Source-linked AI summary
Open-source emulation-based test environment to settle O-RAN-compliant trials
Ramon Fontes, Allan Martins, Vicente Sousa, Kaio Dantas, Lucas Medeiros, Pedro Alves, Marcelo Fernandes, Iago Rego, Eduardo Aranha, Vinícius Filho, Mateus Goldbarg, Wysterlanya Barros, Roger Immich, Augusto V. Neto
TL;DR
The paper addresses the limited availability and high cost of realistic, scalable O-RAN experimentation platforms. It presents Mininet-RAN, an open-source emulator that connects virtual wireless networks, real Linux applications, and near-RT RIC xApps for closed-loop testing. Across V2X and UAV use cases, the authors report that Mininet-RAN is a low-cost and viable alternative for experimenting with O-RAN-compliant algorithms and techniques.
Problem
Full-scale testbeds and SDR platforms are costly and difficult to scale, while accessible open-source tools for realistic O-RAN RIC experimentation remain limited.
Method
The paper develops Mininet-RAN, an open-source emulator that models RAN elements and connects real network traffic, wireless patterns, and O-RAN SC near-RT RIC xApps in a closed loop.
Results
For V2X handover and UAV resource-allocation use cases, Mininet-RAN proved a low-cost and viable alternative for experimenting with O-RAN-compliant algorithms and techniques.
Takeaways & Limitations
Mininet-RAN supports low-cost O-RAN experimentation without requiring deployment of an entire RAN infrastructure.
Takeaways & Limitations
Future 5G support depends on wwan_hwsim and linux-wpan initiatives whose needed updates are not yet available in the Linux kernel main track.
Abstract
from arXiv · showhide
Experimental tools are a key factor in both academic and industrial research communities to create design evaluations of new networking technologies that involve troubleshooting or changing the planning of deployed networks. Physical Software-Defined Radio (SDR) experimental platforms enable a design solution for the quick prototyping of wireless communication systems. However, SDR-based experimental platforms incur high costs, which leads to scalability limitations in the experimental settings. Having said this, network simulators, emulators, and new testbeds have attracted increasing attention. Emulation-based research prototyping can be distinguished from real communication networks and SDR-based platforms by allowing a tradeoff between cost and flexibility. This paper examines the Mininet-RAN emulation tool, which, as well as Radio Access Network (RAN) modeling, provides a way to test Open RAN Intelligent Controller (RIC) services without the need to deploy an entire RAN infrastructure. The Mininet-RAN creates virtual network elements, such as hosts, L2/L3 devices, controllers, and links, by combining some of the best emulator features, hardware testbeds, and simulators. By running the current code of standard practice Unix/Linux network applications and network stack, the Mininet-RAN enables real-world network data traffic patterns to be delivered to the RIC, regarding the most significant aspect of the dynamic generation of wireless system's KPIs. We provide the basic code of Mininet-RAN for the first two O-RAN Alliance-defined use cases involving V2X and UAV. The xApps are being implemented in O-RAN SC near-RT RIC, with Mininet-RAN which provides a closed-loop validation environment.
1. INTRODUCTION
O-RAN experimentation needs affordable, flexible platforms because full-scale testbeds are costly and difficult to reconfigure, while existing open-source emulation options remain limited. The paper introduces Mininet-RAN as an open-source framework for low-cost, close-to-real O-RAN trials and closed-loop RIC validation.
- Full-scale testbeds are expensive, time-consuming, spatially complex, difficult to reconfigure, and hard to replicate across projects.
- Simulation and emulation provide alternative experimentation approaches, trading control and ease of setup against closer-to-real software and hardware integration.
- Mininet-RAN is proposed as an open-source, easy-to-use framework for low-cost, close-to-real O-RAN experimentation and RIC service validation.
- The paper contributes insights into O-RAN simulation and emulation tools, a low-cost evaluator for real traffic and wireless patterns, and a closed-loop experimental scenario.
- The code for Mininet-RAN and its use cases is made available through an open-source repository.
2. OVERVIEW OF O-RAN RIC
O-RAN centers on the RIC, which uses open interfaces and xApps to support interoperable, data-driven RAN control. Near-RT and non-RT control loops coordinate actions ranging from rapid RAN adaptation to longer-term policy and model management.
- The RIC is a central Open RAN component supporting multivendor interoperability and third-party xApps for RAN optimization, AI/ML guidance, QoS, and cost reduction.
- The O-RAN reference architecture includes near-RT RIC, non-RT RIC, and A1, E2, and O1 interfaces that exchange KPIs for 3GPP component control.
- Non-RT control operates above 1000 ms, whereas near-RT control operates between 10 ms and 1000 ms.
- Near-RT xApps request KPIs and return control actions through the E2 interface to adapt RAN parameters such as handovers and carrier frequencies.
- Near-RT and non-RT loops jointly support ML-powered management across short-term MAC procedures and longer-term network decisions.
3. RELATED WORK
Existing O-RAN experimentation solutions include proprietary platforms and several open-source projects, but the paper identifies a gap in highly accurate, end-to-end wraparound testing. Mininet-RAN is presented to address this gap with low-cost, controllable emulation and broader wireless-system support.
