基于量子遗传粒子滤波的WSN目标跟踪算法

Tracking algorithms based on quantum genetic particle filter for WSN

粒子滤波器是解决非线性非高斯运动跟踪的一种有效方法,很适合于无线传感器网络的目标跟踪。但是粒子滤波算法存在严重的退化现象。常规的重采样方法虽可解决退化问题,但容易导致粒子耗尽。本文针对此问题,将量子遗传算法引入粒子滤波,提出了基于量子遗传粒子滤波的无线传感器网络目标跟踪算法。通过量子遗传算法的编码方式增加粒子集的多样性,从而缓解了粒子滤波的退化现象并解决了粒子耗尽问题,量子的并行性也节省了计算时间,提高了跟踪的实时性。仿真结果表明该算法是可行的。

Particle filter is an effective way to solve the problem like tracking objects with Non-Gaussian and Non-linear movement. It is well suited to target tracking in wireless sensor networks, but degeneracy phenomenon is serious. Common re-sampling method can resolve degeneracy phenomenon, but the sample impoverishment is a secondary result. Therefore, tracking algorithms based on quantum genetic particle filter for wireless sensor networks is proposed, in which quantum genetic algorithm is introduced. The diversity of particle sets increase through encoding of the quantum genetic algorithm, thus, the degradation in particle filter is eased and the problem of particle depletion is solved. Quantum parallelism saves the computation time and improved the real-time of tracking. Simulation results show the feasibility of the proposed algorithm.