AI development is often framed as the outcome of isolated research and engineering efforts, yet evidence from deployed systems suggests that language models interact through a shared data ecosystem.While the optimization of individual models is extensively studied, the emergent properties of this interconnected population remain largely unexplored, limiting our ability to predict long-term ecosystem trajectories.We term this process data pollination, the unintentional circulation of synthetic model outputs through shared online platforms and web-scale training corpora, and formalize it as a population-based evolutionary framework to investigate stability dynamics under synthetic data training.Our theoretical analysis and controlled experiments involving 320 language models demonstrate that population dynamics can mitigate the model collapse observed in single-lineage recursive training, yielding stable or improving performance across diverse benchmarks.Crucially, we find that ecological diversity functions as a fundamental resilience mechanism that safeguards the ecosystem against collapse, highlighting the critical importance of maintaining model diversity for sustainable AI development.

Tremor is a common symptom of neurological diseases. The regular assessment of daily tremors facilitates the evaluation of disease progression and assists clinicians in optimizing treatment strategies. However, current home monitoring solutions have difficulty in dealing with user cooperation, privacy concerns, environmental interference, and system generalization, leading to feasibility concerns in activities of daily living (ADL). To this end, we propose Tremerity-Fi, a non-contact and privacy-friendly tremor severity assessment system based on mmWave radar. To realize Tremerity-Fi, we first design an adaptive beamforming algorithm to accurately identify useful but weak signals from numerous reflections captured in the environment. Second, unlike primary reflections commonly used in mmWave sensing, we leverage multipath reflections that carry useful information about the target's motion, even though they are generally considered harmful, to help reconstruct hand signals and improve sensing performance. Furthermore, we propose an unsupervised domain adaptation algorithm to improve the ability to adapt to unseen environments and users. We collect a diverse dataset of 5 patients and 25 healthy subjects in 3 scenarios, such as offices, homes, and hospitals. Extensive experiments show that our system achieves 94.51% accuracy in tremor detection, about 5 higher than the SOTA mmWave radar method, and 89.13% in tremor severity assessment, demonstrating its sufficient potential as a tremor monitoring assistant for patients with neurological diseases.

As a significant biometric identification technology, face recognition (FR) is extensively utilized in identity verification and security surveillance systems. Current research predominantly relies on high-definition RGB camera-based methods. However, these methods are susceptible to various factors such as lighting conditions and disguises. This paper proposes a low-cost face recognition solution called Warm- Face based on thermal array sensors. By leveraging the thermal radiation of the face, we overcome the disturbances caused by lighting conditions and disguises, thereby achieving rapid and highly accurate face recognition. However, face recognition based on thermal array sensors still faces two major challenges. Firstly, in complex scenarios, thermal noise interference can lead to the thermal radiation characteristics of the target face becoming indistinguishable from the background. Secondly, due to their large network parameter sizes and high computational complexity, recognition models face challenges in simultaneously achieving low latency and high accuracy. WarmFace extracts facial regions through a semantic segmentation-based approach, effectively reducing the impact of background interference on recognition performance. Additionally, in the recognition model, we utilize a series of linear transformations instead of convolution operations to process the intrinsic features of images, which reduces redundancy in feature maps while preserving the essential information. Extensive real-world experiments validate the effectiveness of WarmFace in various environments, achieving an average recognition accuracy of 98.6%.

Deep learning technologies have seen widespread adoption in WiFi-based wireless sensing systems. However, they are inherently vulnerable to adversarial perturbation attacks, which has received little attention within the WiFi sensing community. To more comprehensively understand the potential threats posed by perturbation attacks, we present a novel attack method, named WiIntruder, distinguishing itself with universality, robustness, and stealthiness. This paper intends to provide a catalyst that promotes the assessment of security in existing WiFi-based sensing systems. We achieve the three aforementioned salient features in WiIntruder through the following three steps: (1) Maximizing transferability by differentiating user-state-specific feature spaces across sensing models, thereby enabling a universal perturbation attack vector applicable to a wide range of applications; (2) Mitigating the impact of perturbation signal distortion by optimizing key factors of device synchronization and wireless propagation through a heuristic particle swarm algorithm; and (3) Enhancing the diversity and stealthiness of attack patterns by randomly switching among perturbation surrogates generated by a generative adversarial network. Experimental results confirm the threat posed by WiIntruder to four common WiFi-based services, with the average accuracy decrease by 72.9% under black-box attack scenarios.

