VLIW instruction scheduling for DSP processors based on rough set theory
Shu Long Xiao, Edmund M-K Lai, A. P. Vinod · 2006
Power-balanced instruction scheduling for Very Long Instruction Word (VLIW) processors is an optimization problem which requires a good instruction-level power model for the target processor. Conventionally, these power models are deterministic. In this paper, we propose a rough program problem formulation to handle the imprecision involved in the VLIW power models. A problem-specific genetic algorithm is proposed to solve it. Our experiments revealed that the actual occurrence of optimal schedules obtained by integer programming often have a large deviation of the objective function values, due to the ignorance of imprecision accumulation. The results justified our rough programming approach to find an optimal schedule which makes sure that the effects of these imprecision could be minimized. © 2005 IEEE.