VANET Security Optimization with Blowfish Algorithm and Adversarial Transfer Learning

Transportation systems have taken new heights ever since they are combined to wireless technologies. With network protocols being developed and optimized a whole new era of intelligent vehicles is here. However, the issue of security sustains as attacks and malicious elements in networks that can harm the integrity of whole system have grown too. In this paper, we have presented a framework for structuring the vehicular network carrying sensitive data in a open source environment. The unique vehicular identification number is generated by our pre-processing algorithm, followed by encryption provided by Blowfish algorithm. To optimize the data security, we will incorporate the adversarial transfer learning model as a defense mechanism against all known as well unknown attacks. The entire works shows promising results with robustness against cyber-attacks and the scalability rankings to be deployed over a large-scale networks. With our framework we have provided resultant model that have been proven to be more stable and immune to VANET attacks. We have analyzed our work over parameters like maximum stored keys, encryption time, battery capacity required, and memory required, it’s been observed that with optimization, we have provided better results.