Design and implementation of multiplier-less tunable 2-D FIR filters using McClellan transformation

This paper proposes new structures for realizing tunable 2D fan and elliptical filters with different spectral characteristics using McClellan transformation. The 1D prototype is a variable digital filter obtained from the interpolation of a set of desirable impulse responses and is implemented using the Farrow structure. The coefficients of the sub-filters and the transformation parameters in the Farrow structure are represented in SOPOT (sum-of-powers-of-two), which can be easily implemented as simple shift-and-add operations. Furthermore, the transformation part can be shared between the sub-filters, leading to significant saving in hardware complexity. Several design examples are given to demonstrate the effectiveness and feasibility of the proposed approach.