Optimum Design of Linear Phase Paraunitary Filter Bank & its Applications in Signal Processing

Filter Banks plays crucial role in signal processing and image processing as subband processing gives dominant results in time critical applications. In formal years, various Para unitary Linear Phase Filter Banks are proposed by following conventional and computational complex factorization and lattice approaches consisting of complex nonlinear optimization problems. One of the recent methods to design Filter Bank having properties of Linear Phase and Paraunitary is via Singular value decomposition technique which leads to optimum results compared to existing methods as most of the time it deals with matrix operations. In this paper, design benchmark is evaluated as two dominant optimization queries and reasonable key of each optimization query is solved by performing Singular Value Decomposition. Proposed Paper discusses linear phase condition of filter banks satisfying mirror image symmetry at analysis side and perfect reconstruction property at synthesis side. Singular Value Decomposition approach leads to fast and efficient simulation results compared to existing filter banks designs. Proposed method of filter bank design deals with any arbitrary channels and every length of the filters