A Novel OMNeT++-Based Simulation Tool for Vehicular Cloud Computing in ETSI MEC-Compliant 5G Environments

Angelo Feraudo, Alessandro Calvio, Paolo Bellavista

TL;DR

This paper introduces a specialized simulation tool that integrates Vehicular Cloud Computing (VCC) with the ETSI Multi-access Edge Computing (MEC) standard in 5G environments. By extending the popular OMNeT++ and Simu5G frameworks, the authors provide a platform to test how parked vehicles can serve as "far-edge" micro-datacenters to reduce latency and backbone network stress.

What Problem Does This Paper Address?

As 5G and IoT applications grow, traditional "client-to-cloud" architectures struggle with stringent latency requirements. Edge computing (MEC) solves this by placing resources at the network's edge, but infrastructure is expensive to deploy.

Vehicular Cloud Computing offers a solution: using the idle computing power of parked or moving vehicles. However, the research community lacked an integrated simulation framework that could model both the vehicular paradigm and standardized ETSI MEC architectures within a 5G context. Without such a tool, researchers cannot easily design or validate algorithms for distributing applications across these highly dynamic, transient vehicular resources.

Methods

The authors developed their tool by significantly extending OMNeT++ and the Simu5G library. Key technical contributions include:

  • Architectural Extension: They implemented an extended version of the ETSI MEC reference architecture that treats vehicles as "far-edge" nodes.
  • The car Module: A new module that extends the standard New Radio User Equipment (NRUE). It acts as a wrapper for 5G-enabled devices, providing computational resources (CPU, RAM, storage) and running a local virtualization infrastructure.
  • Management Entities: They created specific applications like ClientResApp (to handle resource registration/rewards) and VIApp (to manage local resource allocation) that run on the vehicles.
  • Broker Integration: An external "Broker" entity was added to the MEC system level to manage subscriptions to an Area of Interest (AoI), notifying the system when new vehicular resources enter or leave a specific zone.

Datasets

The authors validated their simulator using two real-world datasets to ensure realistic mobility and demand patterns:

  1. Arnhem Dataset: Open data from parking garages in Arnhem, Netherlands, used to model vehicle entry rates and residency times (how long a car stays parked).
  2. Bologna WiFi Dataset: Data from the city of Bologna, Italy, used to mimic user activity levels and application request frequencies throughout a 24-hour cycle.

Key Insights and Ideas

  • Vehicles as "Far-Edge" Hosts: The paper demonstrates that vehicles shouldn't just be seen as data consumers, but as valid extensions of the MEC host level.
  • The Volatility Challenge: Unlike fixed edge servers, vehicular resources are transient. The "aha moment" here is the necessity of a predictive scheduler. By using Gaussian distributions to predict how long a car will remain parked, the system can drastically reduce "migrations" (moving a running app from a car that is about to leave to another host).
  • Standardization is Key: By sticking to ETSI MEC specifications, the tool ensures that simulated applications and management logic are closer to what would actually be deployed in real-world 5G networks.

Key Results

  • Protocol Efficiency: The "join time" (the time it takes for the MEC system to recognize and integrate a new vehicle) ranged from 13 to 40 ms, depending on the number of cars. The "release time" remained constant at roughly 7 ms.
  • Scalability: Even under a "worst-case scenario" where 300 users requested applications simultaneously, the allocation delay introduced by the tool's management logic remained negligible at approximately 40 ms.
  • Scheduler Performance: In a 24-hour simulation, their Custom Scheduler (which predicts residency time) significantly outperformed standard "Best First" and "Round Robin" algorithms. While the "Best First" approach triggered over 60 migrations at peak hours, the custom predictive scheduler kept migrations around 20, greatly increasing service reliability.

Limitations and Future Work

The current implementation of the tool primarily focuses on parked vehicles as far-edge nodes. While the architecture supports mobility, the specific use case and evaluation centered on the stability of parked resources.

The authors suggest that future versions could incorporate more sophisticated Machine Learning (ML) techniques for the scheduler to better handle the high mobility of vehicles on the move. Additionally, exploring the energy consumption impact on vehicles providing these services remains a promising area for further study.

Summary updated

Abstract

Vehicular cloud computing is gaining popularity thanks to the rapid advancements in next generation wireless communication networks. Similarly, Edge Computing, along with its standard proposals such as European Telecommunications Standards Institute (ETSI) Multi-access Edge Computing (MEC), will play a vital role in these scenarios, by enabling the execution of cloud-based services at the edge of the network. Together, these solutions have the potential to create real micro-datacenters at the network edge, favoring several benefits like minimal latency, real-time data processing, and data locality. However, the research community has not yet the opportunity to use integrated simulation frameworks for the easy testing of applications that exploit both the vehicular cloud paradigm and MEC-compliant 5G deployment environments. In this paper, we present our simulation tool as a platform for researchers and engineers to design, test, and enhance applications utilizing the concepts of vehicular and edge cloud. Our platform significantly extends OMNet++ and Simu5G, and implements our ETSI MEC-compliant architecture that leverages resources provided by far-edge nodes. In addition, the paper analyzes and reports performance results for our simulation platform, as well as provides a use case where our simulator is used to support the design, test, and validation of an algorithm to distribute MEC application components on vehicular cloud resources.
BibTeX
@inproceedings{feraudo2023novel,
  author = {Angelo Feraudo and Alessandro Calvio and Paolo Bellavista},
  title = {{A Novel OMNeT++-Based Simulation Tool for Vehicular Cloud Computing in ETSI MEC-Compliant 5G Environments}},
  booktitle = {SIMULTECH},
  year = {2023},
  doi = {10.5220/0012140900003546},
}

Loading related content...

Figure