Adaptive IIR model identification using chaotic opposition-based whale optimization algorithm

Infinite impulse response (IIR) filter system recognition is a serious issue nowadays as it has many applications on a diversity of platforms. The whale optimization algorithm (WOA) is a novel nature-motivated population-based meta-heuristic algorithm where the hunting techniques of humpback whales are implemented to solve many optimization problems. But the main disadvantage of WOA is its stagnant convergence rate. As the algorithm is population based, the initialization process is very important in finding the best result and to enhance the convergence rate. In this paper, a novel chaotic oppositional-based initialization process is nominated before the start of conventional WOA to improve the performance. To effectively cover the entire search region, a chaotic-based logistic population map consists of both the actual numbers and its corresponding opposite numbers are incorporated into this opposition-based initialization process. When checked out with some classic model of examples, simulation performance authorizes chaotic oppositional-based whale optimization algorithm (COWOA) as a more convenient contender compared to the other evolutionary techniques in terms of accuracy and convergence speed. Convergence profile and mean square error are the performance specifications that are needed to inspect the performance of our recommended algorithm.