利用遗传算法改进基于自适应阈值的小波去噪方法，并验证了该去噪方法的有效性。通过小波分析和经典谱估计方法计算了某款车驾驶员耳旁声信号的能量分布，发现其信号能量主要集中于低频部分，据此提出了利用软阈值和基于遗传算法自适应阈值的联合小波去噪方法。去噪前后信号对比结果表明，该方法可快速衰减信号的高频噪声，降低信号的波动。%Genetic algorithm is used to improve the wavelet denoising method with its effectiveness verified. By applying wavelet analysis and classic spectral estimation, the distribution of acoustic signal energy at driver's ear position of a bus is calculated and it is found that the signal energy mainly concentrates in low frequency range. Ac-cordingly a combined wavelet denoising scheme is proposed by using both soft threshold and genetic algorithm-based adaptive threshold. The comparison between signals before and after denoising indicates that the proposed scheme can quickly attenuate the high frequency noise and reduce signal fluctuations.
展开▼
