摘要
在非平稳非高斯背景噪声下 ,使用经典信号检测理论对信号进行检测往往难以达到理想的效果 .文中利用小波包变换 ,通过将信号及噪声变换到小波包系数域 ,使得噪声在某些尺度及子空间上的小波包系数成为近似平稳高斯噪声 ,从而运用信息融合的思想和经典信号检测理论的分集检测方法 ,对非平稳非高斯噪声下的确知信号检测问题提出了一种新的检测算法 .仿真实验表明该方法是经典信号检测理论的有效推广 .
If the background noise is nonstationary and non-Gaussian, the effect of classic signal detection theory is not satisfying. By wavelet packet transform, the transform coefficients of background noise at some scale and some space can be transited to be stationary and Gaussian. Further more, a new method to detect signal in nonstationary and non-Gaussian white background noise is given by reorganizing the information of useful signal incorporated in those coefficients, according to the thought of information amalgamation along with multisignal detection method. The simulative experiment shows that this method is an effective extension of classic signal detection theory.
出处
《武汉理工大学学报（交通科学与工程版）》
北大核心
2003年第3期354-357,共4页
Journal of Wuhan University of Technology（Transportation Science & Engineering）
基金
教育部骨干教师项目"水中运动目标的模型化探测技术研究"资助
关键词
非平稳
非高斯
小波包变换
信号检测
nonstationary
non-gaussian
wavelet packet transform
detection of signal
