Parameterless self organizing map for improving OFDM system performance

Abstract

This paper presents a new adaptive neural approach to improve the system performance of an Orthogonal Frequency Division Multiplexing (OFDM) based receiver structure. Peak to average power ratio (PAPR) of an OFDM signal is very large. When such a signal passes through high power amplifier, the signal gets nonlinearly distorted. This causes bit error rate (BER) degradation of the OFDM system and also results in spectral regrowth. Multipath fading effects also deteriorate the signal to high extent. Our proposed system takes care of both these factors and thus improves overall performance of the system.