Misbehaviour Prediction in CAV Network using Aggregated Trust Analysis
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Abstract Connected and Autonomous Vehicle (CAV) technology is rising in the transport sector due to its enormous benefits, including better traffic safety and efficiency. The intelligent decision in CAV relies on traffic data from in-vehicle sensors, neighbouring vehicles, and roadside infrastructure. The traffic data is prone to manipulation by internal and external attacks. This NTU’s CAV research focuses on message manipulation attacks, such as false position attacks. In this context, this paper proposes a novel misbehaviour prediction framework to identify malicious positions in CAV networks based on an aggregated trust analysis strategy. The proposed framework utilises Mamdani fuzzy logic for identification and will be applied to the onboard unit of every vehicle without depending on external infrastructure for validation. Through repeated observation of the concerned CAVs’ behaviour and trust values aggregation, the framework can predict potential misbehaviour accurately before it leads to safety-critical issues. The framework is validated using the VeReMi extension dataset and tested with supervised machine learning algorithms, including Decision Tree (DT), k-nearest Neighbour (KNN), Logistic Regression (LR), and Support Vector Machine (SVM). The simulation results show that the DT and SVM yield the highest sensitivity in detecting false positions. Although we used the results with KNN, the default module in previous studies, we discovered that it was competitively effective in identifying all position attacks, doing very well with eventual stop position attacks. Additional examination of the trust aggregation fuzzy model indicates that the F1 measure for constant false position attacks, random position attacks, and eventual stop position attacks is above 0.99 on average. In contrast, the F1 score for random and fixed offsets is less than 0.70. To achieve improved performance, we use weights that are inversely proportional to the sender-receiver distance for post-processing. We also analyse the performance of our proposed model against other fuzzy logic models. Overall, our fuzzy model shows competitive performance in detecting false position data, as indicated by the performance values.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it