机器学习在无线体域网多网共存状态预测中的应用研究
首发时间:2017-12-01
摘要:当一个无线体域网的通信范围内有其它无线体域网存在时可能会出现共存问题,这会导致整个网络通信性能下降,甚至威胁人的生命,因此需要一种能够可靠预测多网共存状态的方法来确保共存问题可以被及时检测和处理。本文研究了机器学习在无线体域网多网共存状态预测中的应用,并采用了决策树和朴素贝叶斯分类器两种监督学习模型,将共存状态分为四类:空态、静态、半动态和动态。本文通过CC2530实验平台比较了当传感器节点分别位于头部、左臂和腰部三个部位时两种模型的表现性能,并最终获得了准确率最高的共存状态预测模型。
For information in English, please click here
Research on the Application of Machine Learning in Coexistence State Prediction in Multiple WBANs
Abstract:The coexistence problem may occur when a wireless body area network (WBAN) meets with other WBANs. It can degrade the communication performance of these WBANs, and even threaten people\'s life. Therefore, a reliable method for coexistence state prediction is required to ensure this problem could be detected and handled in time. The application of machine learning in coexistence state prediction in multiple WBANs is researched in this paper, in which decision tree (DT) and naive Bayes classifier (NBC) as two supervised learning models are adopted to divide the coexistence states into four classes: None, Static, Semi-dynamic and Dynamic.The performance of two kinds of models are compared by CC2530 experimental platform when the sensor node is deployed on head, left arm and waist respectively, and finally the model with the highest accuracy for coexistence state prediction is acquired.
Keywords: Wireless Body Area Network Coexistence Problem Prediction Method Machine Learneing
基金:
引用
No.****
同行评议
共计0人参与
勘误表
机器学习在无线体域网多网共存状态预测中的应用研究
评论
全部评论0/1000