Optimization of energy consumed by real-time applications embedded into DSP systems

This paper deals with consumption of real-time applications running on embedded systems. It is assumed that a significant part of the power dissipated by systems is due to input/output made between all the chips and especially between commonly used digital signal processors and their tiny internal memories and their compared to huge external memories, it is worth settling a strategy to reduce their transfers by keeping inside the processor the more interesting data that will yield to minimize the power. The authors' method is based on the modeling of C source application into two graphs, graph of data dependence and graph of expression dependence that will give information about the behavioral of data, their precise moments of utilization and the best scheduling of instructions so that data reuse is maximal. First results show that by their method it is possible to reduce by 70% the energy of some applications.