Person re-identification (Re-ID) can recognize users based on their clothing, body shape, and other information without the need for clear facial images, and is widely applied in the field of intelligent security. Traditional Re-ID systems mainly rely on high-definition RGB cameras, but the deployment of large-scale high-definition RGB cameras indoors has caused serious privacy and ethical concerns. Recently, wireless-based Re-ID systems (Wi-Fi, RFID, millimeter-wave radar, etc.) have shown promising prospects, but the limited sensing resolution hinders their practical deployment. In this paper, we propose WarmGait, a Re-ID system based on thermal array sensors, which can achieve high-precision Re-ID at low cost and minimize the invasion of user privacy. However, using thermal arrays for Re-ID still faces two major challenges. The first is the low and unclear texture resolution of images caused by low-cost infrared devices. The second is that existing gait recognition methods require maintaining the sequential constraint of gait images, which reduces the flexibility of gait recognition or Re-ID. To address these two challenges, we first designed an edge module inspired by Taylor Finite Difference (TFD) to aggregate image edge information to help improve the resolution of infrared devices. Then, we considered gait as a collection of gait profiles and extracted features from the frame level and collection level for recognition, breaking through the limitations of the number and order of input images. After extensive experimental evaluation, our model can achieve an average recognition accuracy of 87.3% in various scenarios, demonstrating the potential of WarmGait in Re-ID.

Detecting stowaways hidden in various transport vehicles, including cars, trucks, containers, and trailers, is crucial to border security inspection systems. Existing solutions mainly rely on contact-based sensors and manual inspection, which significantly compromise the efficiency of border control operations. Therefore, there is an urgent need for an automated, efficient and fast non-contact vehicle hidden person detection system. In this paper, we propose SiVe, a novel border inspection system that utilizes laser Doppler vibrometer (LDV) to detect hidden people in the vehicle. We extract signals associated with human activities (such as breathing, heartbeat, low-frequency body movements, etc.) from complex laser reflection data to detect the presence of hidden people. Specifically, we first employs the Empirical Mode Decomposition (EMD) algorithm to extract and reconstruct signals associated with human activities in complex and noisy environments. Then based on the characteristics of EMD outputs, we design a Time-series Variation Feature extraction Identification network (TVFI-net) model that accurately captures complex time-varying patterns for efficient and reliable detection of hidden people presence. Extensive real-world experiments validate the effectiveness of SiVe in various environments. The system achieves an average presence detection accuracy of 99.93$\%$for sedans and MPVs, 98.37$\%$for light trucks, 98.07$\%$for heavy trucks, and 95.23$\%$for trailers in non-contact detection of hidden people across twelve different vehicle types, under various indoor and outdoor environments.

LoRa holds significant promise for marine monitoring and communication due to its advantages of long range, low power consumption, and low cost. However, in marine environments, its communication performance is severely degraded by the strong absorption of electromagnetic waves by seawater. To enhance the reliability of communication under weak channels in the marine environment, this article proposes FCLoRa, a LoRa receiver enhancement system for low signal-to-noise ratio (SNR) marine environments. FCLoRa adopts a dual-domain cooperative strategy between the transmitter and receiver. On the receiver side, it employs a multilevel accumulation scheme to detect packets by aggregating the energy of windowed symbol “fragments” and reconstructs complete signals by fusing weak, fragmented signals from multiple gateways. On the transmitter side, it builds a two-dimensional polarization fingerprint library based on antenna attitude sensing and dynamically adjusts the transmission direction to match the polarization characteristics of the base station, thereby minimizing signal loss. Experimental results show that FCLoRa achieves a packet detection rate of nearly 40% at an extremely low SNR of –35 dB, and improves the average SNR by 1.92 dB compared to conventional LoRa reception, demonstrating its practical value in extreme marine scenarios.

In recent years, Wi-Fi sensing has attracted much attention due to the widespread deployment of communication devices. Due to advancements in signal processing algorithms, contactless sensing technology based on Wi-Fi signals has now been widely applied. However, the limited bandwidth of Wi-Fi systems constrains the performance of Wi-Fi sensing, posing challenges for accomplishing more fine-grained tasks (distinguishing more gestures or multiple targets, etc.). To address this challenge, in this paper, we design a spectrogram enhancement network for Wi-Fi channel state information (CSI) based on the characteristics of Wi-Fi signals to improve the sensing capability of Wi-Fi signals. Specifically, we use a neural network to generate super-resolution spectrograms of CSI to distinguish different time-frequency components in the environment at a finer granularity. Through extensive evaluation, we demonstrate that our designed system can achieve finer-grained perception accuracy than the state-of-the-art systems.

