The Georgia Tech digital signal multiprocessor

Techniques for implementation of digital signal processing (DSP) algorithms on synchronous multiprocessors studied for the past ten years are discussed. This research is characterised as a theoretical investigation of optimality criteria for the implementation of DSP algorithms and the development of a unique approach that can achieve these optimality criteria. A result of this ongoing research program has been the development of a set of formal techniques for the automatic realization of optimal synchronous supercomputer implementations for a large class of iterative or recursive DSP algorithms. Three experimental laboratory digital signal microprocessor (DSMP) systems designed to provide experimental verification of the theoretical results are described. This has provided important feedback to drive the theoretical developments. Compilers that take flow graph specifications of algorithms and generate optimal multiprocessor code are also discussed