LFM信号基于自适应预白化处理的GLRT检测器  

GLRT detector for LFM signal based on adaptive prewhitening processing

作　　者：陈鹏[1] 侯朝焕[2] 马晓川[2] 梁亦慧 

机构地区：[1]中国科学院研究生院,北京100039 [2]中国科学院声学研究所,北京100080 [3]船舶系统工程部,北京100036

出　　处：《系统工程与电子技术》2006年第8期1138-1140,1168,共4页Systems Engineering and Electronics

基　　金：国家自然科学基金资助课题(60472101)

摘　　要：针对有色混响噪声背景下线性调频(LFM)信号的检测,提出了基于自适应预白化处理的广义似然比检测GLRT(generalized likelihood ratio test)改进算法。改进算法根据前一段的混响噪声数据估计AR模型参数,利用估计参数构成的白化器对当前段混响噪声数据进行滤波,并对滤波后数据进行单变量的GLRT检测。与双变量的GLRT算法相比,改进算法有效降低了运算量。最后通过蒙特卡罗仿真给出了改进算法的检测性能分析。To the detection of linear frequency-modulated （LFM） signal under colored reverberation noise background, the GLRT （Generalized Likelihood Ratio Test） modified algorithm based on adaptive prewhitening processing is proposed. The modified algorithm estimates the AR model parameters according to the reverberation noise data of the previous blodk, uses the whitener constructed by estimated parameters to filter the reverberation noise data of the current block, and executes the univariate GLRT judgement to the filtered data. Compared with the bivariate GLRT algorithm, the modified algorithm degrades the computations effectively. At last the detection performance analysis of the modified algorithm via the Monte Carlo simulations is given.

关 键 词：混响 AR模型 预白化 线性调频 检测器 

分 类 号：TB566[交通运输工程—水声工程]