With the aging population, there is an increasing demand for rehabilitation services for people with chronic diseases. However, limitations such as medical resources, geographic barriers, and cost make home rehabilitation an option for more patients. Existing wearable devices and vision methods are effective but face problems with portability, cost, and privacy concerns. As for existing wireless sensing methods, they can only extract coarse features for activity recognition. Therefore, we present EchoHealth, which utilizes a smart speaker for rehabilitation exercise detection and assessment. We upgrade the smart speaker into an active sonar system without hardware modification to generate acoustic micro-distance images with motion information. Then, time-domain motion detection and distance-domain feature extraction are utilized to filter out the effects of non-motion time and distance to extract patient motion features for motion recognition. We further assess the patient's rehabilitation exercises from five aspects, based on which EchoHealth provides rehabilitation guidance. Extensive experiments with 15 participants performing 12 rehabilitation motions confirmed that EchoHealth can achieve 97.4% average accuracy in recognition of rehabilitation motion and provide accurate rehabilitation indicators in various environments.

Deep learning-based multi-view stereo (MVS) methods enable dense point cloud reconstruction in texture-rich areas. However, existing methods incur significant computational costs to capture pixel dependencies for complete reconstruction in low-texture regions. Additionally, discrete depth layers in occluded environments hinder the cost volume’s ability to model object information effectively. To address these issues, we propose a spatial-aware multi-view stereo network with attention cost volume, termed SA-MVSNet. The network introduces the pixel-driven spatial interaction (PDSI) module, which integrates the hierarchical spatial location enhancement mechanism (HSLE) and the spatial context aggregation mechanism (SCA). Leveraging an efficient parallel architecture, the PDSI module captures pixel-level spatial dependencies with the HSLE and strengthens global contextual information through the SCA. This design improves the network’s ability to represent features in low-texture regions while maintaining high inference efficiency. Furthermore, SA-MVSNet incorporates an attention weight generation branch that refines the cost volume by aggregating multi-scale depth cues, effectively mitigating the impact of occlusion. Experiments on the DTU dataset and the Tanks and Temples dataset show that our method outperforms other learning-based methods, achieving superior performance and strong generalization ability.

Recent years, point-of-sale (POS) terminals are no longer limited to wired connections, with many relying on Wi-Fi for data transmission. Although Wi-Fi offers the convenience of wireless connectivity, it introduces significant security vulnerabilities. This work presents a non-intrusive method for eavesdropping POS passwords via Wi-Fi sensing, named${\mathsf {BeamThief}}$. Instead of conventional Wi-Fi Channel State Information (CSI) readings, our approach employs Wi-Fi Beamforming Feedback Information (BFI) for an eavesdropping attack. Compared to CSI, which can only be extracted through intruding into the Access Point (AP) or from a limited selection of commercial Wi-Fi cards (e.g., Intel-5300), BFI readings can be more readily obtained from a broad array of commercial Wi-Fi devices. A key technological contribution of${\mathsf {BeamThief}}$is the development of an analysis model for predicting finger motion trajectories. This model is based on the physical relationship between BFI readings and finger motion, thus eliminating the need for extensive labeled training data. Furthermore, we employ Maximum Ratio Combining (MRC) to enhance the BFI series, ensuring performance across various scenarios. We implement${\mathsf {BeamThief}}$using everyday commercial Wi-Fi devices and conduct a series of experiments to assess the impact of this attack. Experimental results demonstrate that${\mathsf {BeamThief}}$achieves an accuracy rate 79$\%$in inferring 6-digit POS passwords within the top-100 attempts.

