Processing of SAR data based on the heterogeneous architecture of GPU and CPU

SAR imaging process is usually computationally burdensome and it is difficult to achieve SAR image in real-time. Recently, the rapid increase in the performance of Graphics Processing Units (GPU), coupled with powerful CPU, has made heterogeneous framework of GPU and CPU into a compelling platform for computationally demanding tasks. This paper presents the parallel implementation on the heterogeneous architecture using Chirp Scaling algorithm. Experimental results suggest that real-time imaging processing is possible for large-scale SAR data.