself. Based on such pre-processed data, we propose a Markov Chain based model to track the behavior of a smartphone user. According to this model, iGuard instantly alarms once the tracked data deviate from the smartphone owner's usual habit. We implement iGuard on Android and evaluate its performance in real environments. The experimental results show that iGuard is accurate and robust in various scenarios.

With parallel decoding for backscatter communication, tags are allowed to transmit concurrently and more efficiently. Existing parallel decoding mechanisms, however, assume that signals of the tags are highly stable, and hence may not perform optimally in the naturally dynamic backscatter systems. This paper introduces FlipTracer, a practical system that achieves highly reliable parallel decoding even in hostile channel conditions. FlipTracer is designed with a key insight: although the collided signal is time-varying and irregular, transitions between signals' combined states follow highly stable probabilities, which offers important clues for identifying the collided signals, and provides us with an opportunity to decode the collided signals without relying on stable signals. Motivated by this observation, we propose a graphical model, called one-flip-graph (OFG), to capture the transition pattern of collided signals, and design a reliable approach to construct the OFG in a manner robust to the diversity in backscatter systems. Then FlipTracer can resolve the collided signals by tracking the OFG. We have implemented FlipTracer and evaluated its performance with extensive experiments across a wide variety of scenarios. Our experimental results have shown that FlipTracer achieves a maximum aggregated throughput that approaches 2 Mbps, which is 6x higher than the state-of-the-art.

Target imaging and material identification play an important role in many real-life applications. This paper introduces TagScan, a system that can identify the material type and image the horizontal cut of a target simultaneously with cheap commercial off the-shelf (COTS) RFID devices. The key intuition is that different materials and target sizes cause different amounts of phase and RSS (Received Signal Strength) changes when radio frequency (RF) signal penetrates through the target. Multiple challenges need to be addressed before we can turn the idea into a functional system including (i) indoor environments exhibit rich multipath which breaks the linear relationship between the phase change and the propagation distance inside a target; (ii) without knowing either material type or target size, trying to obtain these two information simultaneously is challenging; and (iii) stitching pieces of the propagation distances inside a target for an image estimate is non-trivial. We propose solutions to all the challenges and evaluate the system's performance in three different environments. TagScan is able to achieve higher than 94% material identification accuracies for 10 liquids and differentiate even very similar objects such as Coke and Pepsi. TagScan can accurately estimate the horizontal cut images of more than one target behind a wall.

In recent years, unsupervised two-dimensional (2D) dimensionality reduction methods for unlabeled large-scale data have made progress. However, performance of these degrades when the learning of similarity matrix is at the beginning of the dimensionality reduction process. A similarity matrix is used to reveal the underlying geometry structure of data in unsupervised dimensionality reduction methods. Because of noise data, it is difficult to learn the optimal similarity matrix. In this letter, we propose a new dimensionality reduction model for 2D image matrices: unsupervised 2D dimensionality reduction with adaptive structure learning (DRASL). Instead of using a predetermined similarity matrix to characterize the underlying geometry structure of the original 2D image space, our proposed approach involves the learning of a similarity matrix in the procedure of dimensionality reduction. To realize a desirable neighbors assignment after dimensionality reduction, we add a constraint to our model such that there are exact [Formula: see text] connected components in the final subspace. To accomplish these goals, we propose a unified objective function to integrate dimensionality reduction, the learning of the similarity matrix, and the adaptive learning of neighbors assignment into it. An iterative optimization algorithm is proposed to solve the objective function. We compare the proposed method with several 2D unsupervised dimensionality methods. K-means is used to evaluate the clustering performance. We conduct extensive experiments on Coil20, AT&T, FERET, USPS, and Yale data sets to verify the effectiveness of our proposed method.

