An ADPCM with an exponential power estimator based predictor for speech coding

Abstract

The application of the exponential power estimation (EPE) technique to realize computationally simple predictor adaptation algorithms has been investigated. An adaptive differential pulse code modulation (ADPCM) system which employs the EPE-based adaptive predictor provides a computational advantage of 2N multiplications (N being the predictor order) and one division per iteration compared to the ADPCM which employs a block power estimator based adaptive predictor. A performance comparison has been conducted with two conventional ADPCM systems at different bit rates and channel conditions, using different quantizer characteristics. The results indicate that the proposed ADPCM performs identically to the conventional ones at 32 kbps and better than them at 16 kbps. >