Man-In-The-Middle Attacks in Vehicular Ad-Hoc Networks: Evaluating the Impact of Attackers' Strategies

Lu Liu, Farhan Ahmad, Asma Adnane, Virginia N. L. Franqueira, Fatih Kurugollu, Lu Liu 0001Sensors 2018 · DOI

TL;DR

This paper investigates the impact of Man-In-The-Middle (MITM) attacks on Vehicular Ad-Hoc Networks (VANETs) by simulating different attacker strategies: message delay, drop, and tamper. The findings reveal that these attacks significantly degrade network performance, with fleet-based attackers proving more disruptive than randomly distributed ones, highlighting the critical need for robust security in intelligent transportation systems.

What Problem Does This Paper Address?

Vehicular Ad-Hoc Networks (VANETs) are a cornerstone of Intelligent Transportation Systems (ITS), enabling vehicles to communicate with each other (V2V) and with roadside infrastructure (V2I) to enhance traffic efficiency and road safety. This communication relies on the timely and accurate exchange of critical information, such as collision avoidance alerts and steep-curve warnings. However, VANETs are vulnerable to various security threats, with Man-In-The-Middle (MITM) attacks posing a significant risk. In a MITM attack, a malicious node intercepts or manipulates messages between legitimate nodes, potentially leading to compromised information, increased delays, and packet loss. While MITM attacks are known, the specific impact of different attacker strategies and their deployment patterns (e.g., individual vs. coordinated) within the dynamic VANET environment remains underexplored. This paper aims to fill this gap by systematically evaluating the performance degradation caused by various MITM attack types and two distinct attacker distribution strategies.

Methods

The authors employ a simulation-based approach to evaluate the impact of MITM attacks. They utilize the Veins framework, which integrates the SUMO traffic simulator and the OMNET++ network simulator. This setup allows for realistic modeling of vehicular traffic and network communication.

The core of their methodology involves simulating three primary types of MITM attacks:

  • Message Delay: Malicious nodes intentionally delay the forwarding of legitimate messages.
  • Message Drop: Malicious nodes intercept and discard legitimate messages.
  • Message Tamper: Malicious nodes alter the content of legitimate messages, replacing it with misleading information.

Furthermore, they investigate two distinct attacker deployment strategies:

  • Distributed Attackers: Malicious nodes are spread randomly across the network.
  • Fleet Attackers: Malicious nodes operate in a coordinated, fleet-like structure, suggesting a more organized and collaborative attack.

The simulation involves 100 vehicles in a 2.5 km x 2.5 km area, with an accident event triggering message broadcasting. The percentage of malicious nodes is varied from 10% to 50% to observe the scalability of the attacks.

Key performance metrics used for evaluation include:

  • End-to-End Delay (E2ED): The time taken for a message to travel from its source to its destination.
  • Content Delivery Ratio (CDR): The proportion of successfully delivered messages to legitimate nodes.
  • Packet Loss Ratio (PLR): The percentage of packets lost during transmission.
  • Number of Compromised Messages: The count of messages that have been tampered with or delayed.
  • Number of Dropped Messages: The count of messages intentionally dropped by attackers.

Datasets

Not discussed in the paper. The study relies on simulated traffic data generated by SUMO and network traffic simulated within OMNET++.

Key Insights and Ideas

  • Fleet Attackers are More Disruptive: A significant finding is that attackers operating in a coordinated "fleet" structure have a more detrimental impact on network performance compared to randomly distributed attackers. This is attributed to their ability to concentrate their malicious actions, leading to greater congestion and more effective disruption of communication channels.
  • MITM Attacks Severely Degrade VANET Performance: Regardless of the specific attack type or strategy, MITM attacks consistently lead to increased end-to-end delays, reduced content delivery ratios, higher packet loss, and a greater number of compromised messages. This underscores the inherent vulnerability of VANETs to such threats.
  • Message Tampering is Particularly Insidious: While all MITM attacks are harmful, message tampering is highlighted as particularly dangerous because it injects false information into the network, which can lead to incorrect decisions by legitimate vehicles and potentially cause accidents.
  • Attackers Exploit Network Dynamics: The study implicitly shows how the dynamic nature of VANETs, with its high mobility and broadcast communication, can be exploited by attackers. The effectiveness of fleet attacks suggests that coordinated malicious actors could significantly disrupt traffic flow and safety systems.

Key Results

  • Message Delay: Fleet attackers caused significantly higher E2E delays than distributed attackers. For 50% malicious nodes, fleet attackers resulted in 73.44% higher E2E delays compared to distributed attackers.
  • Content Delivery Ratio (CDR): Distributed attackers led to lower CDRs, especially when dropping messages. For 50% malicious nodes, distributed attackers resulted in 44.89% lower CDR compared to fleet attackers in message drop scenarios.
  • Packet Loss Ratio (PLR): Distributed attackers generally resulted in higher PLR, particularly in message drop scenarios, due to their wider attack scope. For 50% malicious nodes, distributed attackers experienced 44.67% higher PLR in message drop attacks.
  • Compromised Messages: Fleet attackers compromised a higher percentage of messages, especially in message tampering scenarios. For 50% malicious nodes, fleet attackers compromised 12.23% more packets than distributed attackers.

Limitations and Future Work

Limitations:

  • The study focuses on a specific simulation environment and map, which may not generalize to all real-world VANET scenarios.
  • The mobility model used might not fully capture the complexities of real-world driving behaviors.
  • The paper primarily focuses on the impact of attacks rather than proposing comprehensive defense mechanisms, although it hints at potential solutions.

Future Work:

  • The authors propose extending this research by evaluating MITM attack models in VANETs with varying node mobility patterns.
  • A key future direction is to design and evaluate trust-based solutions to cater to MITM attacks, aiming to enhance the trustworthiness of information exchanged within VANETs.
  • Further research could explore the integration of plausibility checks and mobility-controlled criteria to mitigate the propagation of compromised messages.

Summary updated

BibTeX
@article{liu2018man,
  author = {Lu Liu and Farhan Ahmad and Asma Adnane and Virginia N. L. Franqueira and Fatih Kurugollu and Lu Liu},
  title = {{Man-In-The-Middle Attacks in Vehicular Ad-Hoc Networks: Evaluating the Impact of Attackers' Strategies}},
  journal = {Sensors},
  year = {2018},
  doi = {10.3390/s18114040},
}
Figure