Device-free localization (DFL), which does not require any devices to be attached to target(s), has become an appealing technology for many applications, such as intrusion detection and elderly monitoring. To achieve high localization accuracy, most recent DFL methods rely on collecting a large number of received signal strength (RSS) changes distorted by target(s). Consequently, the incurred high energy consumption renders them infeasible for resource-constraint networks, such as wireless sensor networks. This paper introduces an energy-efficient framework for high-precision multi-target-adaptive device-free localization (E-HIPA). Compared with the existing methods, E-HIPA demands fewer transceivers, applies the compressive sensing (CS) theory to guarantee high localization accuracy with less RSS change measurements. The motivation behind the proposed E-HIPA is the sparse nature of multi-target locations in the spatial domain. Before taking advantage of this intrinsic sparseness, we theoretically prove the validity of the proposed CS-based framework problem formulation. Based on the formulation, the proposed E-HIPA primarily includes an adaptive orthogonal matching pursuit (AOMP) algorithm, by which it is capable of recovering the precise location vector with high probability, even for a more practical scenario with unknown target number. Experimental results via real testbed demonstrate that, compared with the previous state-of-the-art solutions, i.e., RTI, SCPL, and RASS approaches, E-HIPA reduces the energy consumption by up to 69 percent with meter-level localization accuracy.

Many emerging applications driven the fast development of the device-free localization (DfL) technique, which does not require the target to carry any wireless devices. Most current DfL approaches have two main drawbacks in practical applications. First, as the pre-calibrated received signal strength (RSS) in each location (i.e., radio-map) of a specific area cannot be directly applied to the new areas, the manual calibration for different areas will lead to a high human effort cost. Second, a large number of RSS are needed to accurately localize the targets, thus causes a high communication cost and the areas variety will further exacerbate this problem. This paper proposes FitLoc, a fine-grained and low cost DfL approach that can localize multiple targets over various areas, especially in the outdoor environment and similar furnitured indoor environment. FitLoc unifies the radio-map over various areas through a rigorously designed transfer scheme, thus greatly reduces the human effort cost. Furthermore, benefiting from the compressive sensing theory, FitLoc collects a few RSS and performs a fine-grained localization, thus reduces the communication cost. Theoretical analyses validate the effectivity of the problem formulation and the bound of localization error is provided. Extensive experimental results illustrate the effectiveness and robustness of FitLoc.

Device-free localization, which does not require any device attached to the target, is playing a critical role in many applications, such as intrusion detection, elderly monitoring and so on. This paper introduces D-Watch, a device-free system built on the top of low cost commodity-off-the-shelf RFID hardware. Unlike previous works which consider multipaths detrimental, D-Watch leverages the “bad” multipaths to provide a decimeter-level localization accuracy without offline training. D-Watch harnesses the angle-of-arrival information from the RFID tags' backscatter signals. The key intuition is that whenever a target blocks a signal's propagation path, the signal power experiences a drop which can be accurately detected by the proposed novel P-MUSIC algorithm. The proposed wireless phase calibration scheme does not interrupt the ongoing data communication and thus reduces the deployment burden. We implement and evaluate D-Watch with extensive experiments in three different environments. D-Watch achieves a median accuracy of 16.5 cm for library, 25.5 cm for laboratory, and 31.2 cm for hall environment, outperforming the state-of-the-art systems. In a table area of 2 $\text{m}\times 2$ m, D-Watch can track a user's fist at a median accuracy of 5.8 cm. D-Watch is also capable of localizing multiple targets which is well known to be challenging in passive localization.

Network coding is emerging as a promising technique that can provide significant improvements in the throughput of Internet of Things (IoT). Previous network coding schemes focus on several nodes, regardless of the topology and communication range in the whole network. Consequently, these schemes are greedy. Namely, all opportunities of combinations of packets in these nodes are exploited. We demonstrate that there is still room for whole network throughput improvement for these greedy design principles. Thus, in this paper, we propose a novel network coding scheme, ECS (Efficient Coding Scheme), which is designed to achieve a higher throughput improvement with lower computational complexity and buffer occupancy compared to current greedy schemes for wireless mesh networks. ECS utilizes the knowledge of the topologies to minimize interference and obtain more throughput. We also prove that the widely used expected transmission count metric (ETX) in opportunistic listening has an inherent error ratio that would lead to decoding failure. ECS therefore exploits a more reliable broadcast protocol to decrease the impact of this inherent error ratio in ETX. Simulation results show that ECS can greatly improve the performance of network coding and decrease buffer occupancy.

