基于FastICA的雷达弱目标检测前跟踪方法研究

研究了强噪声干扰下的雷达弱目标检测及跟踪问题。针对信号与噪声干扰之间的相互独立性,提出了一种基于FastICA的弱目标检测前跟踪(TBD)算法。该算法的关键为对分离后的信号进行能量分配,以及求解所构建的一个多目标规划问题,进而实现匹配回波信号的能量积累。仿真实验结果表明,在负信噪比的情况下,无论是高斯噪声还是非高斯噪声,该方法均可以实现检测前跟踪的目的;用能量积累过程中记录的目标状态信息为观测值,则可以通过滤波估计的方法最终实现对目标的稳健跟踪。该方法为弱小目标的跟踪检测技术提供了一种新的思路。

Study on tracking before detecting method based onFastICA for weak radar target

This paper studied weak radar target's detecting and tracking problems with the disturbance by strong noise. Aiming at independence between signals and noise, this paper proposed one method of TBD based on FastICA for weak target. The key points of the method was to dispute energy to signals after blind source separation BSS, and solved a designed multi-objective programming approach problem so as to achieve the energy accumulation of coincident echo wave. Simulation results show that in the case of negative signal to noise ratio SNR the proposed method can detect and track target well either it is Gaussian noise or non-Gaussian noise. Furthermore, used status information recorded during process of energy accumulation as the observed values, and then target would be stably tracked by filter estimation methods, which provided one new way for tracking and detecting technology of weak target.