低信扰比下粒子权重优化的粒子滤波算法

针对低信扰比条件下粒子权重有效评价问题，本文给出了一种粒子权重优化的粒子滤波算法。在算法实现中，首先，通过代价评估粒子滤波中代价函数和风险函数的引入实现粒子权重评价过程中对于当前量测信息的合理利用；其次，通过置信度距离和置信度矩阵的构建及求解完成对于粒子间蕴含冗余和互补信息的充分提取；最终，利用权重平衡因子在融合两种权重度量结果基础上实现粒子权重的合理度量。新算法在实现当前时刻粒子集中信息有效利用的同时，避免了量测噪声先验统计信息的偏差的不利影响，从而使得粒子权重度量结果更加稳定和可靠。理论分析和仿真实验验证了算法的有效性。

A novel particle filter based on particle weights optimization in low signal to interference ratio

Aiming at the effective measure of particle weights in low signal to interference ratio, a novel particle filter based on particle weights optimization is proposed in this paper. In the new algorithm, cost function and risk function are introduced to realize reasonable utilization of the latest observation, and then confidence degree distance and confidence degree matrix are constructed and calculated to extract and use the redundancy and complementary information among particles. Finally, the effective measure of particle weights is realized by the weight balance factor according to the combination of above two weights measure results. The new method not only makes full use of the information from the current time particles set, but only avoids the adverse effect of prior statistical information error, therefore it make evaluation results of particles weights more stable and reliable. The theoretical analysis and experimental results show the efficiency of the proposed algorithm.