Research on long-term coherent accumulation algorithm for high-speed targets

With the development of technology, high-speed targets with characteristics such as fast speed, high maneuverability, and strong stealth capabilities pose great challenges to radar detection performance. In recent years, high-speed target detection has received much attention in the field of radar signal processing. Increasing the time of observation can effectively improve the echo signal-to-noise ratio and the radar's ability to detect weak targets. However, increasing the time of the target's high-speed and maneuverability for observation can lead to range migration and doppler frequency migration, rendering traditional coherent accumulation algorithms ineffective. This article proposes the SKT-HWVD-MCICPF (Second-order Keystone Transform, High-order Wigner-ville Distribution and Modified Coherent Integration Cubic Phase Function) coherent accumulation algorithm for a target model that undergoes variable acceleration motion along the radial direction of the radar. The algorithm solves the problems of range migration and range curvature correction and high-order doppler frequency migration compensation for variable acceleration motion targets. A third-order WVD (Wigner-ville Distribution) matrix is constructed to accurately estimate the third-order phase coefficient, reducing the problem to a second-order phase coefficient estimation problem. The MCICPF (Modified Coherent Integration Cubic Phase Function) algorithm effectively suppresses cross term interference and improves parameter estimation performance.