论文部分内容阅读
目前智能手机大多节能技术是以约束用户行为为前提的。为了兼顾节能与用户体验，提出一种面向用户行为的智能手机能耗优化方法。即集合用户使用习惯，采集用户在屏幕关闭时使用应用程序的网络活动数据，通过挖掘、分类和归纳等技术手段生成决策树，预测应用程序对用户的重要性，建立应用程序网络请求最优控制的数学模型。使用多背包算法求解该最优控制问题，基此限制屏幕关闭后的应用程序的网络请求。实验结果表明，该方法的节能百分比维持在38.4%左右，相比于延迟容忍、批处理等节能技术提高了19.9%。所提方法能够在不影响用户体验的基础上改善智能手机的能耗问题。
At present, most energy-saving smart phones are based on the premise of constraining user behavior. In order to balance energy saving and user experience, a smart phone energy consumption optimization method based on user behavior is proposed. That is, the user habits are collected, and the user’s network activity data of the application is collected when the screen is closed. The decision tree is generated through mining, classification and induction and other technical means to predict the importance of the application to the user and to establish the optimal control of the application network request Mathematical model. Use the multiple backpack algorithm to solve the optimal control problem, based on the network request of the application after the screen is closed. Experimental results show that the energy-saving percentage of this method is maintained at about 38.4%, and energy-saving technologies such as batch processing are increased by 19.9% compared with delay tolerance. The proposed method can improve the smart phone’s energy consumption without affecting the user experience.