无线传感网络中目标定位的研究

在无线传感网络WSNs(Wireless Sensor Networks)中,现存的基于压缩感知的目标定位算法是假定目标均落在预定网格,当不满足此假设时,将极大地降低了目标定位算法的性能.为此,提出基于变分贝叶斯期望最大化的目标定位VBEM-TL (Variational Bayesian Expectation Maximization-based Target Localization)算法.VBEM-TL算法先利用一阶泰勒级数展开算法建立稀疏近似模型,然后将目标定位问题转化成稀疏恢复问题,再利用VBEM算法重构稀疏矢量.最后,依据重构的稀疏矢量估计目标位置.实验数据表明,提出的VBEM-TL算法能够有效地降低定位误差.

Study on Target Localization algorithm in Wireless Sensor Networks

In Wireless Sensor Networks(WSNs),existing compressed sensing-based schemes implicitly assume that all targets fall on a pre-defined grid exactly. However,when the assumption is violated,their performance deteriorates dramatically. Therefore,variational Bayesian Expectation Maximization-based Target Localization(VBEM-TL)algo-rithm is proposed in this paper. We first construct sparse approximate model with its first order Taylor expansion, then then formulate the counting and localization problem as a sparse recovery. At last,the sparse vector is recon-structed by VBEM algorithm. Finally,the location of targets is estimated by sparse vectors. Our numerical results show that VBEM-TL provides excellent accuracy,surpassing the state-of-the-art methods in general.