Resilient AI-driven channel modelling for 6G-based IoV: a controller–worker CNN-LSTM approach

With an increasing need to incorporate advanced communication systems and compliance with safety standards in vehicular communication, the role of sixth-generation (6G) technology is critical towards Internet-of-Vehicles (IoV). Adoption of 6G will facilitate sustainable communication performance towards minimal latency and higher throughput even in adverse environmental conditions in IoV. Current state-of-the-art methods have shown the involvement of Artificial Intelligence (AI) for boosting the proactive operations in IoV communication. A closer insight into the existing system has witnesses shortcomings associated with these frequently adopted approaches. When the AI models themselves are prone to being victimized by adversaries. Therefore, the proposed study presents a novel AI-based computational model that is capable of performing analysis of communication channels when exposed to variable forms of adversaries using convolution neural network (CNN) and long short-term memory (LSTM). The model contributes towards developing a sequential adversary where the input data is altered to degrade the AI model’s sustainability. In contrast, a novel Controller–Worker model has been introduced that can optimize the predictive outcome of channel evaluation. Assessed on an extensive test environment, the outcome exhibits the proposed system to score approximately 65% of minimized processing time and 51% of increased accuracy performance in contrast to baseline standalone as well as hybrid methods.