Differential Pulse Code Modulation Encoder Research Articles

Embedded DPCM for Variable Bit Rate Transmission

Within the bit stream of an embedded digital code is a stream that can be decoded to produce a reasonable replica of the analog source signal. Unlike pulse code modulation (PCM), differential PCM (DPCM), is not an embedded code. If C bits/sample are delected from the bit stream of a DPCM encoder with E bits/sample, the decoded analog signal is substantially noisier than the output of a DPCM codec with D = E - C bits/sample. The penalty is 4-10 dB in signal-to-noise ratio (snr). However, with minor modifications to the encoder and decoder, DPCM becomes an embedded code. Embedded DPCM with E bits/ sample at the encoder and D bits/sample transmitted produces exactly the same output as embedded DPCM with D bits/sample encoding and perfect transmission. The snr of embedded DPCM is slightly lower than the Snr of DPCM. The penalty is 0.5-0.8 dB if the minimum transmitted bit rate is 2 bits/sample. It is less than 0.3 dB if D is at least 3 bits/sample. Combined with an appropriate adaptive quantizer the embedded DPCM codec produces embedded ADPCM (adaptive DPCM) for variable rate transmission ranging from 2 bits/sample up to any desired maximum. Applications exist in speech interpolation, packet switching, and hardware architecture.

DPCM Encoding of NTSC Composite Signals

Two subjective evaluation experiments were performed using an experimental digital television (EDIT) encoder developed at Comsat Laboratories. The first experiment led to the selection of a differential pulse-code-modulation (DPCM) quantizer from among a set of seven quantizers yielding various tradeoffs between quantizing noise and slope overload. In the second experiment, the picture quality of digitally processed pictures sampled at 10.7 MHz and coded with four bits per sample was found to be subjectively equivalent to the quality of pictures with an analog signal-to-weighted-noise ratio of about 47 dB.

A Unified Representation of Differential Pulse-Code Modulation (DPCM) and Transform Coding Systems

We consider a transform coding system that uses a lower-triangular transformation to uncorrelate the data. Based on this transformation we propose a generalized differential pulse code modulation (DPCM) system and show that at high bit rates it performs almost as well as coding by the method of principal components (Karhunen-Loeve transformation). This study connects the transform coding system to the DPCM encoder by showing that the proposed system simplifies to a standard DPCM encoder for Markov data.