认知雷达波形和检测门限自适应跟踪算法

An Adaptive Tracking Algorithm Based on Waveform and Detection Threshold of Cognitive Radars

为有效改善杂波背景下雷达目标跟踪中的不确定性问题,提出一种基于认知雷达的波形和检测门限联合自适应跟踪算法。首先,给出雷达目标跟踪的信息熵不确定模型,提出用滤波误差协方差的行列式来描述目标状态跟踪的不确定性;然后,基于时延-多普勒分辨单元理论,给出单个高斯线性调频脉冲包含波形和虚警概率参数的量测误差协方差的近似表达式;最后,受人类"感知-行动"循环机制启发,设计出基于信息熵最小为代价函数的自适应波形和检测门限跟踪算法。仿真结果表明,所提出的联合自适应算法的信息熵比目标跟踪的不确定性传统算法可减少30%以上,总体跟踪性能得到明显改善。

A joint adaptive tracking algorithm based on waveform and detection threshold of cognitive radars is proposed to effectively improve the uncertainty of radar target tracking in cluttered background. Firstly, an uncertainty model for the information entropy of radar target tracking is given, and the determinant of the covariance of filtering errors is proposed to describe the uncertainty of target state tracking. Then, an approximation of measurement error covariance of a single Gaussian linear frequency modulation(LFM) pulse with waveform and false alarm probability parameters is given based on the theory of time delay-Doppler resolution cell. Finally, inspired by human 'perception-action' cycle mechanism, an adaptive waveform and detection threshold tracking algorithm is designed based on minimizing the information entropy as a cost function. Simulation results and a comparison with the traditional algorithm show that, from the point of view of information entropy, the proposed joint adaptive algorithm reduces the uncertainty of target tracking by more than 30%, and the overall tracking performance is significantly improved.