An anti-RGPO approach based on GRU and Kalman filtering
X X Li, Feng Li, Chang Xiong, Yang Li · IET conference proceedings. · 2026
With the rapid advancement of modern electronic warfare technologies, radar systems face increasingly severe threats from deception jamming, particularly Range Gate Pull-Off (RGPO) jamming which significantly compromises target tracking accuracy. Traditional anti-jamming approaches primarily relying on signal feature analysis or data-level processing demonstrate limited effectiveness in complex electronic warfare environments. This paper proposes a dynamic trajectory selection network based on Gated Recurrent Unit (GRU), integrated with a Kalman filter post-processing framework, to achieve effective suppression of RGPO jamming. By constructing a radar observation dataset encompassing three operational phases including reconnaissance-silence, cooperative deception and forced pull-off and designing a classification head with confidence thresholding, the model enables accurate association of authentic target trajectories. Experimental results demonstrate that the proposed method achieves correct target association under pull-off jamming while the Kalman filter framework effectively predicts missing measurements, significantly enhancing tracking stability. The presented approach provides a novel solution for radar anti-jamming in complex electronic warfare environments.