Design and Implementation of DWMT Transmission Systems using IIR Wavelet Filter Banks

Abstract

Multicarrier modulation using orthogonal frequency division multiplexing (OFDM) is often recognized as an efficient solution for wireless communications. However, waveform bases other than sinusoidal functions can similarly be used for multicarrier systems in order to provide an alternative to OFDM. In this context, DWMT transmission systems are suggested. It is known that such systems suffer obviously from the two main problems of Inter Symbol Interference (ISI) and Inter-Carrier Interference (ICI). ISI may be considerably reduced via channel equalization. The effect of ICI can be reduced using high order Finite Impulse Response(FIR)wavelet filter banks, but at the expense of increased system complexity. Those FIR banks which are required to reduce effect the ICI should have autocorrelation functions approaching an impulse. On other hand, Infinite Impulse Response(IIR) filter banks can be used to replace those FIR counterparts to increase sub-band discrimination while reducing the required filter order, i.e., decreasing the complexity of system. In this paper, a special type of IIR filter banks is designed and implemented for such purpose. Bi-reciprocal lattice wave digital filters (BLWDFs) are utilized in an approximate linear phase design of 11th order IIR wavelet filter bank (FB) for discrete wavelet multi-tone (DWMT) transmission systems. The performances of the proposed DWMT for transmission over simulated wireless channels with additive white Gaussian noise (AWGN) are examined and compared with one of the recent FIR designs. Significant improvements over the classical DWMT approach are observed.