Effect of noise and bandwidth on homomorphic deconvolution

The impulse response of an unknown system is recovered from time-domain reflectometry (TDR) data using the homomorphic deconvolution technique. TDR waveforms obtained from the simulation of resistive and reactive discontinuities are used to investigate the effect of the system´s bandwidth on the homomorphic deconvolution. The effects of the noise content in the processed waveforms as well as the system´s bandwidth on the deconvolution process are studied. The same deconvolution applications are also treated using the optimal compensation deconvolution method, and the results are compared. The homomorphic deconvolution technique is shown to be fairly sensitive to the noise content in the TDR waveform. For an SNR (signal-to-noise ratio) of less than 40 dB, the cepstrum becomes very noisy, and the separation of the convolution components is not practical. It is concluded that noise reduction techniques are needed to improve the SNR, thus enabling the use of the homomorphic deconvolution technique.<>