时频域自适应均值脊波变换滤波及其检测方法.pdfVIP

时频域自适应均值脊波变换滤波及检测方法 朱广平，孙辉，朱凤芹 (哈尔滨工程大学 水声工程学院 黑龙江 哈尔滨, 150001) 摘要：针对Wigner-Ville Hough 变换(WHT) 在低信噪比下检测性能降低的问题，提出了互Wigner-Ville 自适应均值脊波变 Wigner-Ville XWVD Wigner-Ville WVD 换滤波检测方法。该方法首先采用互 （ ）代替 （ ）在时频平面上提高了信噪比，且避 免了信号多分量时各分量间的交叉项干扰。由于信号和噪声的时频聚集特性不同，且考虑到滑动窗长对均值滤波的影响，设 计了自适应轴向均值滤波器及采用脊波变换滤波方法在时频域滤除噪声，最后采用Hough 变换检测信号。给出了详细的滤 波检测算法。仿真和海试结果表明：该方法与WHT检测相比，在低信噪比下通过时频域滤波能有效的滤除噪声从而提高检 测性能。 关键词：信号检测；噪声；混响；时频域；脊波变换 中图分类号 TB56 文献标志码 A ： ： ADAPTIVEMEANANDRIDGELETTRANSFORMFILTERINGINTIME-FREQUENCY ADAPTIVEMEANANDRIDGELETTRANSFORMFILTERINGINTIME-FREQUENCY AADDAAPPTTIIVVEEMMEEAANNAANNDDRRIIDDGGEELLEETTTTRRAANNSSFFOORRMMFFIILLTTEERRIINNGGIINNTTIIMMEE--FFRREEQQUUEENNCCYY DOMAINFORDETECTION DOMAINFORDETECTION DDOOMMAAIINNFFOORRDDEETTEECCTTIIOONN ZhuGuang-pingSunHuiZhufeng-qin (CollegeofUnderwaterAcousticEngineering,HarbinEngineeringUniversity,Harbin,China150001) Abstract: Abstract: AAbbssttrraacctt:: For solving the problem which the performance of detection was reduced in the low signal to noise ratio (SNR) using Wigner-Ville Hough transform (WHT), we proposed the method of XWVD adaptive mean and ridgelet transform filtering. In this method, firstly used XWVD instead of WVD for improve SNR and avoiding the cross-component when the signal were multi-components. Due to the power distribution of signal is different from noise’s, so designed adaptive axial mean filter and using ridgelet transform filter to restrain noise in time-frequency domain. At last, it is to detect the signal using Hough transform. The signaldetection algorithm in detail was given.Theresult of real and simulation experime