Adaptive signal processing for telecommunication network modelling and QoS estimation

Abstract

A novel modulating scheme is presented that improves echo path modelling using system identification techniques in noise contaminated environments. The systems of interest are usually based on a class of digital adaptive filters (DAFs). A novel chaotic based modulation regime utilising the logistic mapping is exploited to whiten the speech power spectral density (PSD) whilst preserving the signal bandwidth requirements. Software simulations are reported for noise-impaired and noise-free circuits. In the noise-free configuration, it is shown that the described modulation scheme significantly improves the convergence rate and guarantees optimal modelling independent of the input speech signal statistics. In the noise contaminated environment, it is shown that stringent operating requirements dictate modelling with echo to noise ratios approaching 0 dB. Uncoded speech is shown to be ineffective in driving the models when noise to echo ratios exceed approximately -10 dB. The noise-impaired environment may invalidate speech driven DAF circuits, however, it is shown that the chaotic coding improves the signal energy and whitens the resultant PSD. As a result, it is demonstrated that the proposed coding scheme allows accurate DAF modelling for noise levels that are 40 dB higher than the corrupted uncoded speech case. Furthermore, it is shown that for a practical range of noise conditions the proposed coding results in a faster convergence rate than the upper bound established for the noiseless uncoded case.