Implementation of parallel optimization for a broadband adaptive beamforming algorithm on localized GPUs

The computational resources and energy carried by Autonomous Undersea Vehicle (AUV) are limited, failing to meet the requirements of real-time operation of the adaptive algorithm in the sonar system. To address these issues, a set of broadband signal parallel processing system based on the adaptive beamforming algorithm is designed and implemented through the utilization of a local GPU. The simulation results indicate that, in contrast to CPU and DSP, GPU has a better calculated energy consumption ratio and calculated speed-up ratio. In the case of complex beamforming calculations, the system demonstrates higher real-time performance and lower energy consumption. Once the system is installed on the AUV platform, it can not only reduce the energy consumption of the platform, enhance the calculation efficiency, but also save some space and weight. Additionally, the design of this system is close to the real scenario and holds the significance of engineering realization.