[Paper Review] Data-centric Misbehavior Detection in VANETs

Sushmita Ruj, Marcos Antônio Cavenaghi+3 more|arXiv (Cornell University)|Mar 12, 2011

Vehicular Ad Hoc Networks (VANETs)19 references9 citations

TL;DR

This paper proposes a data-centric misbehavior detection scheme for VANETs that identifies false alerts and location information by analyzing post-alert vehicle behavior, eliminating the need for voting or certificate revocation. By imposing fines instead of revoking credentials, the scheme reduces communication and computational overhead while maintaining resilience to Sybil attacks and preserving location privacy through pseudonyms.

ABSTRACT

Detecting misbehavior (such as transmissions of false information) in vehicular ad hoc networks (VANETs) is very important problem with wide range of implications including safety related and congestion avoidance applications. We discuss several limitations of existing misbehavior detection schemes (MDS) designed for VANETs. Most MDS are concerned with detection of malicious nodes. In most situations, vehicles would send wrong information because of selfish reasons of their owners, e.g. for gaining access to a particular lane. Because of this (\emph{rational behavior}), it is more important to detect false information than to identify misbehaving nodes. We introduce the concept of data-centric misbehavior detection and propose algorithms which detect false alert messages and misbehaving nodes by observing their actions after sending out the alert messages. With the data-centric MDS, each node can independently decide whether an information received is correct or false. The decision is based on the consistency of recent messages and new alert with reported and estimated vehicle positions. No voting or majority decisions is needed, making our MDS resilient to Sybil attacks. Instead of revoking all the secret credentials of misbehaving nodes, as done in most schemes, we impose fines on misbehaving nodes (administered by the certification authority), discouraging them to act selfishly. This reduces the computation and communication costs involved in revoking all the secret credentials of misbehaving nodes.

Motivation & Objective

• To address the limitations of entity-centric misbehavior detection in VANETs, which focus on identifying malicious nodes rather than false data.
• To detect false alert messages and incorrect location reports by analyzing the consistency of a vehicle's reported position with its actual post-alert movement.
• To reduce communication and computational costs by avoiding full certificate revocation and instead imposing financial fines on misbehaving nodes.
• To maintain location privacy through the use of pseudonyms, preventing long-term tracking of vehicles.
• To design a scheme resilient to Sybil attacks by eliminating reliance on voting or majority decisions.

Proposed method

• Each vehicle independently verifies the consistency between an alert message and the subsequent actual positions of the sender using real-time location data.
• False information is detected when a vehicle's reported position after an alert deviates significantly from the expected trajectory based on prior movement.
• The scheme uses pseudonyms to preserve location privacy, ensuring that a vehicle's identity cannot be inferred from multiple pseudonyms.
• Instead of revoking all credentials of a misbehaving node, the system imposes fines administered by the certification authority to discourage repeated false reporting.
• The detection mechanism does not require message voting or consensus, making it robust against Sybil attacks where multiple fake identities are created.
• Communication overhead is minimized since no certificate revocation list (CRL) is transmitted; only the identity of misbehaving nodes is reported to nearby RSUs.

Experimental results

Research questions

• RQ1 How can false alert messages and incorrect location reports be detected in VANETs without relying on voting or majority decisions?
• RQ2 What mechanisms can detect selfish misbehavior in vehicles—such as falsely reporting congestion or hazards—without classifying the entire node as malicious?
• RQ3 How can communication and computational overhead be reduced in misbehavior detection schemes by avoiding full certificate revocation?
• RQ4 In what ways can location privacy be preserved while still enabling accurate detection of false data in vehicular networks?
• RQ5 How can Sybil attacks be mitigated in a misbehavior detection system that does not depend on centralized trust or voting?

Key findings

• The proposed data-centric misbehavior detection scheme reduces communication overhead by eliminating the need to distribute certificate revocation lists (CRL), as only the identity of misbehaving nodes is reported.
• The scheme achieves a verification time of 5.1ms, calculated as 3T_par + 2T_mul with T_par = 4.5ms and T_mul = 0.6ms, indicating low computational latency.
• By imposing fines instead of revoking credentials, the scheme reduces the overhead associated with key revocation and credential management, especially in high-misbehavior scenarios.
• The method is resilient to Sybil attacks because it does not rely on voting or majority decisions, and detection is based on behavioral consistency rather than reputation aggregation.
• The scheme successfully detects false alerts by comparing the trajectory of a vehicle after sending an alert with the expected path, enabling real-time validation of reported information.
• Limitations include undetected misbehavior when vehicles make U-turns or travel on complex road structures like flyovers, where Euclidean distance may not reflect actual travel distance.