Platform-Based Design has recently emerged as an important design style for embedded System-on-Chip (SoC) architectures, that must be optimally tuned to find the best trade-offs in terms of the selected figures of merit (e.g. energy and delay) for the given class of applications. This tuning process is also called the Design Space Exploration (DSE), aiming at optimally configure the parameterized SoC platform in terms of system-level requirements depending on the given application. In general, the optimization problem involves the minimization (maximization) of multiple objectives making the definition of optimality not unique. The solution of multi-objective or constrained optimization problems consists of finding the points of the Pareto curve.
The main objective of the research consists of defining exploration and optimization techniques for Multi-Processor System-on-Chip (MP-SoC) architectures for embedded multimedia applications.