Abstract
Large Intelligent Surfaces (LiSs) are a major forerunner in the field of communication research. These surfaces are integral constituents employed in the present generation of wireless communication systems. Essentially, they are a tangible set of intelligent elements stacked together to create and route highly directional signals. These active electromagnetic surfaces are molded in any intended way, say as smart reflectors, in an attempt to mitigate blockages, improve coverage and other figures of merit such as signal to interference ratio of the system.Obtaining optimal reflection coefficients of the LiS requires full channel knowledge of the transmitter and the receiver. With increasing reflector array size (LiS), this challenge involves training overhead and complex hardware and higher power consumption otherwise. This project proposes a low complexity algorithm for configuring LiS that only requires partial channel knowledge. We show via simulations that with only partial channel knowledge, performance measures closer to that obtained by full channel knowledge can be obtained. We discuss and simulate different channel estimation algorithms and verify the behavior of the communication system in various cases.