非高斯噪声下的滤波方法

干扰条件下的机动目标跟踪在一些文献[1][2]中已有讨论，但利用多传感器，尤其是被动传感器进行非高斯观测噪声条件下的目标跟踪仍需要研究。本文讨论了被动传感器在随机干扰条件下进行机动目标跟踪的方法，其观测量包含非高斯噪声，也可能包含影响观测值的随机干扰。与基于卡尔曼滤波的常见方法不同，采用动态规划算法进行多假设检验，从而估计目标的状态。仿真试验表明，本文方法能有效地处理非高斯噪声情况下的目标跟踪问题，而基于卡尔曼滤波的跟踪方法，比如EKF，则效果较差。

Filtering Method with Non-Gaussian noise

Tracking a maneuvering target in a nonlinear interference environment has been discussed in some literature, and this paper considers the problem of tracking with multiple passive sensors in the non- Gaussian noise environment. The measurements possibly include random interference and Glint noise. The method of target state estimation in this paper is a dynamic programming approach. Unlike the traditional method, we reduce the estimating problem to a multiple hypothesis-testing problem, and then use the dynamic programming a1gorithm to solve the problem of tracking with non-Gaussian noise. The simulation results show the superiority of the new method.