Error Resilient Turbo Compression of Source Codec Parameters Using Inner Irregular Codes

Abstract:
In this paper, we present a novel near-lossless compression scheme for scalar quantized source codec parameters. The scheme is comparable to a Turbo source coding approach and can inherently incorporate protection against transmission errors. We show that using the concept of EXIT charts and inner irregular convolutional codes of rate > 1, a linear programming optimization problem can be formulated resulting in the minimization of the transmitted number of bits given a certain channel quality. The solution of this problem leads to an irregular inner code offering the best possible compression. The advantage of this method in comparison to a previous proposal is the easy adaptability to varying source conditions.