Programming Real-Time Autofocus on a Massively Parallel Reconfigurable Architecture Using Occam-pi

Recently we proposed occam-pi as a high-level language for programming massively parallel reconfigurable architectures. The design of occam-pi incorporates ideas from CSP and pi-calculus to facilitate expressing parallelism and reconfigurability. The feasability of this approach was illustrated by building three occam-pi implementations of DCT executing on an Ambric. However, because DCT is a simple and well-studied algorithm it remained uncertain whether occam-pi would also be effective for programming novel, more complex algorithms. In this paper, we demonstrate the applicability of occam-pi for expressing various degrees of parallelism by implementing a significantly large case-study of focus criterion calculation in an auto focus algorithm on the Ambric architecture. Auto focus is a key component of synthetic aperture radar systems. Two implementations of focus criterion calculation were developed and evaluated on the basis of performance. The comparison of the performance results with a single threaded software implementation of the same algorithm show that the throughput of the two implementations are 11x and 23x higher than the sequential implementation despite a much lower (9x) clock frequency. The two designs are, respectively, 29x and 40x more energy efficient.