Occupancy Detection Using a Frequency-Modulated Phase- and Self-Injection-Locked (FMPSIL) Radar and Two-Stage Generalized Likelihood Ratio Test (GLRT) Algorithm

• Published in: IEEE transactions on microwave theory and techniques Vol. 72; no. 8; pp. 4908 - 4918
• Main Authors: Shih, Ju-Yin, Zhong, Ji-Xun, Wang, Fu-Kang
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.08.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Computational efficiency; Continuous wavelet transforms; Cost analysis; Detection algorithms; Fast Fourier transformations; Frequency conversion; Frequency locking; Frequency modulation; Frequency-modulated continuous wave (FMCW); generalized likelihood ratio test (GLRT); Injection-locked oscillators; Likelihood ratio; Moving targets; occupancy detection; Phase frequency detectors; phase- and self-injection-locked (PSIL) radar; Probability density function; probability density function (pdf); Probability density functions; Radar detection; Sensors; Signal detection; Target detection; Temperature sensors; Time domain analysis
• ISSN: 0018-9480; 1557-9670
• DOI: 10.1109/TMTT.2023.3349154

This study proposed a frequency-modulated phase- and self-injection-locked (FMPSIL) radar for occupancy detection. Compared to the conventional range-Doppler map, which performs occupancy detection by obtaining the vital sign of the subject using 2-D fast Fourier transform (2-D-FFT), this study proposed a two-stage generalized likelihood ratio test (GLRT) algorithm to overcome the leakage problem in the spectrum and increase occupancy detection accuracy. The use of the probability density function (pdf)-based GLRT analysis in the time domain reduced the required computational cost and observation time. To demonstrate the feasibility of the proposed two-stage GLRT algorithm, experiments were conducted for a 600-min dataset in a cluttered environment, the algorithm achieved a detection accuracy of 97.32% at an observation time of 5 s for a single moving target, and multitarget occupancy detection was also achieved.