Publication

Abstract : A variety of applications in digital signal processing require hardware-efficient digital filters to have the property of tunability over a wide range of pass-band edge frequencies. This paper proposes to employ Newton structure-based Lagrange interpolators to increase the hardware efficiency of the tunable filters. The variable pass-band filter structure consists of a filter with a fixed pass-band frequency placed between two Lagrange interpolators realized using the Newton structure. The variations in the pass-bands and bandwidths are achieved by tuning a sample rate conversion factor without altering the coefficients and the structure, making this structure reconfigurable and hardware efficient. The experimental results show that the proposed tunable filter based on the Newton structure can achieve better hardware efficiency with fewer adders and multipliers with the same performance as the latest existing structures. The proposed architecture is simulated and synthesized in Xilinx Kintex-7 FPGA (xc7k160tfbg484-1). It is observed from the synthesis results that the proposed architecture results in a significant improvement in the hardware resources and power consumption when compared to the Farrow-based implementation.