一种基于QMC-APF的检测前跟踪算法

针对粒子滤波检测前跟踪算法中存在的粒子数目大,导致计算量和存储量大的问题,提出了一种基于拟蒙特卡罗的辅助粒子滤波检测前跟踪算法。该算法通过引入拟蒙特卡罗思想,产生低差异序列代替原来算法中的伪随机序列,使得粒子分布更加均匀,可以有效降低粒子数;采用辅助粒子滤波算法,对粒子进行两次加权操作。实验仿真表明,在对雷达弱目标进行检测与跟踪的过程中,该算法能够在保证算法性能的同时减少算法中的粒子数目,有效降低计算量和存储量。

Tracking Before Detection Algorithm Based on QMC-APF

The particle number is large in the tracking before detection algorithm based on particle filter, which result in a problem of a large amount of computation and storage, the auxiliary particle filter tracking before detection algorithm based on quasi-monte carlo is proposed. The quasi-monte carlo algorithm is proposed to replace the original pseudo-random sequence with low-discrepancy sequence in the algorithm, which can make a more uniform particle distribution and reduce the number of particles effectively.Using auxiliary particle filter to calculate the particle weight for the twice. The simulation results show that in the process of the radar weak target detection and tracking, the algorithm can guarantee the performance of the algorithm but also can reduce the number of particles in the algorithm, so that the amount of computation and storage are reduced effectively.