A Clustering Approach for Attack Detection and Data Transmission in Vehicular Ad-hoc Networks
Atul Barve and Pushpinder Singh Patheja
To support a variety of applications for smart cities and intelligent passage systems, Vehicular Ad-hoc Networks (VANETs) are now becoming more common. To ensure dependable and stable VANETs communications, there are various difficult elements to overcome. The most priceless and endangered species and ecosystems are found in Natura 2000 (N2k), the biggest coordinated network of protected areas in the world. The rigid management structure of N2K sites, which primarily emphasizes conservation practices without a strategic vision for tying into the larger plans, has drawn significant criticism. With the help of three crucial phases that enable effective key management, this study attempts to design a strategy for the supportable running of N2K sites. Network initialization, key generation, and Key distribution. The Deep Neural Network aided Canonical Correlation Analysis (DNN-CCAS) method entails three steps: Cluster Formation (CF), Cluster Head selection (CH), outbreak recognition, and VANET security. The vehicles are initially clustered together by the Amended K-consonance expedient method. Next, the CH is chosen as one of the clusters using the linear measure promenade approach. Finally, if the CH is a normal one after detection, the DNN-CCAS is used to send the data contained in the CH securely to the cloud. In the assessment of prevailing techniques, the planned method achieves an accuracy of 91%.
Keywords: Vehicle ad hoc network, Natura 2000, DNN-CCAS, Amended K-consonance expedient, Linear measure promenade approach