Design of oneM2M-Based Fog Computing Architecture

Abstract

Internet of Things (IoT) technology could provide solutions for many real-world applications, and Big Data analysis of IoT continues the key to realizing smart cities. However, IoT cloud computing cannot yet fully meet the requirements of the low latency and fast response for applications, such as public safety and emergency response services. These deficiencies could be addressed by Fog Computing, which uses front-end devices or serially connected terminal devices to conduct data storage, calculations, or related decentralized control management through decentralized collaborative architecture. According to the original oneM2M IoT architecture, it takes 251.2–537.2 ms of transmission time, for high-resolution image data at front-end edge nodes, but it fails to meet the requirements of low-latency and fast-response applications. To address this, we propose a novel oneM2M-based Fog Computing architecture under which adjacent Fog nodes can coordinate and cooperate with each other to deal with operational requirements. Operations are no longer constrained by a node’s ability to perform tasks or by the requirement of forwarding tasks to the cloud for execution. The proposed computing architecture could transmit high-resolution image data between Fog nodes and then could reduce the transmission time to 16.7–46.2 ms, which is a 94.1% reduction compared with the original oneM2M architecture. In summary, this method could provide low latency and fast response in Fog Computing by maximizing the reduction of the overall end-to-end delay.