智能环境下基于边缘设备规则推理的数据预部署研究

Study on Data Pre-Deployment Based on Inference and Computing of Edge Devices in Smart Environment

在现有的规则推理机制下,大量的传感器数据导致的过大规则匹配期间的实时特征计算量降低了推理实时性,同时边缘设备受限的内存资源难以应对如此庞大的数据量.为此,本文设计了数据预部署方案(Data Pre-De⁃ployment Scheme,DPDS).利用规则解析与预处理模块解析规则集得到的规则网络和轻量级特征表(Light-weight Char⁃acteristic Table,LCT),该方案无需进行实时特征计算,使推理效率和实时性得到显著提高,并大大降低了规则匹配期间的内存占用量.实验表明,即使在规则、数据规模很大的情况下,DPDS仍然具有较高的时间效率和空间效率.

Under the existing rule inference mechanism,the amount of real-time feature calculation during rule matching caused by a large amount of sensor data reduces the inference real-time performance.At the same time,the limit⁃ed memory resources of edge devices cannot cope with such a huge amount of data.For this reason,this thesis designs the Data Pre-Deployment Scheme(DPDS).By utilizing the rule network and Light-weight Characteristic Table(LCT)obtained by the rule analysis and preprocessing module,this scheme enables the rule network to directly reference the characteristic values in the LCT during inference without real-time feature calculations,which significantly improves efficiency and realtime and greatly reduces the memory usage of the inference process.The experimental results show that even in the case of a large amount of data and rules,DPDS still has high time efficiency and space efficiency.