Wavelet transform based adaptive playout algorithm for VoIP

Abstract

In Voice over Internet Protocol (VoIP) applications, adaptive playout algorithms are often used to counteract the effects of network jitter at the receiver. The playout algorithms rely on short term estimations of the characteristics of network delay that would be experienced by up-coming voice packets. AR (auto regressive) algorithm is the traditional one but only suitable for stationary delay series. The latest DIAR (differential auto regressive) algorithm based on ARIMA (auto regression integrated with moving averages) model has been proved to obtain good performance for non-stationary delay series. In this paper, a new adaptive playout algorithm based on wavelet transform is proposed. This algorithm uses wavelet transform to decompose non-stationary delay series into several stationary components, and then all these components are predicted by AR models with different adaptation rate. Subsequently, the final end-to-end delay is obtained by wavelet reconstruction with these predicted components. Simulation results show that the prediction precision of the new algorithm is about 5-14 dB higher than that of the conventional AR algorithm and about 1.5~5 dB higher than that of the latest DIAR algorithm.