Device-free localization, which does not require any device attached to the target is playing a critical role in many applications such as intrusion detection, elderly monitoring, etc. This paper introduces D-Watch, a device-free system built on top of low cost commodity-off-the-shelf (COTS) RFID hardware. Unlike previous works which consider multipaths detrimental, D-Watch leverages the "bad" multipaths to provide a decimeter level localization accuracy without offline training. D-Watch harnesses the angle-of-arrival (AoA) information from the RFID tags' backscatter signals. The key intuition is that whenever a target blocks a signal's propagation path, the signal power experiences a drop which can be accurately captured by the proposed novel P-MUSIC algorithm. The wireless phase calibration scheme proposed does not interrupt the ongoing communication. Real-world experiments demonstrate the effectiveness of D-Watch. In a rich-multipath library environment, D-Watch can localize a human target at a median accuracy of 16.5 cm. In a table area of 2 m×2 m, D-Watch can track a user's fist at a median accuracy of 5.8 cm. D-Watch is capable of localizing multiple targets which is well known to be challenging in passive localization

Operating in unlicensed ISM bands, ZigBee devices often yield poor throughput and packet reception ratio due to the interference from ever increasing wireless devices in 2.4 GHz band. Although there have been many efforts made for interference avoidance, they come at the cost of miscellaneous overhead, which oppositely hurts channel utilization. Our empirical results show that, a specific interference is likely to have different influence on different outbound links of a ZigBee sender, which indicates the chance of concurrent transmissions. Based on this insight, we propose Smoggy-Link, a practical protocol to exploit the potential concurrency for adaptive ZigBee transmissions under harsh interference. Smoggy-Link maintains an accurate link model to describe and trace the relationship between interference and link quality of the sender's outbound links. With such a link model, Smoggy-Link can obtain fine-grained spatiotemporal link information through a low-cost interference identification method. The link information is further utilized for adaptive link selection and intelligent transmission schedule. We implement and evaluate a prototype of our approach with TinyOS and TelosB motes. The evaluation results show that Smoggy-Link has consistent improvements in both throughput and packet reception ratio under interference from various interferer.

Vehicle-to-Vehicle (V2V) communication is a vital component of vehicular ad-hoc networks (VANET) under the situation that infrastructure for vehicle-to-infrastructure (V2I) has not been well deployed due to its cost and suffering. However, messages transmission path is so challenge to be found without infrastructures supporting that packets are inevitably spread in a sparsely or competitive area, in which they should avoid being trapped because of its poor communication links between vehicles, which eventually lowers down the performance of messages propagation. In this paper, we analyze the relationship between vehicular geographical distribution and packets propagation of VANET in a realistic large-scale urban scenario. It is demonstrated that, from a macro perspective, we could guide the path selection of data propagation between source and destination through V2V communication on the basis of the feature of vehicle density in different geographic locations. Furthermore, to be further close to the actual traffic environment, we model the vehicular geographical distribution by four typical real scenes to present the real environment and develop appropriate messages propagation strategies respectively.

Device-free localization (DfL) techniques, which can localize targets without carrying any wireless devices, have attracting an increasing attentions. Most current DfL approaches, however, have two main drawbacks hindering their practical applications. First, one needs to collect large number of measurements to achieve a high localization accuracy, inevitably causing a high deployment cost, and the areas variety will further exacerbate this problem. Second, as the pre-obtained Received Signal Strength (RSS) from each location (i.e., radio-map) in a specific area cannot be directly applied to new areas for localization, the calibration process of different areas will lead to the high human effort cost. In this paper, we propose, FitLoc, a fine-grained and low cost DfL approach that can localize multiple targets in various areas. By taking advantage of the compressive sensing (CS) theory, FitLoc decreases the deployment cost by collecting only a few of RSS measurements and performs a fine-grained localization. Further, FitLoc employs a rigorously designed transfer scheme to unify the radio-map over various areas, thus greatly reduces the human effort cost. Theoretical analysis about the effectivity of the problem formulation is provided. Extensive experimental results illustrate the effectiveness of FitLoc.

