Using Micro-Doppler radar signals for human gait detection

[摘要] ENGLISH ABSTRACT: This work entails the development and performance analysis of a human gaitdetection system based on radar micro-Doppler signals. The system consistsof a tracking functionality and a target classifier. Target micro-Dopplersignatures are extracted with Short-Time Fourier Transform (STFT) basedspectrogram providing a high-resolution signatures with the radar that isused. A feature extraction mechanism is developed to extract six featuresfrom the signature and an artificial neural network (A-NN) based classifier isdesigned to carry out the classification process. The system is tested on realX-band radar data of human subjects performing six activities. Those activitiesare walking and speed walking, walking with hands in pockets, marching,running, walking with a weapon, and walking with arms swaying. The multiclassclassifier was designed to discriminate between those activities. Highclassification accuracy of 96% is demonstrated.

[发布机构] Stellenbosch University