Implementation and Performance of Factorized Back projection on Low-Cost Commercial-Off-the-Shelf Hardware
Abstract
Traditional Synthetic Aperture Radar (SAR) systems are large, complex, and expensive platforms that require significant resources to operate. The size and cost of the platforms limits the potential uses of SAR to strategic level intelligence gathering or large budget research efforts. The purpose of this thesis is to implement the factorized back projection SAR image processing algorithm in the C ++ programming language and test the code's performance on a low cost, low size, weight, and power (SWAP) computer : a Raspberry Pi Model B. For a comparison of performance, a baseline implementation of filtered backprojection is adapted to C from pre-existing MATLAB code. The factorized back projection algorithm shows a computational improvement factor of 2-3 compared to filtered backprojection. Execution on a single Raspberry Pi is too slow for real-time imaging. However, factorized backprojection is easily parallelized, and we include a discussion of parallel implementation across multiple Pis.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2016
- Accession Number
- AD1053861
Entities
People
- Alec S Rasmussen
Organizations
- Air Force Institute of Technology