In recent years, with increasing user demands for convenience, privacy, and personalized experiences, gait recognition has been widely studied across various domains, such as indoor intrusion detection and smart homes. Although computer vision solutions are extensively researched for their visual intuitiveness, Wi-Fi sensing is emerging as a new research focus due to its ability to preserve privacy. However, previous studies have primarily relied on abstract features with limited interpretability or required multiple Wi-Fi links. To address these issues, we propose Wi-GR, which utilizes a Wi-Fi link to extract robust and highly interpretable gait features for user recognition. First, we construct a multi-path gait signal model to establish a clear relationship between Channel State Information (CSI) and gait motion. Then, we design a gait signal separation and enhancement method to mitigate the effects of external non-target reflections and internal multi-part reflections, which significantly impact the extraction and interpretability of gait features. Finally, fine-grained gait features that visualize gait patterns are generated using MUSIC-based and GAN-based multi-part velocity profile generation algorithms, tailored for single-person and multi-person scenarios, respectively. Numerous experiments have demonstrated that Wi-GR achieves single-person recognition accuracies of 95.3%, 94.0%, and 93.2% for 30 persons in the meeting room, corridor, and lobby, respectively, and an average accuracy of 88.3% for two-person recognition.

With the development of Wi-Fi sensing, wireless-based gait recognition has become increasingly important as it supports a wide range of applications (person identification, disease diagnosis, etc.). However, two serious challenges limit the universal deployment of such Wi-Fi vision schemes: i) the limited bandwidth of Wi-Fi severely restricts the granularity of gait recognition, and ii) users non-gait behaviors (e.g., stopping and turning) interfere with the extraction of gait-related features. In this paper, we propose CSID, which can achieve robust gait recognition under the limited bandwidth conditions of commercial Wi-Fi devices. Specifically, we use a neural network to generate super-resolution spectrograms of channel state information (CSI), overcoming the limitation of insufficient Wi-Fi bandwidth. To overcome the challenge of non-gait behavior interference, considering the human-incomprehensible nature of Wi-Fi spectrograms, we adopt cross-domain adversarial training and further extract gait features that are independent of the interference behaviors by learning domain-independent representations. We conducted a large number of experiments in different indoor environments, and the average person identification rate of the CSID system reached 91.6%. These results demonstrate that the CSID system is promising and could be used as a complement to visual person identification systems in the future.

Nowadays, point-of-sale (POS) terminals are no longer limited to wired connections, and many of them rely on Wi-Fi for data transmission. While Wi-Fi provides the convenience of wireless connectivity, it also introduces significant security risks. Previous research has explored Wi-Fi-based eavesdropping methods. However, these methods often rely on limited environmental robustness of Channel State Information (CSI) and require invasive Wi-Fi hardware, making them impractical in real-world scenarios. In this work, we present SThief, a practical Wi-Fi-based eavesdropping attack that leverages beamforming feedback information (BFI) exchanged between POS terminal and access points (APs) to keystroke inference on POS keypads. By capitalizing on the clear-text transmission characteristics of BFI, this attack demonstrates a more flexible and practical nature, surpassing traditional CSI-based methods. BFI is transmitted in the uplink, carrying downlink channel information that allows the AP to adjust beamforming angles. We exploit this channel information to keystroke inference. To enhance the BFI series, we use maximal ratio combining (MRC), ensuring efficiency across various scenarios. Additionally, we employ the Connectionist Temporal Classification method for keystroke inference, providing exceptional generalization and scalability. Extensive testing validates SThief’s effectiveness, achieving an impressive 81% accuracy rate in inferring 6-digit POS passwords within the top-100 attempts.

Wi-Fi signals are commonly used for conventional communication, yet they can also realize low-cost and non-invasive human sensing. However, Wi-Fi sensing in Multi-person scenarios is still a challenging problem. In this paper, we propose M2-Fi to achieve multi-person respiration monitoring using a handheld device. M2-Fi leverages Wi-Fi BFI (beamforming feedback information) performs respiration monitoring. As a compressed version of the uplink CSI (channel state information), BFI transmission is unencrypted, easily obtained using frame capture, and does not require specific firmware to obtain. M2-Fi is based on an interesting experiment phenomenon that when a Wi-Fi device is very close to a subject, near-field channel changes caused by the subject significantly cancel out changes from other subjects. We employed VMD (Variational Mode Decomposition) to eliminate the interference caused by hand movement in the BFI time series. Subsequently, we devised a deep learning architecture based on GAN (Generative Adversarial Networks) to recover fine-grained respiration waveforms from the respiration patterns extracted from the BFI time series. Our experiments on collected 50-hour data from 8 subjects show that M2-Fi can accurately recover the respiration waveforms of multiple persons with handheld devices.

