[Paper Review] 5G enabled Mobile Edge Computing security for Autonomous Vehicles

Daryll Ralph D'Costa, Hassan Abbas|arXiv (Cornell University)|Jan 30, 2022

Vehicular Ad Hoc Networks (VANETs)4 citations

TL;DR

This paper proposes a machine learning-based framework to detect and classify Distributed Denial of Service (DDoS) attacks in 5G-enabled Mobile Edge Computing (MEC) environments supporting autonomous vehicles. It enhances a generic dataset with synthetic 5G network parameters, applies label encoding and minority class oversampling, and trains models to predict DDoS attacks and 5G latency quality, though performance is limited by insufficient real-world 5G data and reliance on simulated, single-machine experiments.

ABSTRACT

The world is moving into a new era with the deployment of 5G communication infrastructure. Many new developments are deployed centred around this technology. One such advancement is 5G Vehicle to Everything communication. This technology can be used for applications such as driverless delivery of goods, immediate response to emergencies and improving traffic efficiency. The concept of Intelligent Transport Systems (ITS) is built around this system which is completely autonomous. This paper studies the Distributed Denial of Service (DDoS) attack carried out over a 5G network and analyses security attacks, particularly the DDoS attack. The aim is to implement a machine learning model capable of classifying different types of DDoS attacks and predicting the quality of 5G latency. The initial steps of implementation involved the synthetic addition of 5G parameters into the dataset. Subsequently, the data was label encoded, and minority classes were oversampled to match the other classes. Finally, the data was split as training and testing, and machine learning models were applied. Although the paper resulted in a model that predicted DDoS attacks, the dataset acquired significantly lacked 5G related information. Furthermore, the 5G classification model needed more modification. The research was based on largely quantitative research methods in a simulated environment. Hence, the biggest limitation of this research has been the lack of resources for data collection and sole reliance on online data sets. Ideally, a Vehicle to Everything (V2X) project would greatly benefit from an autonomous 5G enabled vehicle connected to a mobile edge cloud. However, this project was conducted solely online on a single PC which further limits the outcomes. Although the model underperformed, this paper can be used as a framework for future research in Intelligent Transport System development.

Motivation & Objective

• To address security vulnerabilities in 5G-enabled Mobile Edge Computing (MEC) for autonomous vehicles, particularly DDoS attacks.
• To develop a machine learning model capable of classifying different types of DDoS attacks in 5G network environments.
• To predict the quality of 5G network latency as a performance metric in the context of MEC and autonomous vehicle operations.
• To evaluate the feasibility of using synthetic 5G parameters to augment existing datasets for training robust DDoS detection models.

Proposed method

• Synthetic 5G network parameters were added to an existing dataset to simulate real-world 5G MEC conditions.
• Data preprocessing included label encoding of categorical features and oversampling of minority attack classes to balance the dataset.
• The dataset was split into training and testing sets using standard partitioning techniques.
• Multiple machine learning models were trained and evaluated for DDoS attack classification and 5G latency prediction.
• Model performance was assessed using standard classification metrics, though specific values were not reported in the abstract.
• The study was conducted entirely in a simulated environment on a single PC, relying on publicly available datasets.

Experimental results

Research questions

• RQ1 How effective is a machine learning model in classifying different types of DDoS attacks within a 5G-enabled Mobile Edge Computing environment for autonomous vehicles?
• RQ2 To what extent can synthetic 5G parameters improve the representativeness and utility of existing datasets for DDoS detection in vehicular networks?
• RQ3 Can machine learning models accurately predict 5G network latency quality under DDoS attack conditions in MEC-based ITS systems?
• RQ4 What are the limitations of using simulated, single-machine environments and synthetic data for training and validating DDoS detection models in real-world autonomous vehicle scenarios?

Key findings

• The proposed machine learning model successfully detected and classified DDoS attacks in the simulated 5G MEC environment.
• Despite model functionality, performance was limited due to insufficient real 5G network data and over-reliance on synthetic parameters.
• The dataset used in the study significantly lacked authentic 5G-related information, affecting model generalization and reliability.
• The model's ability to predict 5G latency quality was constrained by the same data limitations and simulation constraints.
• The research identified a critical gap in real-world data availability for training robust DDoS detection systems in 5G-enabled autonomous vehicle networks.
• While the model underperformed, the framework provides a foundational approach for future research in Intelligent Transport Systems and 5G MEC security.