The pedestrian micro-Doppler observed by terahertz radar can be used for pedestrian detection and recognition, which can be widely applied in the field of unmanned vehicle and intelligent driving. In this paper, a 292 GHz terahertz radar is utilized to observe micro-Doppler caused by the swing of pedestrians’ limbs and trunk by setting up the transmitting and receiving environment of 292 GHz terahertz signal. In this experiment, the baseband signal is generated by vector signal generator, and then the baseband signal is up-converted to 292 GHz. The echo signals of the body parts are down-converted to baseband signal. The echo signal is processed by short-time Fourier transform to extract the micro-Doppler spectrum. The experimental results of human micro-Doppler characteristics extracted at 292 GHz are consistent with the simulation results. At the same time, the micro-Doppler results of 292 GHz and 24 GHz radars are compared.