Comparative Performance Analysis of IIR and FIR Filters for 5G Networks

Abstract

This paper comprehensively quantifies and compares Infinite Impulse Response (IIR) and Finite Impulse Response (FIR) bandpass filters that can optimize bandwidth allocation by eliminating noise, side lobes and Intersymbol Interference (ISI) in Orthogonal Frequency Division Multiplexing (OFDM) systems. Our goal is to choose a feasible filter that can accommodate more carriers in a bandwidth, so that the spectral efficiency can be increased. For IIR filtering, we employ, namely, Butterworth, Chebyshev and Elliptic, while for FIR Equiripple, Bohman and Hamming are studied. We evaluate these filters in terms of magnitude response and group delay, and also identify the minimum filter order that characterized nearly to an ideal filter response. The results show that the IIR filter has steep transition region when compared to the FIR filter with the same order, but the trade-off is high group delay. Due to the fastest roll-off, small transition region and low cost, we found that Elliptic filter is a good candidate to be implemented in the filtered-OFDM (f-OFDM) 5G system.