Optimum duration discrete-time wavelets
Joel M. Morris · Optical Engineering · 1997
We present a class of discrete-time orthonormal wavelets that are optimized with respect to time duration. The global optimization is achieved via an adaptive simulated annealing (ASA) approach, and the objective function is the average time duration over all scales of the wavelet as defined by the levels of the corresponding binary, tree- structured, filter bank structure. We present and discuss the algorithm for generating these optimum wavelets; present the defining filter coeffi- cients $g(n)% for filter lengths ofN 5 8, 10, 12, 14, 16, 18, 24, and 32; and provide comparisons with Daubechies' discrete wavelets and other discrete and discrete-time wavelets optimized to other measures based on duration and bandwidth. © 1997 Society of Photo-Optical Instrumentation Engi- neers. (S0091-3286(97)00304-8)