Single-Channel FMCW-Radar-Based Multi-Passenger Occupancy Detection Inside Vehicle

Abstract

In this paper, we provide the results of multi-passenger occupancy detection inside a vehicle obtained using a single-channel frequency-modulated continuous-wave radar. The physiological characteristics of the radar signal are analyzed in a time-frequency spectrum, and features are proposed based on these characteristics for multi-passenger occupancy detection. After clutter removal is applied, the spectral power and Wiener entropy are proposed as features to quantify physiological movements arising from breathing and heartbeat. Using the average means of both the power and Wiener entropy at seats 1 and 2, the feature distributions are expressed, and classification is performed. The multi-passenger occupancy detection performance is evaluated using linear discriminant analysis and maximum likelihood estimation. The results indicate that the proposed power and Wiener entropy are effective features for multi-passenger occupancy detection.

Machine Learning in RADAR-based Physiological Signals Sensing: A Scoping Review of the Models, Datasets and Metrics

“…1) Human detection: We found 7 papers dealing with human detection aided with vital sign monitoring through radars. We can notice three main research lines across human detection papers being occupancy detection for vehicle applications [187], [191], [192], human detection underground [189], [190] and human detection in indoor environments [37], [188].…”

“…Passenger detection is relevant also for the efficiency in energy consumption and comfort of the passengers for a controlled heating in each seat [246]. Three papers implement passenger occupancy detection [187], [191], [192] with all different radar technology, being CW, FMCW and UWB, respectively. The sample size for these papers is limited to three subjects across different scenarios.…”

In-Vehicle Monitoring by Radar: A Review

“…Both resistive 2 Range-doppler map or timefrequency spectrum Artificial intelligence to detect occupied seats [3], [11]- [34] The amplitude of reflected signals Left-behind child [2], [35]- [48] BR and HR difference BR and HR estimation [49]- [51] Gesture recognition to assist drivers Micro-doppler features Artificial intelligence to detect gestures [52]- [59] Occupant status monitoring BR and/or HR estimation None [60]- [77] Sensor placement for accurate BR estimation [78]- [80] Vital sign monitoring in an ambulance [81] Drowsy driving detection [82]- [87] Biometric driver seat [79] Angry driver [88] Multiple targets vital sign monitoring [89], [90] Car vibrations suppression [62], [91]- [93] Changes in the reflected power Distracted driver detection by cellphone [82] Random body movement cancellation [62], [80], [94] Airbag [95] Range-doppler map Distracted/drowsy driver based on head motion [96]- [101] Range doppler map, Changes in the phase of signals Drowsy driver based on eye blink frequency [102]- [107] and inductive sensors have difficulty discriminating between humans and objects [3]. Capacitive sensors which can detect the dielectric dispersion effects on human tissues are prone to high false detections [18]. Camera vision [108], [109], and infrared (IR) sensors [110] are also commonly ...…”

Characterization Technique for a Doppler Radar Occupancy Sensor

“…The installation cost, privacy concerns, lag times, or persistent inability to detect sedentary individuals has prevented sensors that can truly detect the presence of sedentary occupants from penetrating the market. However, recent developments in Doppler-radar occupancy sensing are promising because they detect breathing motion [10,12,13] and are being shown to reliably detect the presence of sedentary occupants in realistic settings [14], including in vehicle cabins [20][21][22].…”