多模噪声背景下信号提取算法的研究

Study on Signal Extraction Algorithm in Multi-modal Hybrid Noise

信号检测是信号处理的一个重要的研究方向,以前的信号检测方法多数是基于在高斯噪声背景下进行讨论,对非高斯噪声研究比较少。非高斯噪声的干扰问题在通信过程中是经常出现的。背景噪声用多模(非高斯)噪声为模型,提出改进(最小均方误差牛顿)LMS Newton算法实现对多模背景噪声的抑制,并给出了多模噪声中信号的检测,达到对有用信号提取的目的。最后,进行了计算机模拟仿真,得出了结果。

Signal detection is one of the most important research topics in the signal process domain.The previous detection methods are mostly based on the Gaussian noise,and seldom on non-Gaussian noise.The disturbance of non-Gaussian noise is very common in communication process.Background noise is with multi-modal hybrid（non-Gaussian） noise as its model,thus the improved LMS Newton algorithm is proposed in order to control the multi-modal background noise,and thus acquire the useful signal.Finally,the computer simulation gives the corresponding results.