基于线性高斯滤波的反干扰跟踪方法

A Method for Anti-Jamming Tracking Based on Linear Gauss Filtering

以往机动目标的跟踪问题大多是针对确定性系统,而对随机跳变系统的研究较少。针对目标随机施放干扰的情况,将线性高斯滤波应用于观测噪声中带有尖头干扰信号的系统中,实现机动目标的反干扰跟踪。其算法是一种基于不同模型间"软切换"的机动目标跟踪方法,用计算的概率权值对这些模型输出进行综合,保证了跟踪精度,大大降低了离散时间结构随机跳变系统最优滤波算法的复杂程度。通过仿真实例可以看出,在观测噪声特性发生剧烈随机跳变的情况下,线性高斯滤波算法对机动目标进行了比较准确的跟踪,其性能显著地优于标准的卡尔曼滤波算法。

Most of former studies on maneuvering target tracking are mainly taken for deterministic system, with few on random changing structures. We use linear Gauss filtering in the system for observing the noise with jamming signals, thus to realize anti-jamming tracking of maneuvering target when the target dispenses jamming randomly. This algorithm is a maneuvering target tracking algorithm based on ＂soft switchover＂ of different models. The obtained probability weights are used to integrate the output of these models, which can ensure the tracking accuracy, and greatly reduces the complexity of discrete time optimal filtering of system with random changing structures. The result of simulation shows that linear Gauss filtering can realize tracking of maneuvering targets exactly, its performance is better than Kalman filtering in the case that the property of observed noise changes severely.