Simon Hüttinger


Analysis and Design of Power-Efficient Coding Schemes

Band 6, Erlanger Berichte aus Informations- und Kommunikationstechnik, Herausgeber: A. Kaup, W. Koch, J. Huber. Shaker Verlag, Aachen, 2004. ISBN 3-8322-2402-5.

Abstract

Analysis and Design of Power-Efficient Coding Schemes is of central importance for digital communications. The superiority compared to traditional analog transmission arises from the power-efficiency and reliability

A break-through was achieved in 1993 with the discovery of the so called turbo-codes. Since that time we know that power-efficiency preferably can be attained by application of concatenated channel codes and iterative decoding methods.

The aim of this work is to motivate and analyze iterative decoding, which originally has been introduced more or less heuristically. From the results of the scientific analysis, methods are derived which permit the design of power-efficient channel codes for different applications.

Firstly, in Chapter 3 both component codes and concatenated channel codes are examined by methods derived from information theory. The analysis of the component codes shows that convolutional codes have the potential to achieve power-efficient communication, only if they possess a rate greater than channel capacity. But, in this region their characteristics are almost optimal. Thus, it can be reasoned, that concatenations result in power-efficient coding schemes, as their components are used at operation points they are very well suited for. If additionally the loss of symbol-by-symbol decoding is considered, it can be recognized, that preferably systematic component encoders should be used. Furthermore, the comparison of optimum decoding and symbol-by-symbol decoding easily motivates iterative decoding, as the loss of symbol-by-symbol decoding is diminished during the iterations. Extensions of the analysis methods to concatenated channel codes lead to a possibility to lower bound the performance, e.g., measured by the bit error ratio, without simulation of the iterative decoding process.

The construction of power-efficient channel coding schemes is exemplified in Chapter 4. As the analysis methods developed in Chapter 3 are of very low complexity, they are suited for the search of the most power-efficient concatenations within relatively large groups of possible concatenated codes. If strategies for the concatenations are chosen, which are suitable for an intended application, this is a way to find extremely power-efficient channel codes under almost arbitrary boundary conditions. For example the multiple-turbo-codes are suited very well for low-rate transmission, as needed, e.g., in deep-space communications or cellular communication schemes based on code division multiple access.