物联网环境下基于上下文感知的智能交互模型

Intelligent Interaction Model of Internet of Things Based on Context Awareness

为了提高物联网环境下平台交互的实时性,在Storm框架的基础上,建立了物联网环境下基于上下文感知的智能交互模型.该模型将异构、非结构化数据抽像为上下文,构建了上下文适配、上下文存储与上下文感知3个核心模块,并采用基于Storm框架的规则匹配算法解决了上下文与规则的持续模糊匹配问题.实例验证结果表明:在样本量接近1 000的情况下,IITCA能够达到与人工操作相同的有效性;具有与基于Hadoop框架构建模型相当的数据吞吐能力;设置Spout组件为3、Bolt为6时,模型数据处理的平均时延稳定于50 ms,满足数据处理的实时性要求.

In order to improve the real-time performance of system in Internet of Things （ lOTs）, an intelligent interaction model of lOTs based on context awareness （IITCA） and Storm framework was proposed. In this model, heterogeneous and unstructured data processed in lOTs are abstracted as context, and three core modules, namely context adaption, context storage and context perception are constructed. In addition, a continual rule matching algorithm based on Storm is put forward to solve the problem of fuzzy matching between contexts and rules. Experiment results show that with a sample size close to 1 000, IITCA is of the same effectiveness with manual operation ; at the same time, IITCA has an equivalent data throughput capability to the model based on Hadoop. Besides, when Spout is set to 3 and Bolt is set to 6, the delay of data processing is stable to 50 ms, which can satisfy the real-time performance requirement.