Mobi
            <sup>2</sup>
            Sense

Zhang, Fusang; Xiong, Jie; Chang, Zhaoxin; Ma, Junqi; Zhang, Daqing

doi:10.1145/3495243.3560518

Cited by 27 publications

(2 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Active RF sensing technologies may compromise their main function. For example, WiFi, one of the most popular human sensing technologies, is unsuitable for human sensing applications in a real WiFi communication environment due to unpredictable back-offs and packet sizes [12], [13]. 2.…”

Section: B Motivation and Proposed Solutionmentioning

confidence: 99%

Human Sensing via Passive Spectrum Monitoring

Mu,

Yuan,

2023

IEEE Open J. Instrum. Meas.

View full text Add to dashboard Cite

Human sensing is significantly improving our lifestyle in many fields such as elderly healthcare and public safety. Research has demonstrated that human activity can alter the passive radio frequency (PRF) spectrum, which represents t he passive reception of RF signals in the surrounding environment without actively transmitting a target signal. This paper proposes a novel passive human sensing method that utilizes PRF spectrum alteration as a biometrics modality for human authentication, localization, and activity recognition. The proposed method uses software-defined radio (SDR) technology to acquire the PR F in the frequency band sensitive to human signature. Additionally, the PRF spectrum signatures are classified and regressed by five machine learning (ML) algorithms based on different human sensing tasks. The proposed Sensing Humans among Passive Radio Frequency (SHAPR) method was tested in several environments and scenarios, including a laboratory, a living room, a classroom, and a vehicle, to verify its extensiveness. The experimental findings demonstrate that the SHAPR system, in conjunction with the random forest (RFR) algorithm, achieves human authentication accuracies of 95.6 % and 98.7 % in laboratory and livin g room scenarios, respectively. In a vehicular settin g, grid-level localization accuracy reaches 99.1 %, and in a laboratory environment, activity recognition accuracy is attained at 99.1 %. Moreover, within a classroom scenario, the SHAPR system, when integrated with the Gaussian process regression (GPR ) model, can realize coordinate-level localization with an error margin of merely 0.8 meters. These results indicate that the SHAPR technique can be considered a new human signature modality with high accuracy, robustness, and general applicability.

show abstract

Section: B Motivation and Proposed Solutionmentioning

confidence: 99%

Human Sensing via Passive Spectrum Monitoring

Mu,

Yuan,

2023

IEEE Open J. Instrum. Meas.

View full text Add to dashboard Cite

show abstract

“…M Illimeter-wave (mmWave) imaging has enormous potential in various applications, including security checks [1], human sensing [2][3][4], indoor mapping [5,6], and autonomous driving [7][8][9]. Its advantages of being penetrative, light-robust, and non-ionizing radiated, make it a highly promising imaging modality compared to optical cameras and X-rays [10][11][12][13][14][15]. In addition, existing works have demonstrated the potential of mmWave imaging in detecting skin cancer [16], providing a non-intrusive approach to medical diagnosis.…”

Section: Introductionmentioning

confidence: 99%

DI-Gesture: Domain-Independent and Real-Time Gesture Recognition with Millimeter-Wave Signals

Zhang

Chen

et al. 2022

GLOBECOM 2022 - 2022 IEEE Global Communications Conference

View full text Add to dashboard Cite

Recent advancements have showcased the potential of handheld millimeter-wave (mmWave) imaging, which applies synthetic aperture radar (SAR) principles in portable settings. However, existing studies addressing handheld motion errors either rely on costly tracking devices or employ simplified imaging models, leading to impractical deployment or limited performance. In this paper, we present IFNet, a novel deep unfolding network that combines the strengths of signal processing models and deep neural networks to achieve robust imaging and focusing for handheld mmWave systems. We first formulate the handheld imaging model by integrating multiple priors about mmWave images and handheld phase errors. Furthermore, we transform the optimization processes into an iterative network structure for improved and efficient imaging performance. Extensive experiments demonstrate that IFNet effectively compensates for handheld phase errors and recovers high-fidelity images from severely distorted signals. In comparison with existing methods, IFNet can achieve at least 11.89 dB improvement in average peak signal-to-noise ratio (PSNR) and 64.91% improvement in average structural similarity index measure (SSIM) on a real-world dataset.

show abstract