Adaptive Array Design using Least Mean Square Algorithm

Abstract

This paper presents the designing and testing of an 8-element linear array for Adaptive Antenna applications using the Least Mean Square (LMS) algorithm towards improving the directive gain of the array. A conventional patch antenna is optimized to operate at 2.35 GHz (4G applications) and this design is extended up to 8 elements using CST Microwave Studio parameterization. The S-parameters, Return Loss, Gain and VSWR of the antenna array are studied for the 2, 4, and 8 elements adaptive array. The simulation results are validated on the hardware setup and found closely matching with the experimental results. The resulting eight-element antenna array geometry is optimized with a coaxial feeding technique. This geometry appears promising in improving the gain from 6.13 to 23.5 dBi for a single element to eight elements respectively. Further, the LMS algorithm is used to compute the optimal complex weights considering different angles for desired User (60o and 30o) and Interferer (10o and 15o) during MATLAB simulation and then these optimal weights are fed to antenna elements using CST for beam steering in a different direction. Maximas are obtained at 54o and 28o when nulls are at 10o and 15o using CST software which is closely matching with MATLAB results.