Performance Improvement of the DWT-OFDM System Using Convolutional Coding

OFDM is an ideal high-speed data transmission scheme using closed space carriers that can be modulated at low data rate. Conventional OFDM systems employ Fast Fourier Transform method for multiplexing signals over the subcarriers. In order to reduce inter-symbol interference, a cyclic prefix is appended before transmission. However, adding a cyclic prefix reduces data throughput and consequently decreases power efficiency. One option to overcome these problems is to use a wavelet transform popularly known as the Discrete Wavelet Transform - Orthogonal Frequency Division Multiplexing (DWT-OFDM). DWT-OFDM outperforms the Fast Fourier Transform - OFDM (FFT-OFDM) by improving the spectral efficiency without introducing cyclic prefixes. Previous studies that investigated the performance of different OFDM systems did not consider integrating coding techniques for error detection and correction. In this paper, a convolutional-coded DWT-OFDM is proposed to improve its bit error rate performance over an uncoded system. Convolutional encoder paired with Viterbi hard-decision decoder blocks were added to the typical DWT-OFDM system. Uncoded and coded DWT-OFDM systems with different code rates have been simulated in order to verify that an improvement in the bit error rate performance has been achieved.