Misbehaviour Detection using Machine Learning in Collective Perception Services - MSc Thesis Defense by: Kruthika Shantha Murthy

Friday, May 9, 2025 - 11:00

The School of Computer Science is pleased to present…

Misbehaviour Detection using Machine Learning in Collective Perception Services
MSc Thesis Defense by: Kruthika Shantha Murthy

Date: Friday, May 9th,2025

Time:  11:00 am

Location: Essex Hall, Room 186

Abstract:

The main goal of Cooperative Intelligent Transport Systems (C-ITS) is to enhance road safety through real-time data exchange between vehicles via Vehicle-to-Everything (V2X) communication. The connectivity between vehicles through V2X communication creates security risks because attackers can easily falsify position and speed data, which leads to vehicle misdirection and traffic system disruption. When operating in dynamic environments, existing rule-based and cryptographic detection systems lack sufficient adaptability and scalability. To address these challenges, this thesis presents supervised machine learning as a solution to detect misbehaviour in Collective Perception Services (CPS). The research uses the SimCPS dataset to simulate six attack types that target position and speed information across different attacker densities and target strategies.

The research presents a detection framework which analyzes collective perception messages to detect malicious activities effectively. Rather than relying on predefined rules, the model learns attack patterns from the data itself instead of using predefined rules, which enables it to handle various scenarios. The results demonstrate strong detection performance, particularly in identifying subtle manipulations that traditional methods often overlook.

Keywords: Cooperative Intelligent Transport Systems (C-ITS), Collective Perception Services (CPS),  Vehicle-to-Everything (V2X) Communication
 
Thesis Committee:

Internal Reader: Dr. Xiaobu Yuan              

External Reader: Dr. Maher Azzouz         

Advisor: Dr. Arunita Jaekel

Chair:    Dr.Asish Mukhopadhyay

Vector Logo

 

Registration Link (only MAC students need to pre-register)