本文采用最大相关熵准则(MCC)对投影近似子空间跟踪(PAST)算法中基于最小平方误差(MSE)准则的目标函数进行修正,推导出适用于冲激噪声环境的韧性投影近似子空间跟踪新算法(MCC-PAST算法).对两种冲激噪声模型包括,稳定分布噪声模型和混合高斯噪声模型环境下的时变波达方向估计问题的实验仿真表明,与基于韧性M估计的PAST改进算法(RLM-PAST算法)相比,MCC-PAST算法可以自适应地调整核长,对于阵列接收数据的变化体现出更好的适应性.
By employing the maximum correntropy criterion(MCC)based cost function in projection approximation subspace tracking(PAST) algorithm, we deduce the MCC_ PAST algorithm which can be utilized for the projection approximation subspace tracking under impulsive noise environments. We examine the performance of our proposed algorithm in the time-varying DOA(Di- rection Of Arrival)estimation problem under two types of impulsive noise environments, including the alpha-stable noise environ- ments and the contaminated Gaussian noise environments. By adjusting the kernel parameter adaptively in the algorithm, MCC_ PAST exhibits superiority over the robust M-estimate based PAST algorithm, especially when there are sudden changes in the array outputs.