A method of signal denoising. Through the analysis of different signal to noise noise intensity under the correlation coefficient and the original signal autocorrelation coefficient, to find out the original signal and the noise autocorrelation coefficient or the maximum difference before and after denoising noise autocorrelation coefficient difference is equal to the autocorrelation coefficient to remove noise results in small, for the best denoising results. The method of the invention comprises the following steps: firstly calculate the signal autocorrelation coefficient, and identify the recorded signal autocorrelation coefficient in delay = 0 original signal autocorrelation coefficients and noise autocorrelation coefficient. Then, the signal is denoised according to different denoising intensities. Then according to the different denoising intensity denoising denoising results calculate the signal correlation coefficient, and identify the record results of denoising signal autocorrelation coefficient at t = 0 of the original signal and noise autocorrelation coefficient. It is found that the difference between the original signal and the noise autocorrelation coefficient is the largest, or the difference between the noise autocorrelation coefficients before and after denoising is equal to the noise cancellation coefficient, which is the best denoising result. The method of the invention takes the autocorrelation coefficient of the denoising result as the criterion, has nothing to do with the denoising algorithm and the tool and the denoising process, and can be used in various existing denoising methods and tools.
【技术实现步骤摘要】
本专利技术是一种信号去噪方法,属于信号处理
技术介绍
信号在产生和测量过程中不可避免地产生和带来噪声。这些噪声与原始信号叠 加,干扰了对原始信号的后续分析和处理。许多研究围绕着信号去噪展开,并且收到了积极 的效果。有许多不同的信号去噪方法,常见的有滤波器去噪,傅立叶变换去噪和小波变换 去噪。由于带噪声信号的原始信号和噪声在实际应用中往往未知,目前为止的各种去噪方 法都是基于对噪声的估计进行去噪,由此产生了对噪声的去噪不足和去噪过分的问题。去 噪不足不能完全去除噪声,去噪过分会使得原始信号畸变。并且,现有去噪技术不能定量地 去除噪声。
技术实现思路
本专利技术提出一种新的克服了现有技术的上述不足的信号去噪方法。这种方法通过 分析不同去噪强度下的信号的噪声自相关系数和原始信号自相关系数的大小,找出其中原 始信号与噪声自相关系数差值绝对值最大(完全去噪)或者去噪后噪声自相关系数的减少 值等于或接近要去除噪声的自相关系数大小(部分去噪)的去噪结果,为最佳的去噪结果。 由于本专利技术方法能够通过计算信号的自相关系数来间接计算噪声,使得对信号的精确去噪 成为可能。本专利技术方法的对噪声和原始信号大小的分析基于以下原理信号x(t)的自相关函数定义为Rx (τ)=厂 x{t)x(t + r)dt(1)其中τ是时间延迟对信号X(t)X(t) = S(t)+N(t) (2)其中S(t)为原始信号,N(t)为噪声,计算自相关系数,结果为Rx(t) = Rs(x)+Rn(x) (3)其中Rs( τ )为原始信号的自相关系数,其幅度与原始信号的大小成正比;Rn( τ ) ...
【技术保护点】
1.一种信号去噪方法,其特征在于该方法包括如下步骤:(1)计算信号的自相关系数曲线,并且识别和记录信号自相关系数中在延时τ=0处的原始信号自相关系数和噪声自相关系数数值。(2)对信号按照不同的去噪强度去噪。(3)对按不同去噪强度去噪的去噪结果计算自相关系数,识别和记录这些去噪结果自相关系数在τ=0处的原始信号自相关系数和噪声自相关系数数值。(4)按照完全去噪判据或者部分去噪判据找出最佳去噪结果。
【技术特征摘要】
...
还没有人留言评论。发表了对其他浏览者有用的留言会获得科技券。