5G Core Foundation: Deploying Open5GS WebUI & MongoDB with Monitoring Stack

“The telephone will unite mankind.” — Alexander Graham Bell, Inventor of the Telephone

Let’s break down the 5G magic into layers:

1. Radio Layer (RAN — Radio Access Network)

• This is the “air interface.”
• Uses millimeter waves (mmWave), sub-6 GHz, and low bands — like traffic lanes:
• Low band: Long range, slower (like highways).
• Mid band: Balanced coverage and speed.
• High band (mmWave): Superfast, short range (like high-speed bullet lanes).
• Uses Massive MIMO (Multiple Input Multiple Output) — think of it as having hundreds of smart antennas focusing beams (beamforming) directly at devices for better efficiency.

2. Transport Layer (Backhaul + Fronthaul)

• Moves data between RAN and the core.
• Fiber or microwave links act like the “arteries” of the network.

3. Core Network (5G Core)

• This is where 5G gets smart.
• Built on Service-Based Architecture (SBA) — everything is modular microservices.
• Supports Network Slicing: You can create multiple “virtual networks” on one physical infrastructure.
• Example: One slice for gamers (ultra-fast), another for IoT sensors (low power), and one for emergency services (ultra-reliable).

4. Edge Computing

• Moves compute power closer to users.
• Instead of sending every request to a distant data center, 5G processes it near the tower — reducing latency and enabling instant responses.

🌐 What is Open5GS?

Open5GS is an open-source implementation of a 4G/5G Core Network — a complete software-based system that lets you build and test your own mobile network (like a mini version of what carriers use, but for research and learning).

In short:

🧠 Open5GS = “5G Core you can install on your laptop or lab server.”

It’s developed in C and designed to interoperate with open-source RAN projects like srsRAN, OAI, or UERANSIM (for simulating user equipment and base stations).

Lets deploy this Open5GS in an EC2 and Visualize the open5gs webui and Mongo

git clone https://github.com/open5gs/open5gs
cd open5gs
meson build - prefix=`pwd`/install
ninja -C build
ninja -C build install
cd ~/open5gs/docker
sudo docker compose up -d

You should see open5gs-webui (in 9999 port) and open5gs-mongodb up and running!. Lets login open5gs-mongodb and insert users if needed

sudo docker exec -it open5gs-mongodb mongo

use open5gs

db.users.insertOne({
    "username": "admin",
    "password": "open5gs",
    "role": "admin"
})

DevOps: Making Telecom Agile

DevOps practices bring automation, CI/CD, and observability into telecom — enabling continuous delivery of network updates safely and fast.

Key Roles of DevOps in 5G:

• CI/CD Pipelines: Automate code integration, testing, and deployment of network functions (e.g., AMF or SMF upgrades).
• Infrastructure as Code (IaC): Tools like Terraform or Ansible spin up network environments dynamically.
• Automated Scaling: Kubernetes autoscaling adapts network capacity in real time to user demand.
• Observability: Monitoring and logging pipelines ensure SLA compliance and fault recovery.

How ELK and Grafana Fit Into the 5G Cloud Ecosystem?

In traditional telecom (3G/4G), network functions (like MME, SGW, PGW) were monolithic and hardware-tied.

1. ELK Stack (Elasticsearch, Logstash, Kibana)

In telecom, every component — from base stations to the 5G Core — generates tons of logs and metrics.

ELK acts as the real-time analytics and troubleshooting powerhouse.

Here we have build some dashboards using scripts below

https://github.com/sivakumarpalanikumar/Grafana-ELK-Stack-Security-Monitoring/tree/main/ELK

2. Grafana: The Visualization Layer

Grafana is often used alongside Prometheus or Loki to visualize real-time network metrics:

• Throughput, latency, session counts, packet loss, CPU/memory usage.
• Used for NOC (Network Operations Center) dashboards.

https://github.com/sivakumarpalanikumar/Grafana-ELK-Stack-Security-Monitoring/tree/main/Grafana

An Open5GS Proof of Concept (POC) can evolve into a next-generation intelligent 5G testbed by integrating UERANSIM which simulates UE and gNB entirely in software and perfect for lab testing — no RF hardware needed.

It can be integrated with srsRAN, and AI/ML analytics within a cloud-native DevOps environment. Open5GS provides the 5G Core, while UERANSIM simulates virtual UE and gNB connections for scalable lab testing, and srsRAN adds real RF-based RAN functionality for near-real-world experimentation.

Conclusion

• Ericsson (on their “On-Demand” 5G core / SaaS platform):
• “Ericsson On-Demand … is more than a product, it’s an enabler of ambition. Today’s CTOs must move fast, scale smart, and lead their organizations into new commercial territory.” newswire.telecomramblings.com+2MarketScreener+2
• Nokia (on AI / automation in networks):
• “We deliver automation and intelligence — across the network, services, and directly to the customer with our AIOps solutions and services.”

By containerizing these components on Kubernetes, automating deployments through CI/CD, and coupling them with observability tools like ELK and Grafana that can view UPF, SMF logs, AMF logs and real-time network data can feed AI/ML models that predict congestion, detect anomalies, and optimize QoS dynamically. This transforms a basic 5G core setup into a self-monitoring, adaptive, and intelligent 5G network lab, ideal for innovation, automation, and research.