Title :
Orthogonal Neural Network Based Predistortion for OFDM Systems
Author :
Rodriguez, Nibaldo ; Cubillos, Claudio
Author_Institution :
Pontifical Catholic Univ. of Valparaiso, Valparaiso
Abstract :
This paper proposes a predistortion scheme based on an orthogonal hidden layer feedforward neural network for reducing nonlinear distortion introduced by a traveling wave tube amplifier (TWTA) over orthogonal frequency division multiplexing (OFDM) signals. In predistorter, the inputs weight are fixed and based on this the output weights are analytically determined. Computer simulation results confirm that once the 16QAM-OFDM signals are predistorted and amplified at an input back-off level of 0 dB there is a bit error rate performance very close to the ideal case of linear amplification.
Keywords :
OFDM modulation; amplifiers; error statistics; feedforward neural nets; quadrature amplitude modulation; telecommunication computing; travelling wave tubes; 16QAM-OFDM signals; bit error rate; nonlinear distortion reduction; orthogonal frequency division multiplexing; orthogonal hidden layer feedforward neural network; traveling wave tube amplifier; Bit error rate; Constellation diagram; Feedforward neural networks; Informatics; Neural networks; Nonlinear distortion; OFDM; Power system modeling; Predistortion; Robots;
Conference_Titel :
Electronics, Robotics and Automotive Mechanics Conference, 2007. CERMA 2007
Conference_Location :
Morelos
Print_ISBN :
978-0-7695-2974-5
DOI :
10.1109/CERMA.2007.4367690
