Design and Implementation of a Low-Complexity Continuously Variable Digital Filter Using a Novel Farrow-Equivalent-Newton Structure-Based Fractional Delay Filter

Abstract

Variable filters with adjustable bandwidth are vital components in diverse communication scenarios. This paper presents an innovative architecture for a continuously variable bandwidth filter using a fixed hardware. Our approach integrates a fixed finite impulse response filter between two arbitrary fractional delay filters implemented through a novel Farrow-equivalent-Newton structure. The proposed architecture provides a low-complexity implementation structure compared to the state-of-the-art approaches. A precise mapping equation for the edge frequencies of the filters generated from the proposed continuously variable bandwidth filter, in terms of a variable parameter called the resampling ratio, is also formulated. Validation experiments encompass the design of continuously variable bandwidth filters tailored to various wireless communication standards. The hardware utilization report of the proposed continuously variable bandwidth filter obtained by synthesizing the structure using Xilinx Vivado 2020.2 on a Kintex-7 device is also included, which proves the hardware complexity reduction and efficiency of the proposed structure.