An Autoregressive Steady-State Compensation Method for Cross-Correlation Interference Suppression in GPS-Based Passive Radar

GPS-based passive bistatic radar (PBR) benefits from global satellite coverage for target surveillance. However, multiple GPS satellites within the PBR mainlobe generate cross-correlation interference (CCI) that severely masks target echoes, reducing the detection probability to zero across significant portions of the surveillance area. Existing reconstruction-based suppression methods rely on iterative frequency estimation, which introduces substantial errors during the convergence stage of the tracking loop, leading to degraded interference suppression performance. This paper proposes an autoregressive steady-state compensation (ARSSC) method to address this limitation. First, a precise carrier frequency estimation model is established to accelerate convergence and improve tracking accuracy. Second, the frequency estimation outputs are partitioned into convergence and steady-state stages, and a p-th order autoregressive (AR) model is fitted to the steady-state estimates. A compensation function is then derived from the AR model to correct the frequency errors in the convergence stage. Finally, the compensated reconstructed CCI signals are used to construct an interference subspace, and a projection-based algorithm suppresses the CCI from the surveillance signal. Simulation results demonstrate that the proposed ARSSC method achieves a maximum interference suppression improvement of 7.4 dB compared to conventional reconstruction approaches. Real-data experiments conducted under different field scenarios further validate the method, yielding a 6.3 dB interference suppression ratio (ISR) improvement over traditional reconstruction techniques in both tested cases.