The low-cost crystal oscillators embedded in wireless sensor nodes are prone to be affected by their working condition, leading to undesired variation of clock skew. To preserve synchronized clocks, nodes have to undergo frequent re-synchronization to cope with the time-varying clock skew, which in turn means excessive energy consumption. In this paper, we propose DualSync, a synchronization approach for low-power wireless networks under dynamic working condition. By utilizing time-stamp exchanges and local measurement of temperature and voltage, DualSync maintains an accurate clock model to closely trace the relationship between clock skew and the influencing factors. We further incorporate an error-driven mechanism to facilitate interplay between Inter-Sync and Self-Sync, so as to preserve high synchronization accuracy while minimizing communication cost. We evaluate the performance of DualSync across various scenarios and compare it with state-of-art approaches. The experimental results illustrate the superior performance of DualSync in terms of both accuracy and energy efficiency.

This paper introduces a wireless phase calibration algorithm that can be applied on cheap commercial off-the-shelf (COTS) radio frequency identification (RFID) system and auto acquire an accurate radio frequency (RF) phase information without any offline training. The key observation is that the raw phase measurements even measured form different RFID tags contain a same set of unknown phase errors. With enough tags' phase measurements, we can determine all the unknown phase errors, since the number of known phase measurements is much larger than the number of unknown phase errors. Real-world experimental results demonstrate the effectiveness of the proposed method.

Device-free localization of people and objects indoors not equipped with radios is playing a critical role in many emerging applications. This paper presents an accurate model-based device-free localization system LiFS, implemented on cheap commercial off-the-shelf (COTS) Wi-Fi devices. Unlike previous COTS device-based work, LiFS is able to localize a target accurately without offline training. The basic idea is simple: channel state information (CSI) is sensitive to a target's location and by modelling the CSI measurements of multiple wireless links as a set of power fading based equations, the target location can be determined. However, due to rich multipath propagation indoors, the received signal strength (RSS) or even the fine-grained CSI can not be easily modelled. We observe that even in a rich multipath environment, not all subcarriers are affected equally by multipath reflections. Our pre-processing scheme tries to identify the subcarriers not affected by multipath. Thus, CSIs on the "clean" subcarriers can be utilized for accurate localization.

Many emerging applications drive the needs of device-free localization (DfL), in which the target can be localized without any device attached. Because of the ubiquitousness of WiFi infrastructures nowadays, the widely available Received Signal Strength (RSS) information at the WiFi Access points are commonly employed for localization purposes. However, current RSS based DfL systems have one main drawback hindering their real-life applications. That is, the RSS measurements (fingerprints) vary slowly in time even without any change in the environment and frequent updates of RSS at each location lead to a high human labor cost. In this paper, we propose an RSS based low cost DfL system named TafLoc which is able to accurately localize the target over a long time scale. To reduce the amount of human labor cost in updating the RSS fingerprints, TafLoc represents the RSS fingerprints as a matrix which has several unique properties. Based on these properties, we propose a novel fingerprint matrix reconstruction scheme to update the whole fingerprint database with just a few RSS measurements, thus the labor cost is greatly reduced. Extensive experiments illustrate the effectiveness of TafLoc, outperforming the state-of-the-art RSS based DfL systems.

Passive localization is fundamental for many applications such as activity monitoring and real-time tracking. Existing received signal strength (RSS)-based passive localization approaches have been proposed in the literature, which depend on dense deployment of wireless communication nodes to achieve high accuracy. Thus, they are not cost-effective and scalable. This paper proposes the RSS distribution-based localization (RDL) technique, which can achieve high localization accuracy without dense deployment. In essence, RDL leverages the RSS and the diffraction theory to enable RSS-based passive localization in sensor networks. Specifically, we analyze the fine-grained RSS distribution properties at a variety of node distances and reveal that the structure of the triangle is efficient for low-cost passive localization. We further construct a unit localization model aiming at high accuracy localization. Experimental results show that RDL can improve the localization accuracy by up to 50%, compared to existing approaches when the error tolerance is less than 1.5 m. In addition, we apply RDL to facilitate the application of moving trajectory identification. Our moving trajectory identification includes two phases: an offline phase where the possible locations can be estimated by RDL and an online phase where we precisely identify the moving trajectory. We conducted extensive experiments to show its effectiveness for this application - the estimated trajectory is close to the ground truth.

