• EdgeRIC main paper
  • MAC scheduling
    • Training Reinforcement Learning Policy
  • Install srsRAN supporting EdgeRIC messages and control
    • Install zmq
    • Dependencies and cloning the repository
    • Build the repository
  • How to run the network
    • Setup the core and srsenb
    • Run the UEs
      • Running UEs with CQI trace files
    • Stream Traffic:
      • Running Downlink iperf traffic
      • Running Uplink iperf traffic
      • Running various other kinds of Downlink Traffic profiles
  • Running EdgeRIC for downlink scheduling control
    • EdgeRIC messenger
    • μApps supported in this codebase
    • Running muApp1 - downlink scheduler
      • Setting the scheduler algorithm manually
    • Running muApp2 - Training an RL policy for scheduling
      • Usage
      • muApp2_train_RL_DL_scheduling.py
      • Repo Structure
      • EdgeRIC Env (edge_ric.py)

• BeamArmor
  • Sections
  • Installing the repo
  • Preparing BeamArmor and MIMO-RIC
  • Files under concern
  • Parameters of MIMO-RIC
  • BeamArmor Demo: Automatic BeamArmor turn ON
  • Setup
    • Basestation/eNB setup
    • User/UE setup
    • Jammer
    • Phases of Demo
      • Demo procedure
      • Timer 1
      • Timer 2
      • Example
  • BeamArmor Demo: Manual BeamArmor turn ON/OFF
  • Setup
    • Basestation/eNB setup
    • User/UE setup
    • Jammer
  • Phases of Demo
    • Demo procedure
      • Timer 1
      • Example

• Tiny Twin

• Windex

• SPARC
  • Install srsRAN supporting E2 and EdgeRIC messages and control
    • Install zmq
    • Other Dependencies and cloning the repository
  • Network Setup
    • Core Network
    • Radio Access Network
    • Running the network:
      • Run the GRC broker
      • Run enb1
      • Run enb2
      • Run the UEs
    • Running Traffic
      • Running Downlink iperf traffic
      • Running Uplink iperf traffic
      • Running various other kinds of Downlink Traffic profiles
  • Hierarchical RIC deployment
    • Internal Messaging Infrastructure
      • Start the Redis database
    • Near RT RIC: Spectrum Monitoring
    • EdgeRIC: Resource Distribution
      • Running edgeric for 1 site
      • Running edgeric for 2 sites

• Open AI-RAN Tutorial: In-RAN Telemetry and Control with jbpf Codelets
  • Architecture
• Part 0: Bring up the testbed
  • Clone and set up the environment
  • Create the cluster
  • Deploy the 5G core and the subscribers
  • Deploy the RAN pod and the jrt-controller
  • Build the images (one time, ~20 min)
  • Start the radio
  • Traffic
• Part 1: Loading and unloading codelets
  • 1.1 The two YAML files you need to understand
  • 1.2 Load the MAC stats codeletset
  • 1.3 Unload
  • 1.4 Load/unload troubleshooting
  • 1.5 Start the dashboard
  • 1.6 Visualize: Grafana + VictoriaMetrics
• Part 2: Writing new codelets
  • 2.0 Anatomy of a codelet
  • 2.1 The design decision: bucket inside the codelet
  • 2.2 Codelet 1 — RLC queuing latency
  • 2.3 Codelet 2 — RLC buffer occupancy
  • 2.4 Advanced — lossless per-packet records
  • 2.5 Build, deploy, observe
  • 2.6 Sanity-check the measurement
  • 2.7 Traps (all of these were hit during development)
• Part 3: A control codelet for RLC buffer management
  • 3.1 Monitor hooks vs. control hooks
  • 3.2 The simplest control codelet: a fixed cap
  • 3.3 Run it
  • 3.4 What the control actually buys you
    • The setpoint ladder
  • 3.5 Closing the loop
• Appendix
  • Stop and status
  • Verification cheat sheet
  • Hooks used in this tutorial
  • Repository layout
  • Further reading