Design of quadrature mirror filter bank using polyphase components based on optimal fractional derivative constraints

Abstract

This paper proposes an optimal design of two-channel quadrature mirror filter (QMF) bank using optimized polyphase component enforcing very efficient global swarm optimization techniques cuckoo search (CS) and modified cuckoo search (MCS) based fractional derivative constraints. The objective function is structured in frequency domain, as sum of energy of errors in passband and stopband. This objective function is optimized using a well recognized true least square method called Lagrange multiplier method enforcing CS and MCS based fractional derivative constraints and perfect reconstruction condition of two-channel QMF bank as a constraint. Fractional derivative constraints are optimized by employing CS and MCS algorithms taking peak reconstruction error (PRE) of two-channel QMF bank as an objective function. The proposed method is evaluated by several performance attributes of two-channel QMF bank which promulgate the excellence of proposed method. Comparative results clearly show the superiority of proposed method over recently developed techniques.