3GPP-based Testbed for Edge Computing; Architecture, Implementation and Application Deployment
Syed Majed Ashraf, Ashish Singh Patel, Arzad Alam Kherani, Brejesh Lall · 2023
5G and beyond together with edge computing are transforming the next generation communication by enabling real-time interactions and data access between the applications. 3rd Generation Partnership Project (3GPP) has developed an edge application enabling architecture (EDGEAPP) which provides an edge computing framework that enables cloud computing capabilities close to the UE. EDGEAPP architecture provides a service platform for implementing numerous applications and services tailored to consumers and industry at the network edge, which opens new markets and enormous business opportunities for telecom operators, service providers, and other stakeholders. This research work demonstrates the implementation and working of 3GPP architecture for enabling EDGE applications. In addition, the edge computing architecture of various other standardization organizations is also evaluated with reference to the 3GPP architecture. Furthermore, the standardization aspects along with the application deployment scenarios are explored and presented. The deployment is performed by demonstrating application instantiation, configuration, management, and communication within various EDGEAPP entities. Furthermore, the Edge is integrated with a 5G core using Open Air Interface (OAI) is also performed. The implementation is validated by demonstrating application instantiation and execution on the user equipment (UE) as a use-case. The deployment involves interaction among various entities, 5G Core, and UE for service provisioning, discovery, registration, de-registration, capability exposure, and secure communication. The EDGEAPP architecture entities are deployed as docker containers with complete data and control paths established between them. Thus, the work provides proof of concept that simulates a real-world deployment of EDGEAPP architecture in a 5G environment to mimic real-time deployments.