Abstract—Intelligent Reflecting Surfaces (IRS) have emerged as a promising technology for enhancing wireless communication by smartly propagating signals with large, reconfigurable meta-surfaces. While recent foundational literature has benchmarked IRS performance under ideal, uncoupled physical models, practical deployments inevitably face non-idealities, such as mutual electromagnetic coupling between adjacent elements. In this work, we present a simulation framework that extends baseline models to account for mutual coupling in large IRS arrays. Our approach enables a quantitative comparison of power and energy efficiency performance between coupled-IRS, uncoupled (ideal) IRS, and conventional Decode-and-Forward (DF) relaying, directly reproducing and extending results from seminal literature. Key results demonstrate that, for example, with strong mutual coupling (γ = 0.9), IRS array gain is reduced by up to 70%, and the required transmit power to achieve a target rate increases by over 10 dB compared to the ideal uncoupled case. The crossover point—where IRS outperforms DF relaying—shifts to IRS surfaces with more than 100 elements or to substantially higher transmit power levels, as quantified in our energy efficiency and rate simulations. These findings provide actionable numerical benchmarks for practical IRS design.  
 
Keywords—intelligent reflecting surface, reconfigurable intelligent surface, decode-and-forward relaying, mutual coupling, wireless simulation, energy efficiency, transmit power, large intelligent surfaces, meta surface modeling, relay systems