无线传感网中基于信息增益最大化的泛在数据收集算法

发布时间：2019-02-15 20:47

【摘要】：基于移动设备的泛在数据收集是目前的研究热点。然而现有的数据收集方案关注的重点是提高数据收集的可靠性和能效,很少考虑传感器数据携带的信息价值。此外,移动用户可能没有足够多的时间收集周围传感器的所有数据。因此,有必要收集对于用户来说价值最大的数据。本文提出一种基于信息增益最大化的泛在数据收集算法。该算法首先利用移动设备可用容量估计值和传感器需求来自动确定数据收集树构建的最大层数,然后通过数据收集树的迁移来支持移动用户在自身移动性不受限制的条件下进行分布式数据收集。此外,移动用户可以根据自身速度,动态估计可以用于数据收集的设备容量。仿真实验结果表明,与当前最新的其他算法相比,本文算法可将信息价值提高50%、能耗降低50%。
[Abstract]:Ubiquitous data collection based on mobile devices is a hot research topic at present. However, the existing data collection programs focus on improving the reliability and energy efficiency of data collection, with little regard to the value of information carried by sensor data. In addition, mobile users may not have enough time to collect all the data around the sensors. Therefore, it is necessary to collect the most valuable data for the user. This paper presents a ubiquitous data collection algorithm based on maximization of information gain. Firstly, the maximum number of layers constructed by the data collection tree is automatically determined by using the estimation of the available capacity of the mobile device and the sensor requirements. Then the migration of data collection tree is used to support mobile users to collect distributed data under the condition of unrestricted mobility. In addition, mobile users can dynamically estimate the capacity of devices that can be used for data collection based on their own speed. The simulation results show that the proposed algorithm can increase the information value by 50% and reduce the energy consumption by 50% compared with other algorithms.
【作者单位】：黄淮学院信息工程学院;
【基金】：河南省科技厅发展计划项目(142102110088)
【分类号】：TN929.5;TP212.9

【相似文献】