Real-time indoor crowd counting plays an important role in many applications such as crowd control, resource allocation and advertisement. Current research predominantly relies on camera-based methods. However, computer vision-based solutions raise severe privacy and ethical concerns. In this paper, we propose a privacy-preserving counting solution called BeamCount based on Wi-Fi sensing. Instead of using conventional Wi-Fi Channel State Information (CSI) readings, we utilize Wi-Fi Beamforming Feedback Information (BFI) for crowd counting estimation. Compared to CSI which can only be extracted from few commodity Wi-Fi cards (e.g., Intel 5300), BFI readings can be obtained from a large range of commodity Wi-Fi devices. We establish a mapping relationship between BFI and headcount and extract headcounts from BFI inputs through a carefully designed adversarial network. Owing to the adversarial network's cross-domain capability, the proposed counting system can achieve high accuracy across different environments, demonstrating its generalization capability. To mitigate the effect of BFI compression on sensing performance, we adopt a novel time series prediction model. Extensive real-world experiments validate the effectiveness of BeamCount in various environments, achieving an average counting accuracy of 93.6%.

Existing rumor detection methods have not ade-quately considered which features should be addressed at the macroscopic or micro levels. Furthermore, the fundamental manifestation of news dissemination-public opinion-has received insufficient attention. We propose a macro-micro based public opinion (MMPO) modeling method to address these gaps. This approach enables a better understanding of news dissemination on social media platforms. Firstly, we incorporate post sentiment as a crucial feature. Secondly, we examine how the propagation at the micro-level influences sub-posts. Next, we investigate the temporal variations in public opinion from a macro perspective. Lastly, we integrate the acquired general propagation information with textual data to obtain more distinct representations of news propagation. By leveraging the macro perspective, the proposed model effectively captures the fine-grained characteristics of news propagation and helps smooth out the noise in user comments. The experimental evaluation showcases the superiority of our model in rumor detection compared to existing methods.

Self-service libraries need self-service book collection and monitoring of book quality to improve user experience This article proposes a privacy-preserving alternative RFbook, a book classification and moisture sensing system formed from an array of passive commercial RFID tags. We have three key observations in designing RFbook for such benefits. The first observation is that when tags are in the vicinity, their interrogation currents can alter each other’s circuit properties, based on which unique phase and amplitude signatures can be obtained from the backscattered signal. The second observation is that books with different thicknesses and sizes of material will have different signal features. Finally, we found that changes in book humidity are reflected in the reader’s received signal strength (RSS). To turn the high-level idea into a practical system, we built a prototype of RFbook and conducted comprehensive experiments to evaluate the system’s performance. The experimental results show that RFbook can distinguish different types of books with an average accuracy rate higher than 96% and monitor the humidity change of the book.

To balance venue safety with public photography rights, this article presents CamShield—a novel system for selective defense against unauthorized photography. Amid dense electromagnetic environments, CamShield reliably identifies cameras by analyzing their unintended electromagnetic emissions. By tracing frequency drift patterns and harmonic spectral movements unique to each device, CamShield can accurately detect cameras despite environmental noise or model similarities. An integrated antenna amplitude ratio module and Kalman filter further localize threats through resilient positioning. Directional ultrasonic beams then focus tuned acoustic interference toward devices, temporarily disrupting visualization in restricted locations while preserving ambient imaging freedoms. Comprehensive evaluations across three state-of-the-art object detectors quantify real-world reliability. With 30 intruding cameras, CamShield exhibited obstruction latencies below 346 ms. Furthermore, CamShield achieves three times the coverage using the same power as traditional Omnidirectional transmission. Together, the breakthroughs in pervasive camera sensing and context-aware actuation contribute toward advancing policy-centric access controls at the edge of cyber-physical convergence. CamShield sets an important precedent on enforcing venue custom protections in bounded secure zones without undermining positive public photography assumptions elsewhere.

Wi-Fi signals contain information about the surrounding propagation environment and have been widely used in various sensing applications such as gesture recognition, respiratory monitoring, and indoor position. Nevertheless, this information can also be easily stolen by eavesdroppers to obtain private information. In this paper, we propose WiShield, a new framework that protects legitimate users using Wi-Fi sensing applications while preventing unauthorized privacy attacks. The implementation of WiShield is based on a simple principle of physically encrypting Wi-Fi channel status information (CSI) to prevent eavesdroppers from inferring sensitive information through stolen CSI. To achieve a balance between encryption strength, sensing accuracy, and communication quality, we design an efficient multi-objective optimization framework that can safely deliver decryption keys to legitimate users and prevent illegal eavesdropping by eavesdroppers. We implemented the WiShield prototype on an SDR platform and conducted extensive experiments to verify its effectiveness in common Wi-Fi sensing applications. We believe that the implementation of WiShield can improve the privacy standards of Wi-Fi sensing applications, and it is also an important step towards making the integration of Integrated Sensing and Communications (ISAC).

