CAJ | 학술논문

源数目估计是阵列信号处理中的一个重要课题,也是声纳探测领域的一个难题.通过研究特征子空间分析方法,提出了一种新的源数目估计算法.该方法在盲源分离结果的基础上,基于信号子空间和噪声子空间正交的原理,建立了子空间代价函数来估计观测空间的源数目.同时,通过仿真和水池实验,对该方法的性能做了研究,并与基于信息论准则中最小长度描述准则的源数目估计算法进行了比较.仿真和水池实验结果表明,该源数目估计算法能得到准确的源数目,并具有较好的抗噪性,在水声环境中具有更好的稳健性.
원수목고계시진렬신호처리중적일개중요과제,야시성납탐측영역적일개난제.통과연구특정자공간분석방법,제출료일충신적원수목고계산법.해방법재맹원분리결과적기출상,기우신호자공간화조성자공간정교적원리,건립료자공간대개함수래고계관측공간적원수목.동시,통과방진화수지실험,대해방법적성능주료연구,병여기우신식론준칙중최소장도묘술준칙적원수목고계산법진행료비교.방진화수지실험결과표명,해원수목고계산법능득도준학적원수목,병구유교호적항조성,재수성배경중구유경호적은건성.
The estimation of the number of sources is an important problem in array signal processing and it is also a problem for sonar detection.Based on the principle of the eigen sub-space,a new algorithm is proposed for the estimation of the number of sources.The proposed algorithm makes use of the blind source separation and is based on the sub-space distance rule.The blind source separation offers the eigen vector to the new algorithm,and the sub-space distance rule is the fact that if the source number estimation is right,the signal sub-space and the noise sub-space would be in an orthogonal relation,with the distance between the signal sub-space and the noise sub-space taking the minimum value.Based on that rule,the sub-space cost function is built.The simulation data and pool test data are used to analyze the performance of the proposed algorithm and the minimum description length method,based on the information criteria.It is shown that the algorithm can provide the true number of sources in both simulation data and pool test data and with a good performance in low SNR in the simulation data.Compared with the minimum description length method,the proposed algorithm is more robust in the pool test data.