Human Sensing via Passive Spectrum Monitoring

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 reception of RF signals surrounding environment without actively transmitting a target signal. This paper proposes novel method utilizes PRF spectrum alteration biometrics modality for authentication, localization, recognition. The proposed uses software-defined (SDR) technology to acquire band sensitive signature. Additionally, signatures are classified regressed by five machine learning (ML) algorithms based on different tasks. Sensing Humans among Passive Radio Frequency (SHAPR) was tested several environments scenarios, including laboratory, living room, classroom, vehicle, verify its extensiveness. experimental findings demonstrate SHAPR system, conjunction with random forest (RFR) algorithm, achieves authentication accuracies 95.6 % 98.7 laboratory room respectively. In vehicular setting, grid-level localization accuracy reaches 99.1 %, environment, recognition attained at %. Moreover, within classroom scenario, when integrated Gaussian process regression (GPR) model, realize coordinate-level an error margin merely 0.8 meters. These results indicate technique be considered new signature high accuracy, robustness, general applicability.