【摘要】 海洋混响是主动声呐检测中主要的背景干扰,它是一种有色干扰噪声,并常被看做是一个非平稳的随机过程,这使得工作在白噪声条件下的最佳匹配滤波检测器的性能在一定程度上受到限制.因此,提出一种次最佳检测器,将混响看成局部平稳,并将接收信号分成许多小块,每小块利用自回归(AR)模型对混响时间序列进行建模,然后利用估计的AR系数构造白化滤波器,对混响进行预白化处理,并将白化后的序列用于匹配滤波检测器.通过计算机仿真和实验数据分析比较未白化与经白化处理的匹配滤波性能,结果表明,次最佳检测器比未白化匹配滤波在检测性能上可提高近3dB.更多还原

【Abstract】 In active sonar detection,reverberation is the dominant background interference. It is a colored disturbance noise and is always regarded as a non-stationary random process, which degrades the performance of the matched filter that is the optimum detector in white background noise. A sub-optimum detector is introduced in this article. In the new detector, the received data is divided into many blocks,each one considered to be locally stationary. An autoregressive process of reverberation sequence is modeled for each block, and whitened filters based on this model are made for pre-whiten processing of reverberation. New sequences after being whitened are applied to the matched filter. The performances of the matched filter and the new detector with whitening are presented in this article. From the result of analyzing the imitated and experimental signal, it can be found that the sub-optimum detector improves the performance of detection about 3dB in reverberation background.更多还原