噪声空间谱预白化小孔径阵列被动目标检测方法  

An Effective Noise Spatial Spectrum Pre-Whiten Approach in Passive Detection Using Small Aperture Array

作　　者：王静[1] 黄建国[1] 张群飞[1] 韩晶[1] 

机构地区：[1]西北工业大学航海学院,陕西西安710072

出　　处：《西北工业大学学报》2012年第3期422-427,共6页Journal of Northwestern Polytechnical University

基　　金：国家自然科学基金(60972152);国家重点实验室基金(9140C230309110C23);航空科学基金(2009ZC53031)资助

摘　　要：为了解决小孔径阵列在低信噪比下不能正确检测被动目标的问题,提出了两种小孔径阵列非白噪声场空间预白化方法:噪声空间谱模值归一化(MNNSS)和噪声协方差矩阵求逆(NCMI)方法。假设噪声场为空间平稳非白噪声场,通过学习得到平稳非白噪声场的空间角谱和噪声协方差矩阵,使用学习的空间角谱和协方差矩阵对阵列接收数据进行预白化处理。将预白化处理后的阵列数据用于空间谱检测器。水池实验数据处理结果表明,采用MNNSS和NCMI方法对阵列接收数据预白化处理可明显改善小孔径阵列噪声场的非白特性,预白化处理后空间谱检测器在检测概率达到90%时的可检测信噪比分别下降了3 dB和7 dB;MNNSS和NCMI方法运算量不大,可以运用到实时处理系统。Target detection using small aperture array under low signal to noise ratio （SNR） is a difficult problem in passive homing system. Two methods for whitening the noise spatial spectrum using small aperture array are pres- ented to solve two difficult problems ： magnitude normalization on noise spatial spectrum （MNNSS） and noise covar- iance matrix inverse （NCMI）. Sections 1 and 2 of the full paper explain the noise spectrum pre-whiten approach mentioned in the title, which we believe is effective and whose core consists of： ＂Assume the noise background is a stationary time series. First, the spatial spectrum and the covariance matrix of noise are computed by learning the stable color noise and then the received data of array is pre-whitened by using the spatial spectrum and the covari- ance matrix of noise. ＂ Experimental results, presented in Tables 2 and 3 and Figs. 1 through 5, and their analysis show preliminarily that： （ 1 ） the color character of noise is restrained by MNNSS and NCMI pre-whiting methods; （2） the detection performance of the spatial spectrum detector is improved by 3dB and 7dB respectively under 90% probability of detection ; （3） MNNSS and NCMI are computationally saving when applied in real time systems.

关 键 词：算法 计算复杂度 协方差矩阵 效率 实验 实时处理系统 频谱分析 检测 预白化 小孔径阵列 

分 类 号：TN911.7[电子电信—通信与信息系统]