Analysis of subspace migrations in limited-view inverse scattering problems

An analysis of subspace migration that occurs in the limited-view inverse scattering problem is considered herein. Based on the structure of singular vectors associated with the nonzero singular values of the multi-static response matrix, we establish a relationship between the subspace migration imaging function and Bessel functions of integer order of the first kind. The revealed structure and numerical examples show why subspace migration is applicable for imaging of small scatterers in limited-view inverse scattering problems.     

连续隐马尔科夫模型在多基地目标识别中的应用

多基地声纳组网探测系统是目前大范围水下安保领域的研究热点。综合利用多基地系统中各个声纳节点的信息进行水下目标识别是亟待解决的问题。利用传统的多传感器融合的方法进行多基地水下目标识别,往往忽略了各声纳节点之间的相关性,效果并不理想。针对这一问题,本文提出了利用连续隐马尔科夫模型(CHMM)进行多基地水下目标识别的方法。首先利用RELAX算法提取了目标在不同分置角上回波的强散射点特征,组成观测向量,利用Baum-Welch方法对CHMM参数进行训练,然后计算待识别目标的特征值观测序列在不同模型下的似然概率。对所有目标重复此过程,取概率最大值对应的目标类别为最后的识别结果。在消声水池开展多基地模拟实验,对四类目标进行了识别,利用CHMM方法得到的多基地水下目标融合识别率比多基地声纳下单声纳节点的最高识别率提高了30%。