一种基于子波的信号复原方法

提出了一种从强噪声中复原瞬态信号的新方法，该方法将要复原的瞬态信号表示成一组子波函数的线性组合，在最小均方能量准则下，用共轭梯度算法求解子波函数的线性组合的幅度、尺度和延时参数，计算机模拟结果表明，该方法的实现结构可表示成具有子波非线性，并且类似于辐射函数ＢＰＦ的人工神经元网络。

Wavelets - Based Signal Restoration

A new scheme is presented for the restoration of transient signals embedded in the strong noise. Transient signals to be restored can be expressed as a linear combination of wavelets whose weight coefficients, dilations and shifts are calculated by the conjugate gradient algorithm in terms of the least mean squared energy criterion.The results of computer simulation demonstrate that this technique can optimally aestore transient signals that contaminated by the strong noise.