No abstract

In this paper, an FIR cascade structure for adaptive linear prediction is studied in which each stage FIR filter is independently adapted using LMS algorithm. The theoretical analysis shows that the cascade performs a linear prediction in a way of successive refinement and each stage tries to obliterate the dominant mode of its input. Experimental results show that the performance of the cascade LMS predictor are in good agreement with our theoretical analysis.

The performance of an FIR cascade structure for adaptive linear prediction is studied, in which each stage FIR filter is independently adapted using LMS algorithm. In this paper, the performance bound is derived for cascade LMS predictor under some assumptions. We discover that this bound is possible to be better than that of the linear predictive coding (LPC) technique using block-based Levinson-Durbin algorithm if the cascade LMS predictor is well selected. We show some examples of this bound for synthetic and real audio signals in which a cascade LMS predictor outperforms the LPC technique using Levinson-Durbin algorithm.

In this paper, we study the issue of the high sampling rate audio modeling for lossless audio coding. We propose a cascade LMS structure to successfully model all high sampling rate audio signals. This cascade structure predictor, not only performs better than its counterpart FIR linear prediction coding (LPC) technique in modeling general audio signals, but also displays a faster convergence and smaller mean square error (MSE) than conventional LMS predictor and low-order stages cascade LMS predictor, while the complexity of the proposed predictor remains simple. The simulation results show that the proposed structure gets better prediction gain compared with Monkey's audio codec and MPEG-4 ALS codec provided by Technology University of Berlin (TUB) for real high sampling rate audio test set. Other adaption algorithms can be used for the single stages.