Space-Time Adaptive Processing (STAP) refers to adaptive radar processing algorithms that takes the signals from both multiple sensors and multiple pulses to cancel interferences and detect a target. Fully-adaptive STAP is known to be optimal, but the required number of operations is overwhelming. Considering the real-time requirements in radar processing, this method is impractical. Hence, many different heuristic approaches are sought to approximate the optimal method with fewer number of operations. The real-time requirement makes parallel processing and its optimization highly desirable in this area.

Previous researches have shown that these heuristic methods can be described in terms of basic tasks performed on different sub-set of data in different orders. In this work, we introduce a framework for prototyping various STAP methods on a parallel system: describing STAP algorithms, building execution time models for basic building blocks, and using these timing models to automatically optimize performance of the algorithms. We also present the performance results from a COTS PC cluster.