Accurate head movement tracking is crucial for virtual reality and Metaverse in ubiquitous human-computer interaction (HCI) applications. Existing works for head tracking with wearable VR kits and wireless signals require expensive devices and heavy algorithmic processing. To resolve this problem, we propose HeadTrack, a low-cost, high-precision head motion tracking system that uses commercially available wireless earphones to capture the user’s head motion in real-time. HeadTrack uses smartphones as ‘sound anchors’ and emits inaudible chirps picked up by the user’s wireless earphones. By measuring the time-of-flight of these signals from the smartphone to each microphone on the earphone, we can deduce the user’s face orientation and distance relative to the smartphone, enabling us to accurately track the user’s head movement. To realize HeadTrack, we use the cross-correlation method to optimize the Frequency Modulated Continuous Wave (FMCW) based acoustic ranging method, which solves the problem of insufficient wireless earphone bandwidth. Moreover, we solve the problems of asynchronous startup time between devices and the existence of sampling frequency offset. We conduct excessive experiments in real scenarios, and the results prove that HeadTrack can continuously track the direction of the user’s head, with an average error under 6.3° in pitch and 4.9° in yaw.

The increasing misuse of commercial drones for illicit activities poses significant challenges in their detection and identification. Existing methods, such as acoustic-based, radio frequency-based, and computer vision approaches, face limitations due to factors like miniaturization, stealth, and background noise. In this paper, we propose E-Argus, a system that leverages the electromagnetic radiation (EMR) emitted by the memory of drones. It is a basic fact that, with all types of drones, the implementation of arbitrary behavior must be digested in the built-in memory, and electromagnetic radiation is thus generated. Specifically, the memory clock drives the switching regulator causing current fluctuations that generate EMR signals at the clock frequency. E-Argus combines the relationship between the flight pattern of the drone and the memory EMR signal, analyzes the unique side-channel signatures, and utilizes advanced neural network-based identification; E-Argus can accurately detect and identify various types of illegal drones. We designed a system prototype based on USRP B210 and conducted experiments in a wide range of scenarios. The evaluation shows that E-Argus has low latency, high accuracy, and robustness in real environments.

Accurate perception of intelligent vehicles is critical for the safe operation of autonomous vehicles. However, current perception methods often struggle to effectively detect intelligent vehicles when obstacles block their field of view. Collaborative perception, although attracting considerable attention, presents challenges in terms of privacy and data trust. In this study, we present a novel design for Enhancing Intelligent Vehicle (), a cost-effective and comprehensive perception system for intelligent vehicles. We discovered that during the process of memory caching raw sensing data in the intelligent vehicle’s system-on-chip (SOC), continuous fluctuating currents inside the memory result in the emission of Electromagnetic Radiation (EMR). As a result, intelligent vehicles actively expose themselves on the electromagnetic spectrum. is based on a set of specially designed antenna arrays that scan the spectrum and utilize a joint Kalman filtering algorithm to enhance EMR signals. The micro-Doppler signature of each EMR signal is then analyzed to identify signals from intelligent vehicles and construct a vehicle database. A multi-antenna joint estimation algorithm is also designed to further estimate the position, distance, and direction of the target vehicle. Our experiments demonstrate that offers advantages in terms of timeliness, robustness, and accuracy.

Head motion tracking is a promising research field with vast applications in ubiquitous human-computer interaction (HCI) scenarios. Unfortunately, solutions based on vision and wireless sensing have shortcomings in user privacy and tracking range, respectively. To address these issues, we propose IA-Track, a novel head motion tracking system that combines inertial measurement units (IMU) and acoustic sensing. Our wireless earphone-based method balances flexibility, computational complexity, and tracking accuracy, requiring only an earphone with an IMU and a smartphone. However, we still face two challenges. First, wireless earphones have limited hardware resources, making acoustic Doppler effect-based method unsuitable for acoustic tracking. Second, traditional Kalman filter-based trajectory restoration methods may introduce significant cumulative errors. To tackle these challenges, we rely on IMU sensor data to recover the trajectory and use smartphones to emit ”inaudible” acoustic signals that the earphone receives to adjust the IMU drift track. We conducted extensive experiments involving 50 volunteers in various potential IA-Track usage scenarios, demonstrating that our well-designed system achieves satisfactory head motion tracking performance.

