High Performance Computing for Radar Signal Processing Acceleration
Abstract
Radar processing, remote sensing, and object detection represent a class of Army applications. Those future army applications use high definition images, videos and context information to provide assistance to soldiers to provide lethality protection. Those floating point algorithms are difficult to accelerate on general purpose processors with limited power consumption. Traditionally, radar signal processing application were processed by high performance computers off-line. Therefore, designers are challenged to develop hardware by exploring their design space in terms of excessive storage cost, limited bandwidth, and a restrictive battery power. Hence, the traditional radar signal processing should be addressed to meet the dynamic nature of advances in unmanned vehicle technologies. Our solution to this problem is to use a High Performance Computing environment augmented with Field Programmable Gate Arrays by addressing the problem of executing radar signal processing applications directly to FPGA co-processor architectures. Our methodology demonstrates order of magnitude speedup for radar signal applications, showing the effectiveness of our framework for both development and testing of scientific algorithms. This proposal presents a reconfigurable acceleration environment dedicated to radar signal processing and remote sensing. We believe that the Synthetic Aperture Radar is a representative remote sensing application where our proposed methodology is able to help designers to build efficient systems. We also present the case study of co-processor implementation with Discrete Wavelet Transform for speckle reduction in SAR. As an outcome of this research, the institution will be able to demonstrate acceleration up to 10-100x for various radar signal processing. In this proposal, we suggest the acquisition of a high performance computing instrument to support various remote sensing projects.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 12, 2017
- Source ID
- W911NF1610217
Entities
People
- Youngsoo Kim
Organizations
- Army Contracting Command
- San José State University
- United States Army