摘要
关于水声信号检测优化问题,对在极低信噪比情况下被背景噪声所淹没的微弱信号进行检测时,由于舰船航行时接收目标信号的噪声较大,使检测更加困难。通过对谐波小波变换和支持向量回归算法的分析,在谐波小波函数的窄带信号分析支持向量回归基础上,提出了一种谐波小波核函数-支持向量回归的信号检测算法,实现小样本情况下微弱信号的检测。通过仿真信号和海上实测噪声数据的分析,利用改进算法可以很好地检测出在极低信噪比情况下噪声背景中的微弱信号,从而验证了改进算法在低信噪比情况下检测线谱信号的有效性。
To detection the weak signal under low SNR, based on the narrow-band signal analysis capability of the Harmonic Wavelet Function, and combined with the support vector regression, the Harmonic Wavelet Kernel Support Vector Regression algorithm was proposed for detecting the line spectrum signals under the condition of small samples. The simulation results with measured noise data show that the algorithm can detect the line spectrum signal in the Gaussian noise background effectively under the condition of small samples.
出处
《计算机仿真》
CSCD
北大核心
2013年第1期263-267,共5页
Computer Simulation
基金
陕西省教育厅科研计划项目资助(2010JK841
2011JK0937)
关键词
谐波小波函数
支持向量回归
线谱信号
信号检测
Harmonic wavelet function
Support vector regression
Line spectrum signal
Signal detection
