Very-Large-Scale Integration-Friendly Method for Vital Activity Detection with Frequency-Modulated Continuous Wave Radars DOI Creative Commons
Krzysztof S̀lot, Piotr Łuczak, Paweł Kapusta

и другие.

Sensors, Год журнала: 2025, Номер 25(7), С. 2151 - 2151

Опубликована: Март 28, 2025

A simple algorithm for respiratory activity detection in data produced by Frequency-Modulated Continuous-Wave (FMCW) radars is presented this paper. The proposed computational architecture can be directly mapped onto custom digital-analog VLSI hardware, which a unique approach research on intelligent FMCW sensor development, offering potential energy-efficient analysis solution target applications, such as preventing human trafficking or providing life-sign under limited visibility. comprises two main modules. first one summarizes radar-produced into descriptor reflecting the amount of motion that occurs within appropriately determined time intervals. second classifies sequence descriptors using recurrent neural network composed gated units. To ensure algorithm's implementation feasibility, an analog circuit comprising its functional blocks has been designed, manufactured, and tested, constraints model derivation. adverse effects primary constraint, severe restriction admissible weight resolution, have handled introducing novel training loss component mechanism diversifying effective sets different neurons. Experimental evaluation method, performed dataset indoor recordings, indicates simple, hardware implementation-friendly provides over 94% accuracy similar F1 scores.

Язык: Английский

Motion Cancellation Technique of Vital Signal Detectors Based on Continuous-Wave Radar Technology DOI Creative Commons

Min‐Seok Kwon,

Yuna Park, Jinhee Park

и другие.

Sensors, Год журнала: 2025, Номер 25(7), С. 2156 - 2156

Опубликована: Март 28, 2025

Continuous-wave (CW) radar sensors can remotely measure respiration and heartbeat by detecting the periodic movements of internal organs. However, external disturbances, such as random body motion (RBM) or environmental interference, significantly degrade signal-to-noise ratio (SNR) reduce accuracy vital sign detection. The various cancellation techniques that have been proposed to enhance robustness against RBMs include improvements in architecture, advanced signal processing algorithms, studies on electromagnetic propagation characteristics. This paper provides a comprehensive review recent advancements for CW radar-based detectors discusses future research directions improve detection performance dynamic environments.

Язык: Английский

Процитировано

0
Very-Large-Scale Integration-Friendly Method for Vital Activity Detection with Frequency-Modulated Continuous Wave Radars DOI Creative Commons
Krzysztof S̀lot, Piotr Łuczak, Paweł Kapusta

и другие.

Sensors, Год журнала: 2025, Номер 25(7), С. 2151 - 2151

Опубликована: Март 28, 2025

A simple algorithm for respiratory activity detection in data produced by Frequency-Modulated Continuous-Wave (FMCW) radars is presented this paper. The proposed computational architecture can be directly mapped onto custom digital-analog VLSI hardware, which a unique approach research on intelligent FMCW sensor development, offering potential energy-efficient analysis solution target applications, such as preventing human trafficking or providing life-sign under limited visibility. comprises two main modules. first one summarizes radar-produced into descriptor reflecting the amount of motion that occurs within appropriately determined time intervals. second classifies sequence descriptors using recurrent neural network composed gated units. To ensure algorithm's implementation feasibility, an analog circuit comprising its functional blocks has been designed, manufactured, and tested, constraints model derivation. adverse effects primary constraint, severe restriction admissible weight resolution, have handled introducing novel training loss component mechanism diversifying effective sets different neurons. Experimental evaluation method, performed dataset indoor recordings, indicates simple, hardware implementation-friendly provides over 94% accuracy similar F1 scores.

Язык: Английский

Процитировано

0