应用小波变换进行信号消噪处理

小波分析在时域与频域同时具有良好的局部化性质，可以利用信息信号和噪声信号在小波变换下具有截然不同的奇异性来区分信息信号与噪声信号。根据信号与噪声在二进制小波变换随尺度参数减小时信息信号和白噪声信号的小波变换的模极大值点变化的不同性质做消噪处理，然后再重建消噪后的信号。采用本研究所给出的方法对实际数据进行处理，其结果表明应用小波分析可以明显地抑制噪声，提高信噪比。此方法具有很好的实用价值。

Denoising Method With Wavelet Transform

Wavelet analysis is well localized characteristic both in time and frequency fields.Signal and white noise can be discriminated based on their different singularities with wavelet transform.The signals can be reconstructed after noise of signals is decreased by analyzing the changes of wavelet transform modulus maxima of signals and noises when the scales of dyadic wavelet transform become small.The results processed by the method with real data indicates that the ratio of signal and noise is raised and noise is suppressed evidently by using wavelet transform.This method is well suitable for practical application.