In hostile environment, Global Navigation Satellite System (GNSS) could be disturbed by intentional jamming. Many adaptive algorithms have been developed to deal with these threats, among which use of antenna arrays is one of the most efficient. However, most of them have been designed under stationary hypothesis and their performances in harsher environments are questionable. For instance, when a GNSS receiver is placed near rotating bodies, the signal undergo complex and non-stationary effects called Rotor Blade Modulation (RBM). These variations can degrade significantly anti-jamming performance. This paper investigates the impact of the RBM on three conventional space-time adaptive processing (STAP). First, to simulate the RBM, the signal received by an antenna mounted on a helicopter is computed thanks to electromagnetic (EM) asymptotic methods. Then, to quantify precisely the loss in performance of each algorithm, we compare post correlation carrier to noise ratio (post - C/N0) and covariance matrix estimation with respect of the time. Finally, the simulation results are confirmed by experiments conducted on an EC-120 helicopter with an L-band Continuous Wave (CW) jammer.
