Indonesian Journal of Electrical Engineering and Computer Science Vol. 31, No. 1, July 2023, pp. 170~179 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v31.i1.pp170-179  170 Journal homepage: http://ijeecs.iaescore.com Underdetermined direction of arrival estimation for multiple input and multiple outputs sparse channel based on Bayesian learning framework Anughna Narayanaswamy, Ramesha Muniyappa Department of Electrical, Electronics, and Communication Engineering, GITAM University Bengaluru, Bengaluru, India Article Info ABSTRACT Article history: Received Sep 27, 2022 Revised Feb 17, 2023 Accepted Feb 20, 2023 Direction of arrival (DOA) estimation for a sparse channel has attracted serious attention recently. Better signal analysis and denoising achieve accuracy in DOA determination. This paper proposes an underdetermined DOA estimation for multiple input and multiple outputs (MIMO) sparse channels. A novel multi-kernel-based non-negative sparse Bayesian learning (MK NNSBL) framework is implemented using the multiplied form of basis vector within the manifold matrix for a defined grid. Meanwhile, virtual antenna locations are reconfigured by exploiting the conventional cuckoo search algorithm (CCSA) for the fine reception of incoming signals on a non- uniform linear array (NULA). The simulated results reveal that the novel approach outperforms in its optimal root mean square error (RMSE) for various signal-to-noise ratio (SNR) limits and the compilation time. The convergence comparative graph indicates the improved performance in the proposed framework over existing algorithms. Keywords: CCSA Direction of arrival MIMO NNSBL Non-uniform linear array This is an open access article under the CC BY-SA license. Corresponding Author: Anughna Narayanaswamy Department of Electrical, Electronics, and Communication Engineering, GITAM University Bengaluru Bengaluru, India Email: anughna.7@gmail.com 1. INTRODUCTION Wireless mobile communication plays a vital role in today’s life. To satisfy the stringent demands of dense users’ requirements such as higher data rates, signal-to-noise ratio (SNR), and high accuracy with fewer errors different generations of cellular networks have evolved right from early first-generation (1G) to current fourth generation (4G). Now, the fifth generation (5G) network is the new version of the mobile communication system which is commercialized to deploy for usage. Nowadays, the need for direction of arrival (DOA) estimation is increasing rapidly day by day in 5G wireless mobile communication systems, radar, sonar, electronic surveillance, seismology, re-configurable intelligent surfaces (RIS), and medical diagnosis. DOA estimation is performed virtually using computers rather than the manual method to avoid the need for physical adjustment of an antenna and extra phase shifter for beam steering. DOA estimation means finding the exact direction of transmitted electromagnetic signals that impinges on the receiving antenna elements in an array over a noisy channel. In general, DOA estimation techniques have three main classifications such as spectral estimation, parametric subspace-based estimation (PSBE), and deterministic parametric estimation (DPE) [1]. The popular parametric subspace-based methods multiple signal classification (MUSIC) and estimation of signal parameters via rotational invariance technique (ESPRIT) [2] are implied to a greater extent among others, but sparse Bayesian learning (SBL) is a popular method that can be used for sparse signal recovery (SSR) in compressive sensing (CS) [3]. The MUSIC and ESPRIT algorithms comparison based on implementation time