• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International conference on computers and their applications >Modeling Power Consumption of Mobile Devices Using a Hybrid Learning Approach with Crowdsourced Datasets
【24h】

Modeling Power Consumption of Mobile Devices Using a Hybrid Learning Approach with Crowdsourced Datasets

机译：使用具有众包数据集的混合学习方法对移动设备的功耗进行建模

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Power consumption has been one of the major issues in the design of mobile devices. In the past, the concerns of device users mainly focus on which factors influence power consumption. For design of the products, context awareness of product usage is regarded as one of the most crucial process for understanding users' requirements. The view is taken, therefore, in this work we developed a data-driven approach to characterize the power consumption of mobile devices using collected data by crowdsourcing. First, we used a multiple regression method to analyze correlation among the variables, and filtered the specified datasets as training data for experimenting with support vector machines classifiers. Further, we used a support vector regression technique to verify the prediction results of multiple regression analysis. The experimental results demonstrated that our analytical approach can be used to characterize power consumption of the devices, in order to determine an effective energy saving strategy. For example, when the number of screen wake locks increased by 14 times in one day, battery discharge time shortened by 1.49 hours. Users can estimate how these changes are expected to influence the battery consumption.
机译：功耗一直是移动设备设计中的主要问题之一。过去，设备用户的关注主要集中在哪些因素会影响功耗。对于产品的设计，产品使用的上下文意识被认为是了解用户需求的最关键过程之一。有鉴于此，因此，在这项工作中，我们开发了一种数据驱动的方法，以通过众包收集的数据来表征移动设备的功耗。首先，我们使用多元回归方法分析变量之间的相关性，并过滤指定的数据集作为训练数据，以使用支持向量机分类器进行实验。此外，我们使用了支持向量回归技术来验证多元回归分析的预测结果。实验结果表明，我们的分析方法可用于表征设备的功耗，从而确定有效的节能策略。例如，当一天的屏幕唤醒锁定次数增加14次时，电池放电时间缩短了1.49小时。用户可以估计这些变化将如何影响电池消耗。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号