基于贝叶斯网络的WSNs链路质量评估机制研究

Research on link quality estimation mechanism for WSNs based on Bayesian networks

在无线传感器网络(WSNs)应用中,链路质量的有效评估是保障数据可靠传输和上层网络协议性能的基础性问题。针对现有无线链路质量评估研究中,链路质量等级划分仍无统一标准和模型缺乏环境自适应性问题,提出了一种基于贝叶斯网络的链路质量评估机制。从链路质量多属性角度出发,采用贴近度分析法对链路质量等级进行划分,利用贝叶斯网络对链路质量进行不确定性推理与评估建模。通过多应用场景的训练与测试,最后实际测试表明:基于贝叶斯网络的评估模型在不同应用环境具有较高准确率和泛化能力,体现了良好的自适应性。

In application of wireless sensor networks（ WSNs）,effective estimation for link quality is a basic issue in guarantying data reliable transmission and upper layer network protocol performance. Aiming at problem that in existing wireless link quality estimation study,link quality hierarchy division has no uniform standard and models are lack of environmental adaptability,a link quality evaluation mechanism based on Bayesian networks is proposed. As link quality has multiple attributes,close degree analysis method is used for link quality grade division,and use Bayesian networks for link quality uncertainty reasoning and estimation modeling. Through training and testing in multi-application scenarios,practical testing results demonstrate that the link quality estimation model based on Bayesian network has high accuracy and generalization ability in different application environments,reflect good adaptability.