无线传感网中多传感器特征融合算法研究

曹红兵; 魏建明; 刘海涛

doi:10.3724/SP.J.1146.2008.01800

无线传感网中多传感器特征融合算法研究

doi: 10.3724/SP.J.1146.2008.01800

计量
- 文章访问数: 3252
- HTML全文浏览量: 116
- PDF下载量: 1101
- 被引次数: 0
出版历程
- 收稿日期: 2008-12-26
- 修回日期: 2009-06-11
- 刊出日期: 2010-01-19

Research on Multi-Sensor Feature Fusion Algorithms in Wireless Sensor Networks

摘要

摘要: 面向无线传感器网络在地面目标识别方面的应用需求，该文提出了一种基于改进局域判别基(Local Discriminant Bases, LDB)和二进制粒子群优化(Binary Particle Swarm Optimization, BPSO)方法的多传感器特征融合算法。利用新的基于概率密度估计的相对微分熵可分性测度来改进LDB，以提取目标信号的特征频段，然后分别利用一种改进的和一种全新的BPSO来实现特征融合。基于实地采集到的地面目标的声音和震动信号，仿真实验表明，该方法减少了所需分类器的数目，降低了特征维数，并在一定程度上提高了目标的正确识别率，具有实际的应用价值。
- 无线传感器网络; 特征融合; 局域判别基; 二进制粒子群优化; 可分性测度
Abstract: A multi-sensor feature fusion algorithm based on improved Local Discriminant Bases (LDB) and Binary Particle Swarm Optimization (BPSO) is proposed in this paper to satisfy the requirement of application on classification of ground targets in wireless sensor networks. LDB is improved by a new discriminant measure using relative differential entropy based on probability density estimation and used to extract the characteristic frequency band of signals. Then an improved and a new BPSO are used for feature fusion respectively. Based on real acoustic and seismic signals of ground targets, experiment results indicate that this method can decrease the classifier number needed, reduce the dimension of features, and improve the performance of classification at a certain extent, so it is practically valuable for application.

HTML全文

参考文献(1)

[1] Yick J, Mukherjee B, and Ghosal D. Wireless sensornetwork survey [J].Computer Networks.2008, 52(12):2292-2330 [2] Duarte M and Hu Y H. Vehicle classification in distributedsensor networks [J].Journal of Parallel and DistributedComputing.2004, 64(7):826-838 [3] 聂伟荣, 朱继南, 郭亚军等. 地震动信号的分析与目标识别[J].电子科技大学学报, 2003, 32(1): 26-30.Nie Wei-rong, Zhu Ji-nan, and Guo Ya-jun, et al.. Seismicsignals analysis and identification [J]. Journal of UEST ofChina, 2003, 32(1): 26-30. [4] Mazarakis G P and Avaritsiotis J N. Vehicle classificationin sensor networks using time-domain signal processing andneural networks [J].Microprocessors and Microsystems.2007, 31(6):381-392 [5] Wu H W and Mendel J M. Classification of battlefieldground vehicles using acoustic features and fuzzy logic rulebasedclassifiers [J].IEEE Transactions on Fuzzy Systems.2007, 15(1):56-72 [6] Malhotra B, Nikolaidis I, and Harms J. Distributedclassification of acoustic taargets in wireless audio-sensornetworks [J].Computer Networks.2008, 52(13):2582-2593 [7] Kuncheva L I, Bezdek J C, and Duin R P W. Decisiontemplates for multiple classifier fusion: An experimentalcomparison [J].Pattern Recognition.2001, 34(2):299-314 [8] Pan Q, Wei J M, and Cao H B, et al.. Improved DSacoustic-seismic modality fusion for ground-moving targetclassification in wireless sensor networks. PatternRecognition Letters, 2007, 28(16): 2419-2426. [9] Mallat S. A Wavelet Tour of Signal Processing [M]. SanDiego: Academic Press, 1998: 322-338. [10] Saito N and Coifman R R. Local discriminant bases andtheir applications [J]. Journal of Mathematical ImagingVision, 1995, 5(4): 337-358. [11] Umapathy K, Krishnan S, and Rao R K. Audio signalfeature extraction and classification using local discriminantbases [J].IEEE Transactions on Audio, Speech, andLanguage Processing.2007, 15(4):1236-1246 [12] 柳革命, 孙超, 刘兵等. 局域判别基空间能量的水声目标特征提取 [J]. 声学技术, 2007, 26(6): 1089-1093.Liu Ge-ming, Sun Chao, and Liu Bing, et al.. Featureextraction based on subspace energy of local discriminantbasis [J]. Technical Acoustics, 2007, 26(6): 1089-1093. [13] Kennedy J and Eberhart R C. Particle swarm optimization[C]. IEEE 1st International Conference on Neural Networks,IV. Piscataway, NJ: IEEE Service Center, 1995: 1942-1948. [14] Eberhart R C and Kennedy J. A discrete binary version ofthe particle swarm algorithm [C]. IEEE Conference onSystem, Man, and Cybernetics, Orlando, FL, 1997, 5: 4104-4109. [15] Chuang L, Chang H, and Tu C, et al.. Improved binaryPSO for feature selection using gene expression data [J].Computational Biology and Chemistry.2008, 32(1):29-38 [16] Huang C and Dun J. A distributed PSO-SVM hybridsystem with feature selection and parameter optimization[J].Applied Soft Computing.2008, 8(4):1381-1391 [17] Seekings P and Potter J. Classification of marine acousticsignals using wavelets and neural networks [C]. Proceedingsof Underwater Defense Technology Conference, Singapore,2003: 1-6. [18] Clerc M and Kennedy J. The particle swarmexplosion,stability, and convergence in a multidimensional complexspace [J].IEEE Transactions on Evolutionary Computation.2002, 6(1):58-73

施引文献

资源附件(0)

访问统计