Record Detail Back
Blind Equalization and System Identification
- Share to:
Discrete-time signal processing has appeared to be one of the major momenta to the advances of science and engineering over the past several decades be- cause of the rapid progress of digital and mixed-signal integrated circuits in processing speed, functionality, and cost effectiveness. In many science and engineering applications, the signals of interest are distorted by an unknown physical system (channel) and only a set of discrete-time measurements with system (channel) induced distortion is available. Because both the source signals and measurements are basically random in nature, discrete-time statistical signal processing has played a central role in extracting the source signals of interest and estimating the system characteristics since the 1980s. When the system is known or can be accurately estimated at extra cost, estimation of the source signals via a statistical optimum filter (such as the Wiener filter) is usually straightforward. A typical example is that in digital communications; training or pilot signals are often contained regularly or periodically in transmitted signals to facilitate channel estimation at the receiving end at the expense of bandwidth. Other than digital communications, such an arrangement usually cannot apply to other fields such as seismic exploration, speech analysis, ultrasonic nondestructive evaluation, texture image analysis and classification, etc. These therefore necessitate exploration of blind equalization and system identification, which have been challenging research areas for a long time and continue to be so.
1st Edtion
1-84628-218-7
NONE
Blind Equalization and System Identification
Mathematics
English
Springer-Verlag London Limited
2006
London
1-479
LOADING LIST...
LOADING LIST...