Intermittent connection of wireless links, caused by low duty-cycle radio operation, harsh working environment, movement of sensor nodes, etc., makes clock synchronization a challenging task. Prior synchronization approaches in wireless sensor networks (WSNs) typically require that nodes exchange time messages frequently with the reference clock, which is difficult in networks with low or intermittent connectivity. This poster presents RobSync, a robust design for clock synchronization in intermittent-connected wireless networks. Having recognized that clock skew is highly correlated to the voltage supply, we use the local voltage information as a reference for clock self-calibration, which helps reduce the frequency of time-stamp exchanges. To prevent a misuse of the voltage information, leading to error accumulation, a re-synchronization interval adjustment design is developed to make a trade-off between accuracy and energy consumption. We present the theory behind RobSync, and provide preliminary results by experiments to compare our approach and the recent approach.

This poster introduces JRD, a novel device-free localization system which can achieve high accuracy with low cost and little human effort, and is even robust to different scenarios. Unlike the previous Radio Signal Strength (RSS)-based systems which depend on the dense deployment to provide high accuracy, JRD extracts the fine-grained RSS distributions of a single link and presents a voting algorithm based on multi-link to identify the object location accurately while maintaining a low-cost deployment. Furthermore, JRD is flexible to different scenarios by using the transferring technique with less time-consuming and human effort. Experimental results show that JRD can improve the localization accuracy by up to 50% with less cost as compared with the existing RSS approaches.

For a long history, stampede is one of the high potential disaster when thousands of people gathered. Current monitoring systems, however, can only detect the presence of a small number of sparsely located targets, rather than to monitor the change of people flow where there are large number of dense crowd in the environment. This paper presents DanSen, a low-cost people flow monitoring system for sensing the potential danger using the existing wifi infrastructures. Inspired by the dynamic light scattering (DLS) theory, the designed DanSen calculates the correlations between the initial channel state information (CSI) data and all the history CSI data to monitor the changes of people flow and also estimates the sharpness of the changes. By doing so, DanSen can be utilised to perceive the potential danger. Real-world experimental results illustrate the advantage and effectiveness of DanSen.

Many emerging applications and the ubiquitous wireless signals have accelerated the development of Device Free localization (DFL) techniques, which can localize objects without the need to carry any wireless devices. Most traditional DFL methods have a main drawback that as the pre-obtained Received Signal Strength (RSS) measurements (i.e., fingerprint) in one area cannot be directly applied to the new area for localization, and the calibration process of each area will result in the human effort exhausting problem. In this paper, we propose FALE, a fine-grained transferring DFL method that can adaptively work in different areas with little human effort and low energy consumption. FALE employs a rigorously designed transferring function to transfer the fingerprint into a projected space, and reuse it across different areas, thus greatly reduce the human effort. On the other hand, FALE can reduce the data volume and energy consumption by taking advantage of the compressive sensing (CS) theory. Extensive real-word experimental results also illustrate the effectiveness of FALE.

This poster introduces MoveLoc, a system that locates a moving object without carrying any devices from detecting Doppler shifts reflected off the moving object. It works even if the object walking or running in different directions without any training. MoveLoc does not require the user to carry any communication devices, yet its accuracy exceeds current moving object localization systems using Radio Signal Strength(RSS). We implement the system and evaluate the performance on moving person by experiments. Experimental result shows that MoveLoc achieves an average location accuracy of 0.69 meters and reduces the equipment deployment density compared with other known wireless moving object localization approaches using RSS.

