Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GenOnet: Generative Open xG Network Simulation with Multi-Agent LLM and ns-3

Farhad Rezazadeh, Amir Ashtari Gargari, Sandra Lagén, Josep Mangues, Dusit Niyato, Lingjia Liu

TL;DR

The paper addresses the challenge of testing open, interoperable 6G architectures by proposing GenOnet, a generative, multi-agent LLM-based framework that generates, debugs, executes, and interprets ns-3 network simulations for open xG environments. By integrating LangChain/LangGraph, Streamlit, and Retrieval-Augmented Generation, GenOnet automates the creation of ns-3 scripts (in Python and C++) for 5G-LENA NR scenarios while adhering to O-RAN and 3GPP standards. The work demonstrates end-to-end capabilities with concrete examples, including channel models and performance metrics, and discusses current limitations such as the need for manual debugging to achieve bug-free code. Overall, GenOnet aims to lower the barrier to open xG network simulations, enabling faster testing and validation of next-generation wireless architectures and protocols, with plans to broaden coverage to full 5G/6G scenarios and real-time analytics.

Abstract

The move toward Sixth-Generation (6G) networks relies on open interfaces and protocols for seamless interoperability across devices, vendors, and technologies. In this context, open 6G development involves multiple disciplines and requires advanced simulation approaches for testing. In this demo paper, we propose a generative simulation approach based on a multi-agent Large Language Model (LLM) and Network Simulator 3 (ns-3), called Generative Open xG Network Simulation (GenOnet), to effectively generate, debug, execute, and interpret simulated Open Fifth-Generation (5G) environments. The first version of GenOnet application represents a specialized adaptation of the OpenAI GPT models. It incorporates supplementary tools, agents, 5G standards, and seamless integration with ns-3 simulation capabilities, supporting both C++ variants and Python implementations. This release complies with the latest Open Radio Access Network (O-RAN) and 3GPP standards.

GenOnet: Generative Open xG Network Simulation with Multi-Agent LLM and ns-3

TL;DR

The paper addresses the challenge of testing open, interoperable 6G architectures by proposing GenOnet, a generative, multi-agent LLM-based framework that generates, debugs, executes, and interprets ns-3 network simulations for open xG environments. By integrating LangChain/LangGraph, Streamlit, and Retrieval-Augmented Generation, GenOnet automates the creation of ns-3 scripts (in Python and C++) for 5G-LENA NR scenarios while adhering to O-RAN and 3GPP standards. The work demonstrates end-to-end capabilities with concrete examples, including channel models and performance metrics, and discusses current limitations such as the need for manual debugging to achieve bug-free code. Overall, GenOnet aims to lower the barrier to open xG network simulations, enabling faster testing and validation of next-generation wireless architectures and protocols, with plans to broaden coverage to full 5G/6G scenarios and real-time analytics.

Abstract

The move toward Sixth-Generation (6G) networks relies on open interfaces and protocols for seamless interoperability across devices, vendors, and technologies. In this context, open 6G development involves multiple disciplines and requires advanced simulation approaches for testing. In this demo paper, we propose a generative simulation approach based on a multi-agent Large Language Model (LLM) and Network Simulator 3 (ns-3), called Generative Open xG Network Simulation (GenOnet), to effectively generate, debug, execute, and interpret simulated Open Fifth-Generation (5G) environments. The first version of GenOnet application represents a specialized adaptation of the OpenAI GPT models. It incorporates supplementary tools, agents, 5G standards, and seamless integration with ns-3 simulation capabilities, supporting both C++ variants and Python implementations. This release complies with the latest Open Radio Access Network (O-RAN) and 3GPP standards.
Paper Structure (4 sections, 4 figures)

This paper contains 4 sections, 4 figures.

Table of Contents

  1. Introduction
  2. System Overview
  3. Evaluation
  4. Conclusion and Future Work

Figures (4)

  • Figure 1: The graphical user interface of application.
  • Figure 2: An example of simulation generation for using the 5G-LENA helper with the standards such as the channel model.
  • Figure 3: The experimentation shows the execution and interpretation of a setup simulation using the channel model from TR 38.901 based on the 5G-Lena module.
  • Figure 4: Execution and interpretation of a Python-based example.