Simulation of communication time for a space-time adaptive processing algorithm on a parallel embedded system

This thesis involves the investigation of parallelization and performance improvement for a class of radar signal processing techniques known as space-time adaptive processing (STAP). The assumed platform, which consists of multiple DSPs, is the commercially available Mercury RACE System. The main contribution of the thesis is the design and implementation of a network simulator for the RACE system. This simulator allows for the performance of various parallel STAP algorithm implementations to be predicted for existing or future RACE system configurations.

A major challenge of implementing parallel STAP algorithms on multiprocessor systems is determining the best method for distributing the 3-D data cube across CEs of the multiprocessor system (i.e., the mapping strategy) and the scheduling of communication within each phase of computation. It is important to understand how mapping and scheduling strategies affect overall performance. The network simulator developed in this thesis is used to evaluate the performance of various mapping and scheduling strategies.