-Intelligent Reflecting Surfaces are supposed to bring about the next generation of wireless communication systems--especially those at 5G-level and beyond. The IRS system is made up of a large number split passive-effecting element that can change the phase of signals striking them. This enables intelligent control over signal propagation. The benefit is a vastly improved cover age and transmit rate for wireless networks. However, to really reap the benefits from IRS technology, we need a means of estimating the Channel State Information (CSI) accurately. So far, this remains a formidable obstacle because IRS components are all passive. By means of MATLAB-based simulations, the solutions are assessed in different situations: What is the impact of SNR level on the performance? How does user density influence the outcomes? In particular, how does a difference antenna array configuration affect thing, and what kind of pilot signal allocation gets mapped into them. Using the metric Normalized Mean Square Error (NMSE), we judged performances. The MLSR method tends to be even superior to the ML-IRS approach in low-SNR environments, as illustrated by simulation results. Moreover, the study investigates the impact of both pilot availability and channel sparsity on the accuracy of re estimation. Accurate CSI achievement requires the use of adaptive pilot strategies together with sparse channel techniques to minimize necessary pilot overheads. The research introduces three future directions for advancements, namely machine learning integration in channel estimation and hybrid analog-digital IRS development, and wideband millimeter-wave (mmWave) and terahertz frequency band extensions.