Date of Award


Degree Type


Degree Name



Electrical and Computer Engineering

First Advisor

Kwan, Hon


Artificial Intelligence; Digital filters; Filter design; Firefly algorithm; Neural Network; Optimization




Firefly algorithm is an evolutionary algorithm that can be used to solve complex multi-parameter problems in less time. The algorithm was applied to design digital filters of different orders as well as to determine the parameters of complex neural network designs. Digital filters have several applications in the fields of control systems, aerospace, telecommunication, medical equipment and applications, digital appliances, audio recognition processes etc. An Artificial Neural Network (ANN) is an information processing paradigm that is inspired by the way biological nervous systems, such as the brain, processes information and can be simulated using a computer to perform certain specific tasks like clustering, classification, and pattern recognition etc. The results of the designs using Firefly algorithm was compared to the state of the art algorithms and found that the digital filter designs produce results close to the Parks McClellan method which shows the algorithm’s capability of handling complex problems. Also, for the neural network designs, Firefly algorithm was able to efficiently optimize a number of parameter values. The performance of the algorithm was tested by introducing various input noise levels to the training inputs of the neural network designs and it produced the desired output with negligible error in a time-efficient manner. Overall, Firefly algorithm was found to be competitive in solving the complex design optimization problems like other popular optimization algorithms such as Differential Evolution, Particle Swarm Optimization and Genetic Algorithm. It provides a number of adjustable parameters which can be tuned according to the specified problem so that it can be applied to a number of optimization problems and is capable of producing quality results in a reasonable amount of time.