Detection of Sybil Attacks in Vanet using Transformer With CNN Hybrid Architecture and Optimized by AdamW Optimizer

Vehicular Ad Hoc Networks (VANETs) help improve road safety by letting cars talk to each other but they still face Sybil attacks. In a Sybil attack, bad nodes claim many identities and weaken the network. This paper offers a Transformer-CNN model that uses the AdamW optimizer to spot Sybil attacks in VANETs. It uses spatial as well as time data from vehicular communications. The model works best at finding these attacks. The system is measured with metrics like accuracy, precision, and it shows big gains compared to older methods.