In this paper, we propose a novel clock calibration approach, which addresses two key challenges for clock calibration in Wireless Sensor Networks: excessive communication overhead and the trade-off between accuracy and cost. To achieve this, our approach leverages the fact that the clock skew is highly correlated to temperature, which can serve as both an assistant for clock skew estimation and a regulatory factor for the duty-cycled design. Our approach is one order of magnitude more power-efficient than communication based approaches since the calibration largely relies on local temperature information. In addition, our approach provides a nice feature of self-adaptive period, which can substantially promote the system flexibility. We present the theory behind our approach, and provide preliminary results of a simulated comparison of our approach and some recent approaches.

This poster presents GuideLoc, a highly efficient aerial wireless localization system that uses directional antennas mounted on a mini Multi-rotor Unmanned Aerial Vehicle (UAV), to enable detecting and positioning of targets. Taking advantage of angle and signal strength information of frames transmitted from targets, GuideLoc can directly fly towards the targets with the minimum flight route and time. We implement a prototype of GuideLoc using ArduCopter and evaluate the performance by simulations and experiments. Experimental results show that GuideLoc achieves an average location accuracy of 2.7 meters and reduces flight distance more than 50% compared with other known wireless localization approaches using UAV.

Without relying on devices carried by the target, device-free localization (DFL) is attractive for many applications, such as wildlife monitoring. There still exist many challenges for DFL for multiple targets without dense deployment of sensor nodes. To fit the gap, in this paper, we propose a multi-target localization method based on compressive sensing, named LCS. The key observation is that given a pair of nodes, the received signal strength (RSS) will be different when a target locates at different locations. Taking advantage of compressive sensing in sparse recovery to handle the sparse property of the localization problem, (i.e., the vector which contains the number and location information of k targets is an ideal k-sparse signal), we presented a scalable compressive sensing based multiple target counting and localization method i.e., LCS, and rigorously justify the validity of the problem formulation. The results from our realistic deployment in a 12m×12m open space are promising. For 12 people with 24 nodes, the worst localization error ratio and counting error ratio of our LCS is no more than 8.3% and 33.3% respectively.

The Radio Signal Strength Indicator (RSSI)-based localization algorithm is an effective solution for passive object localization. However, the localization accuracy of the existing methods highly depends on the transceiver distance and deployment density. Generally speaking, to obtain higher accuracy, we need a denser sensor node deployment, which results in a higher deployment cost and more communication overheads. In this paper, we investigate this problem based on extensive measurements. According to the measurement results, we propose to localize objects using an RSSI distribution based localization (RDL) model to identify the object location by different RSSI distributions of the communicating links. Experimental results show that the RDL method can achieve higher localization accuracy with less sensor nodes.

Earthen site is an important part of our cultural heritage, so monitoring and protecting its living condition is vital. Like other country, in china many sites also face collapse even destruction danger. In this paper, we present a novel framework and a site survival model to monitor Danfeng Gate of Daming Palace, analyze the data collected by wireless sensor networks (WSNs) for early warning. In order to forecast the health trend, we also adopt the Neural networks and Bayesian model processing internal and external environment data to predict the health status.

In this demo, we design a monitoring system based on sensor network for zoologists to research the activity budget of Rhinopithecus roxellana. We designed the hardware platform of the system for wild deployment and presented an analysis of the relationship between environmental factors and survival conditions of Rhinopithecus roxellana. The system can locate and track Rhinopithecus roxellana by individual identification. The prototype has been deployed in Wildlife Reserve of Qiling Mountain, China.

A novel approach of dynamic graphic software watermark is proposed. In this scheme, many fake watermarks created through encoding multi-constant, which's structure is similar to the only true one's, are introduced to enhance the stealth and anti-attack ability of dynamic graph watermark and keep software from sabotage. Meanwhile, a detailed analysis in theory is made in terms of the principle, feasibility and merits of this approach. A series of tests are made on the basis of the prototype system, analyzing the two sub-systems separately about their validity, robustness and performance overload caused by watermark embedding. Moreover, bit-rate of the three graph watermark structures in the system is analyzed in theory.

This paper proposes a new method aiming at integrating the safety analysis in software design and development. Model Driven Architecture is used for the basic framework. By making use of UML extension, UMLsec models the Platform Independent Model of software safety, which achieves more safety requirements in the initial stage of the system design cycle. This approach reduces the risk and the cost of software development and improves the reusability of software.