Blink detection is essential for various human-computer interaction scenarios, such as virtual reality and driving state detection. It has gained significant attention from industry and academia alike in recent years. Existing non-contact detection systems (cameras, acoustics, etc.) have made significant progress, but various issues have prevented their widespread adoption, including privacy concerns, line-of-sight requirements, and cost issues. Therefore, there is a critical need for a simple and robust system that can detect eye blinks using common commercial equipment. In this paper, we propose BlinkRadar, which uses a low-cost customized impulse-radio ultra- wideband (IR-UWB) radar for non-contact and fine-grained blink detection. BlinkRadar can reliably detect driver blinks in driving conditions, making it possible to infer drowsy driving. To effectively extract the eye blink signal, we analyzed real experimental data to study the characteristics of the eye blink pattern and successfully used the multi-sequence variational mode decomposition (MS-VMD) algorithm to separate the blink signal from the noise signal. We conducted extensive experiments in two different environments (a quiet room and moving vehicles) and found that BlinkRadar had an average blink detection accuracy of over 96.2%. Our results demonstrate the feasibility of using UWB radar for non-contact eye blink detection.

Passenger counting is crucial for many applications such as vehicle scheduling and traffic capacity assessment. However, most of the existing solutions are either high-cost, privacy invasive or not suitable for passengers the vehicle scenarios. In this work, we propose thePa-Count, an effective real-timePassengerCounting system deployed inside the vehicle via using Wi-FiCSI(Channel State Information). Specifically, in Pa-Count, we design a set of combined filters to eliminate environmental interference and enhance CSI quality. In so doing, we can identify the fluctuation of weak CSI caused by passengers’ subtle movement, i.e., the fidgeting, and then obtain the distribution of fidgeting period and silent period. Following that, we describe the subtle movements of passengers via power law with exponential cutoff distribution and establish a counting model based on the queuing theory. A mathematical inference method with a priori probability is devised to calculate the number of real-time passengers through CSI. We evaluate the performance of the Pa-Count by conducting a set of experiments in real-world vehicle scenarios (including private car and subway). Experimental results show that Pa-Count can achieve robust performance with an average accuracy of over 92$\%$.

The contact-free sensing nature of Wi-Fi has been leveraged to achieve privacy breaches such askeystroke inference(KI). However, the use ofchannel state information(CSI) in existing attacks is highly questionable due to its signal instability and hardness to acquire. Moreover, such Wi-Fi-based attacks are confined to only one victim because Wi-Fi sensing offers insufficient range resolution to physically differentiate multiple victims. To this end, we propose MuKI-Fi to enable, for the first time,multi-personKI, leveragingbeamforming feedback information(BFI), a new feature offered by latest Wi-Fi hardware, transmitted in clear-text by smartphones. BFI's characteristics, clear-text communication and signal stability, make it readily acquirable and usable by any other Wi-Fi devices switching to monitor mode without the need forlow-levelhacking on hardware. Moreover, to improve upon existing KI methods offering very limited generalizability across diversified application scenarios, MuKI-Fi innovates in an adversarial learning scheme to enable its inference generalizable towards unseen scenarios. Finally, we discover that, as a smartphone is in close proximity to a victim, the variations of BFI caused by that victim's keystrokes in suchnear-fieldsubstantially outweigh those caused by other distant victims; this phenomenon naturally allows for multi-person KI. Our extensive evaluations clearly demonstrate that MuKI-Fi can effectively eavesdrop on the keystrokes of multiple subjects, achieving 87.1% accuracy for individual keystrokes and up to 81% top-100 accuracy for stealing passwords from mobile applications(e.g., WeChat) on average.

