用于纯方位跟踪的简化粒子滤波算法及其硬件实现

洪少华; 史治国; 陈抗生

doi:10.3724/SP.J.1146.2007.01166

用于纯方位跟踪的简化粒子滤波算法及其硬件实现

doi: 10.3724/SP.J.1146.2007.01166

计量
- 文章访问数: 4594
- HTML全文浏览量: 90
- PDF下载量: 4043
- 被引次数: 0
出版历程
- 收稿日期: 2007-07-16
- 修回日期: 2008-09-29
- 刊出日期: 2009-01-19

Simplified Algorithm and Hardware Implementation for Particle Filter Applied to Bearings-only Tracking

摘要

摘要: 针对粒子滤波运算量大，硬件复杂性高的问题，该文提出了一种用于纯方位跟踪的简化粒子滤波算法，该算法引入了一种新的基于阈值的重采样方法，降低了硬件实现的复杂度。在算法研究的基础上，论文研究了基于FGPA的硬件电路实现方法，给出了系统的整体硬件结构及重采样/采样模块的实现方案，讨论了粒子滤波硬件实现的资源优化及时间优化问题。仿真结果表明，对于纯方位跟踪问题，该粒子滤波算法具有优于扩展Kalman滤波器(EKF)的性能；硬件电路实验表明，该滤波器可以实现对被动目标的纯方位跟踪，并具有比通用粒子滤波器较快的处理速度。
- 纯方位跟踪;粒子滤波;重采样;FPGA
Abstract: A simplified particle filter algorithm, which introduces a compact threshold-based resampling algorithm and features lower computing power and hardware complexity, is proposed for the bearings-only tracking problem. Based on the proposed algorithm, this paper lays emphasis on the efficient hardware implementation of particle filters on FPGA platform, and presents the hardware architecture of the resample/sample unit and the whole system. Simulation results show that the simplified algorithm outperforms the extended Kalman filter. Experimental study indicates that the implemented particle filter can be used to solve the bearings-only tracking problem and has rather fast processing rate.

HTML全文

参考文献(1)

Song T L and Speyer J L. A stochastic analysis of a modifiedgain extended Kalman filter with application to estimationwith bearings-only measurements. IEEE Trans. on AutomaticControl, 1985, AC-30(10): 940-949.[2]郭福成, 李宗化, 孙仲康. 无源定位跟踪中修正协方差扩展卡尔曼滤波算法[J].电子与信息学报.2004, 26(6):917-922浏览[3]Ristic B, Arulampalam S, and Gordon N. Beyond theKalman Filter: Particle Filters for Tracking Applications.Boston, London: Artech House, 2004, Chapter 5-12.[4]Doucet A, de Freitas N, and Gordon N (Eds.). SequentialMonte Carlo Methods in Practice. New York: Springer,2001, Chapter 15-26.[5]Doucet A and Wang X D. Monte Carlo methods for signalprocessing: A review in the statistical signal processingcontext. IEEE Signal Processing Magazine, 2005, 22(6):152-170.[6]Gordon N, Salmond D, and Smith A. Novel approach tononlinear/non-Gaussian Bayesian state estimation[J].IEEProceedings on Radar and Signal Processing.1993, 140(2):107-113[7]Zhai Y and Yeary M. A new particle filter tracking algorithmfor DOA sensor system. Proc. of Instrumentation andMeasurement Technology, Warsaw, 2007: 1-4.[8]Bolic M, Athalye A, and Djuric P M, et al.. Algorithmicmodification of particle filters for hardware implementation.Proc. of the European Signal Processing. Conference, Vienna,Austria, 2004: 1641-1646.[9]Bolic M, Djuric P M, and Hong S. Resampling algorithms forparticle filters: A computational complexity perspective.EURASIP Journal of Applied Signal Processing, 2004, (15):2267-2277.[10]Athalye A, Bolic M, and Hong S, et al.. Generic hardwarearchitectures for sampling and resampling in particle filters.EURASIP Journal of Applied Signal Processing, 2005, (17):2888-2902.

施引文献

资源附件(0)

访问统计