Comparative Study on Error in MIMO Radar DOA Estimation

Abstract

Hardware implementation for faster real-time sig- nal processing in multiple-input multiple-output (MIMO) radar beamforming by estimating direction of arrival (DOA) using multiple signal classification (MUSIC) is attractive. The above subspace-based method works with a basic principle of eigenvalue decomposition and is found suitable for application-specific in- tegrated circuit (ASIC), field-programmable gate array (FPGA). In that regard, the cyclic Jacobi method is used for computing eigenvalues and eigenvectors, by orthonormal plane rotations in accomplishing eigenvalue decomposition. However, the fasterness and foremost, the accuracy of the DOA estimator realized in such above hardware depends on the involved number of antenna elements in an array, the number of snapshots, and the signal-to- noise ratio. Therefore, the comparative study on DOA estimation error on the above mentioned parameters in MUSIC and QR- based algorithms is presented in this paper