基于峰均功率比的多目标检测方法性能分析

为了避免多目标治理环境中由于存在强度差导致检测概率降低的情况,给出了3种基于峰均功率比的多目标检测方法.利用采样协方差矩阵的特征向量受噪声方差影响小的特性,计算其峰均功率比值,将峰均功率比值应用于信息论准则和盖尔圆半径,检测信号源个数.仿真结果表明,对于不等强双目标,PAPRT、PGAIC和PGMDL方法的检测性能优于AIC和MDL方法.随着强度差的增大,给出方法的检测概率稳定在100%.

Performance analysis of multi-target detection methods based on peak-to-average power ratio

Three methods based on peak-to-average power ratio are involved in the presence of the intensity difference among multiple targets to avoid the detection probability′s decrease.By exploiting the characteristic of slight noise variance sensitivity to correlation matrix eigenvectors,the peak to average power ratio is calculated.The peak-to-average power ratios modified by Gerschgorin radii and information criterion are obtained to detect the number of signal source.Simulation results show that three methods which are PAPRT、PGAIC and PGMDL can achieve better source number estimation performance under the presence of intensity difference between the dual-targets than AIC and MDL algorithm.In particular,the detection probability of proposed approaches remains stable at 100 percent with the increase of the intensity difference.