Multi-Frame Coherent Integration for Detecting High-Speed Manoeuvring Target in Bistatic Space-Based Early Warning Radar

Abstract

Long-time coherent integration of high-speed manoeuvring weak targets is critical for enhancing detection performance in bistatic space-based early warning radar systems. The signal-to-clutter-plus-noise ratio after single-frame coherent integration often remains insufficient for reliable detection, motivating the need for multi-frame processing. However, multi-frame long-time coherent integration is severely challenged by several factors: high-speed target manoeuvring induces range and Doppler cell migration (RCM/DCM), variations in pulse repetition frequencies across frames lead to phase inconsistencies and strong clutter environments complicate effective coherent processing. To overcome these challenges, this paper proposes a joint intra-frame and inter-frame integration framework. Firstly, space-time adaptive processing is applied for clutter suppression, followed by Keystone transform and matched filter processing to correct intra-frame RCM and DCM simultaneously. Subsequently, an adaptive phase estimation algorithm is proposed in which the initial phase of each frame is estimated from the dominant eigenvector of the covariance matrix formed by the echo data. Based on this, a phase-compensated generalised Radon transform technique is developed to realise coherent phase alignment and integration across frames. Simulation results show that the proposed method achieves near-optimal coherent gain and significantly improves the detection performance for weak targets.