<p>System identification plays a crucial role in various fields like control systems, communication networks, biomedical signal processing, and many more. Among the different system identification techniques, infinite impulse response (IIR) system offer flexibility in capturing complex dynamics. However, accurately estimating the parameters of an IIR model can be challenging due to its inherent nonlinearity and potential instability. This paper presents a novel method for IIR system identification utilizing an enhanced arithmetic optimization algorithm (IAOA) in order to overcome this problem. The IAOA leverage the strengths of evolutionary computation and numerical optimization techniques to improve the precision and efficacy of the parameter estimation process. By combining concepts from genetic algorithms, particle swarm optimization, and simulated annealing, the proposed algorithm aims to overcome the limitations of traditional optimization methods and provide a more robust and effective solution. The performance of the IAOA is evaluated through comprehensive comparisons and simulations with existing optimization methods on various benchmark IIR system identification problems. The results demonstrate its superiority in terms of parameter estimation and convergence.</p>