Design of DSP‐based robust antenna beamformer using novel MVDR algorithm: Extension to SLL reduction

• Published in: International journal of RF and microwave computer-aided engineering Vol. 32; no. 4
• Main Authors: Duppala, Venkata Ramana, Misra, Iti Saha, Sanyal, Salil K.
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01.04.2022
• Subjects: Algorithms; Antennas; Beamforming; Complexity; diagonal loading; Digital signal processing; digital signal processor; Digital signal processors; Direction of arrival; Fourier transforms; Frequency shift; Microprocessors; MVDR; Real time operation; Robustness; side lobe level reduction; Sidelobes; space selective digital filter; space shifting
• ISSN: 1096-4290; 1099-047X
• DOI: 10.1002/mmce.23039

The mismatch between the actual direction of arrival (DOA) of signal of interest (SOI) and the estimated DOA drastically deteriorates the performance of the antenna beamformer. Diagonal loading (DL) is a popular method to improve the robustness of the antenna beamformer against such mismatch. However, the evaluation of optimum diagonal load level (DLL) involves much complexity. This paper proposes a simple and user parameter free novel data dependent algorithm for the development of minimum variance distortion less response (MVDR) based robust antenna beamformer which provides uniform output SINR over the entire range of mismatch. The behavior of the beamformer for varying DLL has been studied and observed the fact that, under optimum DLL condition, the main lobe of beam pattern is focused towards the estimated DOA of SOI within a mismatch of 40. The novelty includes the logical exploitation of the analogy between the frequency and space domains Fourier Transforms followed by the frequency shift property and convolutions with necessary modifications. Next, the proposed algorithm has been extended to minimize the side lobe level (SLL) below −35 dB, in addition to achieving the desired robustness. Another advantage of this algorithm is the requirement of minimum additional computational burden of M complex multiplications and one convolution. The implementation of the robust antenna beamformer has also been done through the Digital Signal Processor TMS320C5416 supported with Code Composer Studio (CCS) for real time operations. Extensive comparative simulations have been included to justify the superiority of the proposed algorithm over the existing algorithms.