In this paper, the effect of the change of Pulse Repetition Frequency (PRF) on the performances of the STAP with reduced rank is discussed. The numerical evaluation is based on two different models of changing the PRF, namely quadratic and chaotic but with two methods of reduction of the rank: Principle Components (PC) and Signal to Interference Noise Ratio (SINR metric). Rank reduction reduces the computing time and cost. The simulation results show that the change of PRF solves well the problem of ambiguities while using a reduced rank of covariance matrix of the environment. Indeed, the notches of the clutter are eliminated. Also, a comparative study between the two methods, quadratic change and chaotic change, shows that the chaotic change is better and this is due to the simplicity of the generation of the impulses on one hand, while on the other hand, the chaotic change has the possibility of taking a large variety of PRFs at the same time. In fact, we use the properties of a chaotic system, such as sensitivity of initial conditions, mixture, and density of periodic points to sweep all the space. The changes of frequencies used are numerous, and are generated (according to an equation modelling the environment).