Digital communications for low latency and applications for constant envelope signalling

En savoir plus sur le livre

This thesis considers the problem of digital point-to-point communication with low latency, where the data are protected by an error-correction code. A solution was found by G. Ungerböck in 1982 where he invented TCM. His approach deploys a convolutional code for signal points in a larger signal constellation that are most difficult to distinguish on a noisy channel and by this achieves a high bandwidth- and power-efficiency. However, if communication takes place over a channel introducing ISI and, additionally, a flexible rate-adjustment is desired to adjust the transmission schemes to different transmission environments, the classical TCM approach reaches its limits. In such a communication scenario, the receiver faces two problems: that of mitigating the effects of ISI (equalization or detection) and that of decoding. In practice, both can be approached separately at a loss in performance. The major contribution of this thesis is to extend classical TCM for joint equalization and decoding, and in particular, for decoding of punctured convolutional codes. It will be shown that by means of efficiently merging the ISI-channel with the convolutional encoder, a reduced-complexity but still optimum joint equalization and decoding can be performed at the receiver. Furthermore, even though a flexible rate is achieved by puncturing (i. e. discarding of code symbols), an optimal decoder will be derived that can also be extended for ISI-channel and higher transmission rates. Finally, the proposed algorithms are adapted to CPM which enables the use of HPA at the transmitter and in battery-powered devices. The first part of the thesis focuses on the theoretical foundation of the considered communication sys