基于分数低阶统计量的卡尔曼滤波目标跟踪算法

摘要

针对杂波干扰环境中的非高斯特性,发现海杂波噪声、闪烁噪声等具有显著尖峰的非高斯噪声,可以采用α稳定分布建立更符合实际的噪声模型.根据统计信号处理最新理论和技术,利用p阶分数相关和分数低阶协方差替代传统相关和协方差来改进Kalman滤波器,优化获得基于分数低阶统计量的Kalman滤波目标跟踪新算法(FLOS-KF).仿真验证了FLOS-KF滤波跟踪新算法可以更好地适应非高斯复杂环境,得到稳健的跟踪效果.%Aimed at non-Gaussian characteristics in clutter jamming environment, this paper studies significant peak non-Gaussian noise for sea clutter noise and glint noise described by alpha stable distribution,establishing noise model accord with reality. According to the lastest theory and technology of statistics signal processing, make use of p order fractional associated and fractional lower order covariance to modify Kalman filter, optimized fractional lower order covariance Kalman filter new algorithm, which is introduced to the target tracking for complicated non-Gaussian situation and achieve the solid tracking effect.