Broadband Beamforming Using the Least Square Method Improved via Adaptive Diagonal Loading

Abstract

The least square method (LSM) is a commonly used beamforming method. However, it limits the beamforming frequency bandwidth and is vulnerable to noise; therefore, diagonal loading is used to overcome these limitations. The diagonal loading proposed in this study can be used for broadband beamforming in cases in which noise does not exist, and it makes the loading factor adaptive to a singular value distribution. The simulation results show that the proposed method works stably above the beamforming frequency limit, even for an extended beamforming frequency. We expect that the proposed method can be used to form frequency-invariant beam patterns used in monopulse algorithms to detect signals of unknown frequencies.