ASP-based optimized mapping in a simulink-to-MPSoC design flow

This paper presents an approach to the automated identification of optimal mapping choices in a Simulink-to-MPSoC design flow. The mapping process relies on an appropriately chosen model of computation, capturing the high-level structure of the Simulink application as well as enabling formal checking of several relevant properties, such as boundedness, liveness, as well as throughput and latency formulas. The optimization approach exploits an emerging logic programming language, Answer Set Programming (ASP), for design space exploration. The proposed ASP-based solution can be used in the context of Simulink-to-MPSoC translation as it provides a technique to automate the optimization of design choices aimed at resource utilization and execution time. A case-study and the related experimental results, presented at the end of the paper, demonstrate the effectiveness of the proposed approach.