This thesis deals with the low complexity implementation of single carrier wireline communications systems. An important technique, that is widely used in modern transceivers, in order to keep the number of analogue components low, is all digital sampling rate conversion. The sampling clock needs then not to be adjusted. In literature, underlying techniques are almost exclusively described for baseband signals. In the thesis, however, it is shown that resampling of the original passband signal is in many cases more efficient. Corresponding generalizations of well known sampling rate conversion methods are given in the first part of the theses. When the samples of the input signal are spaced by the appropriate frequency, equalization can be carried out in a following processing step. State of the art telecommunication standards oftentimes do not define a receiver training sequence and equalizer adaptation has then to be performed unsupervised. In this context, assuring global convergence of high performing adaptive decision feedback equalizers terms out to be difficult. Strategies to overcome this problem are derived in the second part of the theses. Finally, low power implementations of the examined algorithms are investigated.
«
This thesis deals with the low complexity implementation of single carrier wireline communications systems. An important technique, that is widely used in modern transceivers, in order to keep the number of analogue components low, is all digital sampling rate conversion. The sampling clock needs then not to be adjusted. In literature, underlying techniques are almost exclusively described for baseband signals. In the thesis, however, it is shown that resampling of the original passband signal i...
»