Analysis of Power Minimization for Modified Force Directed Scheduling and Integer Linear Programming

This paper presents a novel approach to minimize peak power consumption in data flow graphs using a modified force-directed scheduling algorithm. The proposed method is compared with Integer Linear Programming (ILP)-based scheduling to optimize high-level synthesis for low-power operations. The primary objective is to reduce peak power while scheduling data flow graphs. This work also provides an overview of existing scheduling algorithms, including: - As Soon As Possible (ASAP) scheduling, - As Late As Possible (ALAP) scheduling, Force-Directed Scheduling. The results show that adjusting the clock cycle and applying the modified force-directed scheduling algorithm successfully lowered peak power consumption, validating the proposed method's potential for power-efficient high-level synthesis.