Acoustic target tracking has shown great advantages for device-free human-machine interaction over vision/RF-based mechanisms. However, existing approaches for portable devices solely track a single target, incapable of the ubiquitous and highly challenging multi-target situations such as double-hand multimedia controlling and multi-player gaming. In this paper, we proposeAMT$^+$, a pioneering smartphone MIMO system to achieve centimeter-level multi-target tracking. The challenge of multi-target occlusion is effectively addressed by employing multiple speaker-microphone pairs. However, the unique challenge raised by MIMO is the superposition of multi-source signals due to the cross-correlation among speakers. Initially, we tackle this challenge by designing a weak cross-correlation signal to reduce interference passively. InAMT$^+$, we’ve further integrated self-interference cancellation for active minimize interference. The most distinguishing advantage ofAMT$^+$lies in the elimination of the raised multipath effect, which is commonly ignored in previous work by hastily assuming targets as particles.AMT$^+$employs Doppler filtering over delay subtraction for echo suppression. Further, by non-particle target reflections modeling results, we introduce a distance-projection-based method for continuous target identification and tracking. Implemented on commercial smartphones,AMT$^+$achieves on average 0.54 cm, 1.37 cm, and 2.13 cm errors for single, double, and triple target tracking respectively, and on average 97.0% classification accuracy for 14 controlling gestures.

The contact-free sensing nature of Wi-Fi has been leveraged to achieve privacy breaches, yet existing attacks relying on Wi-Fi CSI (channel state information) demand hacking Wi-Fi hardware to obtain desired CSIs. Since such hacking has proven prohibitively hard due to compact hardware, its feasibility in keeping up with fast-developing Wi-Fi technology becomes very questionable. To this end, we propose WiKI-Eve to eavesdrop keystrokes on smartphones without the need for hacking. WiKI-Eve exploits a new feature, BFI (beamforming feedback information), offered by latest Wi-Fi hardware: since BFI is transmitted from a smartphone to an AP in clear-text, it can be overheard (hence eavesdropped) by any other Wi-Fi devices switching to monitor mode. As existing keystroke inference methods offer very limited generalizability, WiKI-Eve further innovates in an adversarial learning scheme to enable its inference generalizable towards unseen scenarios. We implement WiKI-Eve and conduct extensive evaluation on it; the results demonstrate that WiKI-Eve achieves 88.9% inference accuracy for individual keystrokes and up to 65.8% top-10 accuracy for stealing passwords of mobile applications (e.g., WeChat).

Middle ear effusion is a common symptom of otitis media, the reactive physical manifestation of otitis media (OM) in children's middle ear. However, diagnosing MEE for little children at home is troublesome due to their difficulty cooperating and the caregiver's lack of medical knowledge. To this end, we propose EarSonar, a novel acoustic-based MEE diagnostic system. The principle behind EarSonar is that the acoustic absorption effect exists in ear scenarios, and the volume of middle ear fluid can markedly affect the absorbed spectrum energy. By automatically eliminating the impact of potential interference factors and identifying the representative frequency range with the typical reaction of acoustic absorption, EarSonar captures fine-grained signal features on absorbed spectrum energy and models the intrinsic relationship between acoustic absorption and the volume of the filler fluid in the eardrum. On that basis, EarSonar extracts the features of the MEE signal segment and uses k-means clustering to classify middle ear effusion status. We conducted a test on 112 adolescents aged 4–6. We divided the degree of middle ear effusion into three grades. The final average detection accuracy rate exceeds 92%, which is 8 % higher than the previous method. We have implemented a proof-of-concept prototype of EarSonar by building upon earphones embedded with a microphone and speaker. Experimental results demonstrate a feasible and effective way to turn earphones into potential home-use MEE screening tools.

The eye-blink pattern is crucial for drowsy driving diagnostics, which has become an increasingly serious social issue. However, traditional methods (e.g., with EOG, camera, wearable, and acoustic sensors) are less applicable to real-life scenarios due to the disharmony between user-friendliness, monitoring accuracy, and privacy-preserving. In this work, we design and implement BlinkRadar as a low-cost and contact-free system to conduct fine-grained eye-blink monitoring in a driving situation using a customized impulse-radio ultra-wideband (IR-UWB) radar which has superior spatial resolution with the ultra-wide bandwidth. BlinkRadar leverages an IR-UWB radar to achieve contact-free sensing, and it fully exploits the complex radar signal for data augmentation. BlinkRadar aims to single out the eye-blink induced waveforms modulated by body movements and vehicle status. It solves the serious interference caused by the unique characteristics of blinking (i.e., subtle, sparse, and non-periodic) and from the human target itself and surrounding objects. We evaluate BlinkRadar in a laboratory environment and during actual road testing. Experimental results show that BlinkRadar can achieve a robust performance of drowsy driving with a median detection accuracy of 92.2% and eye blink detection of 95.5%.