基于MA(q)模型的动态电源管理预测策略

摘要

This paper presented a novel PIF DPM policy based on MA(q) model,which was made up of MA(q) model,CI model and PF model. The experimental results show that the PIF policy is stable and the average power consumption of system can be further reduced as compared to other predictive policies.%基于MA(q)模型提出了一种新颖的PW( predictive-issue-feedback)预测策略,该策略由MA(q)预测模块、CI控制模块、PF反馈模块三部分组成.实验结果表明,PIF策略具有稳定性,而且与其他预测性策略相比,该策略能够进一步降低系统的平均功耗.