Open AccessMixture Modeling based Multikernel Sparse Learning for Directional of Arrival Estimation
Author(s) -
Awab Fakih,
S Shashidhar
Publication year - 2019
Publication title -
international journal of engineering and advanced technology
Language(s) - English
Resource type - Journals
ISSN - 2249-8958
DOI - 10.35940/ijeat.f1132.0886s19
Subject(s) - underdetermined system , computer science , matlab , convergence (economics) , direction of arrival , algorithm , maximum likelihood , signal (programming language) , maximization , pattern recognition (psychology) , speech recognition , mathematics , artificial intelligence , mathematical optimization , statistics , telecommunications , antenna (radio) , economics , programming language , economic growth , operating system
Direction of Arrival (DOA) estimation problem is defined as the problem of Sparse Signal Recovery (SSR) in researches published on the Uniform or Non Uniform array based implementations. This Paper attempts a Multikernel Sparse learning (MSL) approach with mixture modeling for the SSR problem to improve the performance parameters including the PSNR and the RMSE of the estimated sparse signal in the underdetermined condition. The Expectation Maximization algorithm is exploited to obtain the convergence in the mixture modeling MSL method. The virtual array response problem thus developed uses the mixture modeling MSL to estimate the DOA. Matlab based implementation is carried out and the results are found to be satisfactory.