Title :
Detection of continuous-phase modulated (CPM) signals with artificial neural networks
Author :
Grossekatthoefer, Alwin
Author_Institution :
Alcatel Telecom Deutschland, Stuttgart, Germany
Abstract :
A coherent receiver for CPM signals with RC (raised-cosine) pulse shaping is introduced. The receiver mainly consists of a multilayer ANN (artificial neural network) that evaluates the first derivative of the phase of the IF (intermediate frequency) signal. After training the ANN transforms the signal phase into the appropriate digital symbols. Further an easy to implement symbol synchronisation scheme is introduced. The symbol synchronisation also exploits the differentiated phase of the received signal. Therefore, the coherent receiver, does not need any extra synchronisation signal. Both applications are described in detail and a performance analysis is presented
Keywords :
continuous phase modulation; learning (artificial intelligence); multilayer perceptrons; signal detection; synchronisation; telecommunication computing; CPM signals; IF signal; artificial neural network; artificial neural networks; coherent receiver; continuous-phase modulated signals; digital symbols; intermediate frequency signal; multilayer ANN; multilayer perceptron; performance analysis; raised cosine pulse shaping; received signal; signal detection; signal phase; symbol synchronisation; training; Amplitude modulation; Amplitude shift keying; Artificial neural networks; Baseband; Frequency synchronization; Maximum likelihood estimation; Phase modulation; Phase shift keying; Pulse shaping methods; Quadrature amplitude modulation;
Conference_Titel :
Global Telecommunications Conference, 1996. GLOBECOM '96. 'Communications: The Key to Global Prosperity
Conference_Location :
London
Print_ISBN :
0-7803-3336-5
DOI :
10.1109/GLOCOM.1996.594357