单通道信源数估计的改进方法

摘要

本文对单通道接收信号的源数估计方法进行了研究, 提出了对现有方法的改进措施.将单通道数据通过延迟处理转换为多通道形式, 然后引入阵列信号处理中的信源数估计算法, 如盖氏圆盘估计法(Gerschgorin's Disk Estimation, GDE)和最小描述字长法(Minimum Dscription Lengh, MDL).基于信息理论标准(ITC)的MDL方法在低SNR条件下获得比GDE更好的性能, 但是它无法处理包含有色噪声的信号.GDE方法虽然可以克服有色噪声的影响, 但是其在低SNR下的性能欠佳.基于上述考虑, 本文对这两种方法进行了改进.采用对角加载技术改善MDL方法的性能, 并引入Jackknife切法优化数据协方差矩阵, 以提高GDE方法的性能.模拟实验结果表明: 本文提出的方法使原有方法的性能得到很大改善.%Source number estimation methods for single channel received signal are studied and the improvement for each method is proposed in this paper.The singlechannel datais converted to multi-channel form by delay process.Then the algorithms used in the array signal processing, such as Gerschgorin's disk estimation (GDE) and minimum description length (MDL), are introduced to estimate the source number contained within the received signal.The MDL method, which is based on information theoretic criteria (ITC), obtains a better performance than GDE at low SNR condition, but it has no ability to handle the signals that contained with colored noise, while the GDE method can overcome the influence of colored noise.But its performance at low SNR has been less than satisfactory.Based on the above consideration, this paper makes improvement on these two methods.A diagonal loading technique is employed to ameliorate the MDL method and a jackknife technique is introduced to optimize the data covariance matrix to improve the performance of GDE method.The results of simulation experiment show that the boosting we put forward makes the performance of original methods get improvement largely.