The ability to track, detect and monitor human activities in real time conditions is very important for security and surveillance operation. The movements of the human body and limbs generate unique microDoppler features which enable the identification and classification of a wide diversity of human motions. In this paper, we propose a method to simulate the signal generated by human motion, by modeling human movement using equation for each limb. The simulation method is tested on five activities: walking in circle, falling, walking straight, punching, and making squats. The simulated signals are compared with measurement results. The results end up being satisfying.