Low Power Instructions Scheduling Based on Ant Colony Optimization

In this paper, we focus on the problem of instructions scheduling based on new meta-heuristic algorithm named Ant Colony Optimization (ACO). During the execution of the program, in fact, the flow goes from one instruction to the other. This procedure leads to binary transition just out of the differences between the binary code of two consecutive instructions. Our method pursues a related optimization strategy to reduce the total number of binary transition during the execution. It´s very hard to find the best scheduling solution because of being a NP-Hard. ACO is used to research for an approximate but effective solution, to achieve the goal of combinational optimization. The final goal is to minimize the dissipation of power at lower level.