Date of Award
Electrical and Computer Engineering
Kwan, Hon Keung
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Digital filters can be divided into finite impulse response (FIR) digital filters and infinite impulse response (IIR) digital filters. Evolutionary algorithms are effective techniques in digital filter designs. One such evolutionary algorithm is Cuckoo Search Algorithm (CSA). The CSA is a heuristic algorithm which emulates a special parasitic hatching habit of some species of cuckoos and have been proved to be an effective method with various applications. This thesis compares CSA with Park-McClellan algorithm on linear-phase FIR Type-1 lowpass, highpass, bandpass and bandstop digital filter design. Furthermore, a multiobjective Cuckoo Search Algorithm (MOCSA) is applied on general FIR digital design with a comparison to Non-dominated Sorting Genetic Algorithm III (NSGA-III). Finally, a constrained multiobjective Cuckoo Search Algorithm is presented and used for IIR digital filter design. The design results of the constrained MOCSA approach compares favorably with other state-of-the-art optimization methods. CSA utilizes Levy flight with wide-range step length for the global walk to assure reaching the global optimum and the approach of local walk to orientate the direction toward the local minima. Furthermore, MOCSA incorporates a method of Euclidean distance combing objective-based equilibrating operations and the searching for the optimal solution into one step and simplifies the procedure of comparison.
Liang, Jiajun, "Digital Filter Design Using Multiobjective Cuckoo Search Algorithm" (2018). Electronic Theses and Dissertations. 7431.