CREST lets VANET vehicles abstain on weak event evidence
Original title: From Trust Scores to Driver Decisions: An Abstaining Evidential Decision Layer for Event Validation in VANETs
In a simulation of fourteen attacks, CREST cut two trust systems' decision error from 0.162 and 0.106 to 0.051.
The paper presents CREST, a vehicle-side framework for deciding whether a reported VANET event is real. Vehicles weight reports by sender trust, proximity and freshness, fuse them with Yager's rule, and abstain until the fused margin meets a threshold. Unresolved cases follow an explicit application policy. A pre-registered simulation with fourteen attacks, including position spoofing and trust-gated camouflage, reports lower decision error than two published trust-based dissemination systems.
Unconfirmed: Results come from the authors' pre-registered simulation, not a deployed VANET.
- A single report can change the decision margin by at most its weight.
- Attacker-coalition influence is bounded by the coalition's trust.
- The best-guess policy lowered both missed hazards and unnecessary reroutes.
- CREST beat the forced binary rule on both error types in every cell.
- All unspecified parameters were tuned by the same procedure.
From the original article
1 min read at arxiv.orgTrust management in vehicular ad hoc networks is highly active, yet the decision that a driver needs, whether a reported event is real, has received far less attention; the schemes that make it force every decision into two outcomes, even on thin or conflicting evidence. This paper presents CREST, a vehicle-side event-decision framework that can withhold a decision its evidence does not support.
By İpek Abasıkeleş Turgut · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.