Since Ian Chen mentioned that Free5GC UPF doesn't have Deep Packet Inspection (DPI) feature in CNDI course, it caught my interest to implement a DPI feature. After doing some research, most DPI feature is handle with hardware support such as P4 programmable switch. We just happend to have a P4 switch in our Lab, so it's a perfect timing to start this research project. Construct hardware-support DPI solution is easy and straight forward. It's easy for me to construct this since I have some experience on P4 Switch. To make it more challenge, We trying to find a Hardware-Independent solution that can achieve efficient as same as hardware-support solution (Like, P4 Switch). It's interesting that discovery a hardware-independent solution and it increase the possibility that we can integrate into Free5GC UPF.
For the information about our PoC, please refer this blog.
This project implements a high-performance parallel NGAP message processing architecture in the free5GC AMF, replacing the sequential SCTP handling model to resolve Head-of-Line blocking. It utilizes a lock-free dispatching mechanism with a UE-ID based hashing scheduler to distribute signaling loads across worker pools. This design guarantees per-UE message ordering while maximizing multi-core utilization, resulting in a 3.66x speedup in registration time and significantly higher success rates under simulated signaling storms.
This project is an MCP (Model Context Protocol) server implemented in Go that provides AI assistants (like GitHub Copilot) with tools to manage free5GC 5G core network subscribers and configurations.
- Core Network Control: Start, stop, and monitor all free5GC network functions (NRF, AMF, SMF, UPF, etc.)
- Subscriber Management: Full CRUD operations for 5G subscribers
- Tenant User Management: Query users within tenants
- JWT Authentication: Automatic authentication with free5GC webconsole
- VS Code Integration: Works seamlessly with GitHub Copilot in VS Code
Group 4: free5GC-trace
This project implements end-to-end distributed tracing for the UE Registration procedure in a free5GC-based 5G Core, using OpenTelemetry and Grafana Tempo. It correlates SBI calls across six key NFs (AMF, AUSF, UDM, UDR, NSSF, NRF) into a single trace so that each UE registration can be visualized and debugged as one coherent flow.
Group 5: UPF Acceleration App using DOCA
This project implements an accelerated User Plane Function (UPF) datapath using NVIDIA DOCA on a DPU platform (BlueField). The goal is to offload packet processing, GTP encapsulation/decapsulation, and routing functions into hardware to greatly improve throughput while reducing CPU load.
This repository demonstrates how DOCA can accelerate a UPF in a complete test environment, using free5GC as the 5G Core control plane, UERANSIM to simulate UE and gNB traffic, and a DOCA Flow–based UPF application running on the DPU.
Group 6: Automation of free5GC ci-test
A 5G UPF observability platform powered by eBPF. Features real-time traffic monitoring, granular packet drop detection, and PFCP session correlation for free5GC without source code modification.