A 5G Deployment Evaluation and Performance Optimization Testbed (DEPOT)

5G Deployment Evaluation and Performance Optimization Testbed (DEPOT) is designed as a full-stack 5G testbed which can help users assess network and user performance as well as optimize network architecture. The extensive performance evaluation capabilities of the testbed will provide detailed analysis of achieved throughput, latency and other network performance metrics within the Core Network components and to the user. Additionally, the testbed will have awareness of resource consumption of the Core Network to provide required Quality of Service (QoS) to end users. This will enable DEPOT to propose optimal network architecture and topology with QoS requirements as a parameter. To achieve all this, DEPOT will utilize the existing 5G testbed developed by Kryptowire as part of the Open, Programmable and Secure 5G (OPS 5G) program, as well as MantisNet’s network performance evaluation framework – Tawon CVF.   DEPOT will fulfill the following objectives: Develop a full-stack 5G network deployment with remote access that can accommodate a wide variety of use cases. The deployment will utilize open-source implementations of microservices. Individual microservices will be deployed as Docker containers in a Kubernetes cluster. This will allow straightforward replacement of any microservice for testing purposes. Additionally, the described deployment design will accelerate adoption of cutting-edge technology to support the latest standards. Furthermore, the open-source plug-and-play deployment will facilitate future functionality enhancements of the system. The network can be deployed on any commercial off-the-shelf (COTS) hardware with enough resources for the required number of users and given use-cases. Accessible execution of various types of simulated or emulated users with real use-case traffic models. The traffic models will be created by collecting real data from devices running applications such as video streaming, video conferencing, IoT for smart home, VR, online gaming, etc. High fidelity user emulation will facilitate experiments with any real-world use-case scenario. This will enable realistic evaluation of network microservices for any type of QoS that the tester may require. Extensive performance evaluation of 5G components using MantisNet’s Tawon CVF. The framework is vendor agnostic and can provide deep visibility into the topology, data flow, protocol metadata, encrypted session data and packets within the 5G network. Aggregating performance evaluation and analysis tools will be added to simplify network and user performance assessments. Public repositories with network deployment and performance evaluation framework tutorials. Publicly available instructions will allow testers to deploy the network and measurement tools on their own hardware or cloud computing services, and provide guidance on how to run various types of experiments.