- Commercial O-RAN emulation solutions can support different-scale RIC trials but are considered too expensive for this work’s scope.
- O-RAN SC, SD-RAN, and Open Air Interface provide open-source software and emulation options for O-RAN-ready RANs.
- The O-RAN SC sim-e2-interface primarily tests E2AP message exchange, while proposed KPI integrations include a low-cost ns-3-based closed loop.
- Open Air Interface provides an end-to-end option using an nFAPI emulator, while SD-RAN supplies an E2 agent for E2AP.
- The paper reports that existing related solutions do not provide highly accurate wraparound O-RAN workflows and proposes a low-cost platform with PHY abstraction, MAC features, and real-time end-to-end service delivery.
4. MININET-RAN DESIGN AND WORKFLOW
Mininet-RAN combines a near-RT RIC architecture, xApps, external applications, and configurable Linux-based network emulation to support closed-loop KPI collection and control. Its workflow connects virtual wireless networks with dashboards, simulators, and RIC actions through APIs and messaging components.
- The closed-loop workflow supplies network information such as KPIs to xApps, applies resulting actions, and reflects those actions in subsequent measurements.
- The proposed near-RT RIC architecture contains four xApps, two external applications, RMR messaging, SDL storage, and connections to Mininet-RAN and a browser dashboard.
- The KPI Monitor requests KPIs through the xApp Interface, serializes them, and stores them in the SDL.
- The Anomaly Detection xApp evaluates SDL KPIs with a trained ML algorithm and sends resulting actions through RMR to subscribed services.
- The xApp Interface communicates with Mininet-RAN to request KPIs or execute emulator actions and provides network and RIC data to the dashboard.
- Mininet-RAN uses Python topology scripts, REST APIs, Scapy-based KPI collection, Linux wireless tools, and runtime commands, and interoperates with SUMO and CoppeliaSim.
5. CASE STUDIES
The case studies demonstrate Mininet-RAN’s integration with near-RT RIC for O-RAN use cases involving proactive V2X handover and dynamic UAV channel allocation. ML-powered xApps process emulated KPIs and trigger control actions, while dashboards monitor the resulting behavior.
- 5. CASE STUDIES: Mininet-RAN implements two O-RAN Alliance use cases: context-based dynamic handover management for V2X and flight path-based dynamic UAV resource allocation.The source code and reproducible results are publicly available, and the experiments use SUMO for vehicle mobility and CoppeliaSim for UAV flight paths.
- 5.1 Use case 1: Context-based dynamic handover management for V2X: 99.84% model accuracy was achieved for V2X handover prediction with a two-hidden-layer MLP, while similar training and test accuracy suggested no overfitting.GridSearchCV selected hidden layers of 40 and 20 neurons, and the model used four inputs with three BS-switching probability outputs.
- 5.1.2 Results of use case 1: The V2X xApp receives BS measurements through the near-RT RIC and recommends handover as the BS3 SNR exceeds BS1, while delaying the decision to avoid ping-pong effects.The vehicle begins connected to BS1, receives a high-resolution video stream, and moves along a SUMO-defined route while KPIs are measured.
- 5.2 Use Case 2: Flight path-based dynamic UAV resource allocation: The UAV xApp dynamically selects channels for moving UAVs to mitigate same-channel interference between land BSs and airborne UAVs.The near-RT RIC measures KPIs, supplies the ML-powered xApp, and triggers channel changes as UAV power levels and channel assignments evolve.
- 5.2.1 Machine learning of use case 2: Similar training and testing accuracy indicated that the UAV channel-selection model adapted to its training data and was likely to perform well on unseen data, while making four classification mistakes.The model used a 70% training and 30% testing split; errors involved Channels 5 and 12 as false selections and missed Channels 1 and 3.
6. LIMITATIONS OF THE STUDY AND SUGGESTION FOR FUTURE WORK
Mininet-RAN currently lacks support for key 5G elements and O-RAN E2 signaling, while future work targets Linux kernel support for LTE and 5G NR experimentation.
- Mininet-RAN does not currently support 5G technologies such as gNodeBs or their typical signaling signals.
- O-RAN E2 signaling is planned but has not yet been added to Mininet-RAN.
- Future 5G support is expected to use the wwan_hwsim Linux Kernel module.
- The future 5G direction also includes linux-wpan5, whose kernel updates are not yet in the Linux main track.
7. CONCLUSION
The paper presents Mininet-RAN as a low-cost platform for open-source 5G and O-RAN experimentation, evaluated through V2X and UAV use cases. It supports closed-loop experimentation with O-RAN-compliant algorithms and techniques.
- Mininet-RAN is a low-cost, viable alternative for experimenting with new O-RAN-compliant algorithms and techniques.
- The evaluation covers context-based dynamic handover management for V2X and flight path-based dynamic UAV resource allocation.
- The study does not focus on machine-learning efficiency or proposing new efficient techniques for the evaluated scenarios.
- Mininet-RAN is reported as the only related solution achieving complete PHY abstraction and end-to-end evaluation with O-RAN interfacing.