Adaptive Playout Algorithm Based on Wavelet Transform and Kernel Estimation in VoIP

Adaptive playout algorithms are used at the receiver to combat jitter problems in voice over IP (VoIP). The playout algorithms rely on short term estimations of the characteristics of network delay that would be experienced by up-coming voice packets. This paper proposes a new adaptive playout algorithm based on wavelet transform and kernel estimation. The proposed algorithm uses wavelet transform to decompose non-stationary delay series into several stationary components, and then predicts all these components with kernel estimation. Finally, the overall end-to-end delay is obtained by wavelet reconstruction with these predicted components. Experimental results show that the prediction precision of the new algorithm is about 6-17 dB higher than that of the conventional auto regressive (AR) algorithm, about 2.5~8 dB higher than that of the differential auto regressive (DIAR) algorithm and 1~3 dB higher than that of our latest proposed wavelet transform based auto regressive (WTAR) algorithm.