Parallel algorithm mapping to memory multidimensional signals

Abstract

The signal processing algorithms are typically described in a high-level programming language. In data-dominated applications, particularly in the multimedia and telecommunication domains, the code of these behavioral specifications is organized in sequences of loop nests; the main data structures are multidimensional arrays. This paper proposes a memory management algorithm for mapping multidimensional signals (arrays) to the physical memories. The proposed technique is well-suited to a dynamic multithreading implementation, which makes it computationally efficient. Another advantage is that it can be applied to multi-layer memory hierarchies, which makes it particularly useful in embedded systems design.