Real-time Implementation of Joint Domain Localised Algorithm for High Frequency Surface Wave Radar using GPU

Abstract

The performance of target detection and tracking is primarily limited by ionospheric interference in High Frequency Surface Wave Radar (HFSWR). Joint Domain Localised (JDL) has been proved to be an effective algorithm for ionospheric clutter suppression in HFSWR. However, the implementation of JDL in the traditional CPU platform cannot afford the real-time requirement in HFSWR. With the help of the tremendous parallel computational horsepower in GPU, in this paper we investigate the real-time implementation of JDL algorithm for HFSWR using Graphics Processing Unit (GPU). We also perform a comparative analysis in terms of the performance using the CPU-based implementation and the GPU-based implementation. Experimental result shows that the GPU-based implementation accelerates the computation by over 24.72 times as compared to the CPU-based implementation which meets the real-time requirement of HFSWR.