DocumentCode
1606956
Title
A new artificial neural network based adaptive non-linear equalizer for overcoming co-channel interference
Author
Hussain, Amir ; Soraghan, J.J. ; Durrani, T.S.
Volume
2
fYear
1996
Firstpage
1422
Abstract
This paper describes a new approach for the decision feedback equalizer (DFE) based on the functional link neural network. The new structure termed the decision feedback functional link equalizer (DFFLE) is applied to the problem of adaptive equalization in the presence of ISI, additive white noise and co-channel interference. It is shown through simulation results for a severe amplitude distorted co-channel system, that the DFFLE provides significantly superior BER performance characteristics compared to the conventional linear transversal equalizer (LTE) and the non-linear radial-basis function neural network and feedforward functional link equalizer (FFLE) based structures
Keywords
Gaussian noise; adaptive equalisers; cochannel interference; coding errors; decision feedback equalisers; error statistics; interference suppression; neural nets; telecommunication computing; white noise; AWGN; BER performance characteristics; DFE; ISI; adaptive equalization; adaptive nonlinear equalizer; additive white noise; amplitude distorted cochannel system; artificial neural network; cochannel interference; decision feedback equalizer; decision feedback functional link equalizer; feedforward functional link equalizer; functional link neural network; linear transversal equalizer; nonlinear radial basis function neural network; simulation results; Adaptive equalizers; Adaptive systems; Additive white noise; Artificial neural networks; Bit error rate; Decision feedback equalizers; Interchannel interference; Intersymbol interference; Neurofeedback; Nonlinear distortion;
fLanguage
English
Publisher
ieee
Conference_Titel
Global Telecommunications Conference, 1996. GLOBECOM '96. 'Communications: The Key to Global Prosperity
Conference_Location
London
Print_ISBN
0-7803-3336-5
Type
conf
DOI
10.1109/GLOCOM.1996.587680
Filename
587680
Link To Document