维纳滤波信号传输畸变补偿方法性能影响因素分析

Factors affecting performance of the signal compensation method based on Wiener filtering

  • 摘要: 研究表明,基于维纳滤波的信号同轴电缆传输畸变补偿方法对于多种不同类型的脉冲信号均表现出优异的补偿性能和补偿效率,可以有效解决脉冲信号同轴电缆传输畸变问题,具有很好的应用价值。但是,目前尚未对影响该方法补偿效果的因素进行分析,也未明确其最佳应用条件或参数设置。深入分析了影响基于维纳滤波信号传输畸变补偿方法补偿性能的因素,确定主要影响因素为输出信号信噪比SNR、同轴电缆S21曲线测量步长Δf以及功率谱估计方法。针对上述影响因素开展了同轴电缆信号传输畸变补偿试验,结果表明:当SNR小于25 dB时,SNR越低,补偿方法补偿性能越差,而当SNR大于25 dB时,随着SNR的增大,补偿性能趋于稳定不变;对于S21参数测量频率步长Δf,当Δf较小时,补偿性能几乎保持不变,当Δf较大时,随着Δf的增大,补偿精度缓慢降低;而对于功率谱估计方法,常用的三种功率谱估计方法(周期图法、Welch法和Burg法)中,Burg法补偿性能最佳。研究结果可为维纳滤波实际应用提供指导。

     

    Abstract: The signal compensation method based on Wiener filtering has been demonstrated to have excellent compensation performance towards the signal distortion induced by the long-distance transmission in coaxial cable. However, factors affect the compensation performance of this method had not been investigated and analyzed, thus we carried out a systematic study on the effect of the parameters and the means of this modified method. The results show the main factors are the signal-to-noise ratio (SNR), the sampling frequency interval Δf in S21 parameter measurement, and the power spectrum estimation method. When the SNR is less than 25 dB, the compensation performance continuously improves as the SNR increases. Once the SNR exceeds 25 dB, the relative error (δ) between the compensated signal and input signal nearly keeps unchanged. The compensation performance keeps unchanged when Δf is small, and it slowly deteriorates as Δf exceeds a certain value. It is proved that among the traditional power spectrum estimation methods, the Burg method can obtain the best compensation performance. This study can provide a beneficial reference for the application of the signal compensation method based on Wiener filtering.

     

/

返回文章
返回