Bayesian Inspired Grid Search for AoA Estimation

  • 31



Name of the Speaker: Mr. Akshay Sharma(EE18S008)
Guide: Dr. Sheetal Kalyani
Venue/Online meeting link:
Date/Time: 31st March 2023 (Friday) at 11:00 AM to 12:00 PM

Angle of Arrival (AoA) estimation has been of great utility to communication systems. Ranging from military applications like enemy aircraft detection to civilian use cases like targeted advertising, AoA estimation has shown widespread potential. In this thesis a method called BayesOptAoA is proposed for estimating AoA. The proposed method is inspired from Bayesian optimization, where new set of parameters are updated based upon previous parameter evaluations thus making the proposed method 'smart' in a sense. BayesOptAoA is compared with three existing algorithms- Sparse Bayesian Learning (SBL) algorithm, Expectation Maximization (EM) and Space Alternating Generalized EM (SAGE). Simulation results show that the proposed method is not sensitive towards initialization, unlike EM or SAGE and is at par with SBL.