Service Localization and User Satisfaction Enabled Multi-Tier Computing IoT System

The rapid expansion of the Internet of Things (IoT) poses challenges to certain applications, such as Digital Twin (DT). While data from user devices can be filtered by human intelligence, this is not feasible for IoT devices. Owing to the voluminous data generated by IoT devices that require transmission and computing, traditional cloud computing architectures may no longer guarantee the Quality of Experience (QoE), even causing network congestion. To address this issue, we propose a novel Cloud-Network-Edge-Terminal (CNET) model, which includes an intelligent edge layer for filtering IoT data. The computing paradigm shift indicates that the network will provide services at the edge rather than in the cloud, which is so-called service localization. To demonstrate the benefits of service localization, we use integrated user requirement descriptions to measure QoE, specifically the concepts of Service Requirement Zone (SRZ) and User Satisfaction Ratio (USR). Additionally, we conduct extensive numerical simulations to evaluate the model's performance under varying Degrees of Localization (DoL). Our results show that service localization can significantly improve USR even in changing network conditions.