From EdgeRIC to Intelligent RAN

EdgeRIC is a platform for Real-Time AI-in-the-Loop decision making and control in cellular networks.

Our goal is to build an intelligent RAN stack capable of real-time decision making by integrating:

• Next-generation Fine-Grained Telemetry for deep network observability

• Reasoning over Network Dynamics for adaptive control

• Experience-Aware Learning for continuous improvement of network performance

• Mar 2026 - Stay tuned for upcoming events!

• Feb 2026 - Check out our latest updates!

Our Projects

Cloud Gaming via 5G

mmsRAN: Multi-Modal Sensing

SCOUT: Packet Centric Telemetry

QoE-aware Networking

Tiny-twin

Meet Our μApps Family

Our Projects

• Cloud Gaming on 5G
• MultiModal Sensing
• Telemetry: Packet Centric
• QoE-aware Networking
• Tiny-twin
• Other μApps

Demos

Demos

• Demo

BeamArmor

Anti-jamming: Controlling MIMO weights in realtime to steer the beam null toward the jammer

EdgeRIC: AI Scheduling

RL-based scheduling policy trained to maximize overall system throughput

SPARC: Multi-site Management

Interference-aware resource distribution across sites with Near-RT RIC

Open Source Repositories

Open Source Repositories

• Open Source Repositories

srsRAN-4G + EdgeRIC

Release Date: TBD

Description: EdgeRIC integrated with srsRAN 4G

View on GitHub

srsRAN-5G (2024) + EdgeRIC

Release Date: 2024

Description: EdgeRIC on srsRAN 5G 2024 release

View on GitHub

srsRAN-5G (25.10) + EdgeRIC

Release Date: 25.10

Description: EdgeRIC on srsRAN 5G 25.10 release

View on GitHub

OAI + EdgeRIC

Release Date: Oct 01, 2025

Description: EdgeRIC integrated with OpenAirInterface

View on GitHub

Events

EdgeRIC Events

• OAIC-EdgeRIC Workshop ‘24
  • EdgeRIC tutorial
• srsRAN Workshop ‘24
  • Our presence at the workshop

2024

OAIC Workshop 2024

Leading the Deployment: OAI + EdgeRIC Integration Hands-on Sessions with Ushasi Ghosh.

View details

2024

srsRAN Workshop 2024

Beyond the EdgeRIC: Ushasi Ghosh's Live Guide in EdgeRIC Integration, Experiments, and Testbed.

View details

2024

NSDI 2024

Woo Hyun Ko & Ushasi Ghosh: EdgeRIC Main Paper Presentation.

View details

2022

OAI North American Workshop 2022

Charting the NextG Frontier: Prof. Dinesh Bharadia on the Evolution of 5G, RAN, and Digital Twins.

View details

Team

Team

• Team

Dinesh Bharadia

Professor, UCSD

dineshb@ucsd.edu

Srinivas Shakkottai

Professor, TAMU

sshakkot@tamu.edu

Ish Jain

Professor, RPI

jaini2@rpi.edu

Shunsuke Saruwatari

Professor, Osaka University

ssaruwatari@ucsd.edu

Woo Hyun Ko

Senior Research Engineer, TAMU

whko@tamu.edu

Ushasi Ghosh

PhD Student, UCSD

ughosh@ucsd.edu

Sushila Seshasayee

PhD Student, UCSD

sseshasa@ucsd.edu

Ali Mamaghani

PhD Student, UCSD

amamaghani@ucsd.edu

Archana Bura

Postdoc, UCSD

archanabura@tamu.edu

Qingyuan Zheng

MS Student, UCSD

qiz066@ucsd.edu

Frederik Zumegen

Visiting Researcher, UCSD

fzumegen@ethz.ch

Sesha Sai Rakesh

MS Student, UCSD

sjonnavithula@ucsd.edu

WCSNG

Lab @ UCSD

github.com/ucsdwcsng

Publications & Tutorials

Publications & Tutorials

• EdgeRIC main paper
  • MAC scheduling
  • Install srsRAN supporting EdgeRIC messages and control
  • How to run the network
  • Running EdgeRIC for downlink scheduling control
• BeamArmor
  • Sections
  • Installing the repo
  • Preparing BeamArmor and MIMO-RIC
  • Files under concern
  • Parameters of MIMO-RIC
  • BeamArmor Demo: Automatic BeamArmor turn ON
  • Setup
  • BeamArmor Demo: Manual BeamArmor turn ON/OFF
  • Setup
  • Phases of Demo
• Tiny Twin
• Windex
• SPARC
  • Install srsRAN supporting E2 and EdgeRIC messages and control
  • Network Setup
  • Hierarchical RIC deployment

EdgeRIC

Empowering Real-time Intelligent Optimization and Control in NextG Cellular Networks.

Paper@NSDI'24 Demo@SIGCOMM'23 Demo Project Website Code on GitHub Tutorial

Beamarmor

Seamless anti-jamming in 5g cellular networks with mimo null-steering.

Paper@HotMobile'24 Paper@MobiCom'24 Demo Project Website Code on GitHub Tutorial

Tiny-Twin

A CPU-Native Full-stack Digital Twin for NextG Cellular Networks.

Paper@DySPAN'26 Poster@HotMobile'24 Demo Tutorial

Windex

Realtime Neural Whittle Indexing for Scalable Service Guarantees in NextG Cellular Networks.

Main Paper Demo@MobiCom'24 Project Website Code on GitHub Tutorial

SPARC

Spatio-Temporal Adaptive Resource Control for Multi-site Spectrum Management in NextG Cellular Networks.

Paper@CoNEXT'25 Demo Project Website Code on GitHub Tutorial

EdgeRIC Core Architecture

EdgeRIC Architecture

• RT-E2 interface (Real time E2 interface)
• RT-E2 Report Message
• RT-E2 Policy Message
• REDIS database
• μApps - EdgeRIC microservices
  • how to write μApps?

RT-E2 interface (Real time E2 interface)

Messaging framework between the RAN stack and EdgeRIC, built on ZMQ and protobuf, with TTI-level synchronization.

RT-E2 Report Message

Per-UE KPI report structure (cqi, buffers, TBS, rates) sent every TTI from the RAN to EdgeRIC.

RT-E2 Policy Message

Control-action messages from μApps back to the RAN, including scheduling weights and blanking decisions.

REDIS database

How Redis is used for model storage, μApp lifecycle management, and dynamic configuration.

μApps – EdgeRIC microservices

How μApps subscribe to metrics, compute policies, and send control via the EdgeRIC messenger.

Datasets

Datasets

• EdgeRIC Main Dataset
• Scout Dataset

EdgeRIC Dataset

Channel traces (CQI) from over-the-air mobility experiments, used for Tiny-twin emulation and RL training.

Scout Dataset

Log dataset for SCOUT: packet-centric telemetry logs, per-packet latency, queue dynamics, and scheduling impact analysis.

5G Testbed

5G Testbed

• 5G Testbed

Funding

This work was funded primarily by NSF Grants CNS 2312978, CNS 2312979 and in part by CNS 1955696, ECCS 2030245, ARO grant W911NF- 19-1-0367.