Pre-Predictive Congestion-Based Data Allocation for Sixth Generation Cooperative Intelligent Transportation Systems

Abstract

Cooperative intelligent and autonomous transportation systems rely on intelligent sensing, computing, and actuating technologies for unmanned freight and public movements. The information gained from neighbors and communication infrastructures provides efficient actuation for safe and sustained transportation. This article resolves traffic data management congestion using sixth-generation (6G) communication and computing techniques. Terahertz and machine-type communications are exploited for swift information exchange, bypassing the congestion effects. Congestion occurs when demand for road space exceeds supply. This proposal incorporates prediction-based learning to compute the feasibility of handling traffic information and cooperative intelligent transportation. This model is named Congestion-aware Pre-predictive Data Allocation (CPPDA). The traffic flows causing congestion in the data exchange process are predicted for re-allocation and independent channel utilization. In this learning, the pre-predicted instances are updated with the actual identified utilization-to-congestion rate. Therefore, the congestion-causing channels for sensing are identified with ease, reducing the outage. The outage is examined for a basic inter-vehicle data link. Through the optimal allocation of channels for actuation, cloud-aided resources are utilized to a maximum level, leveraging infrastructure support. In addition to an outage of 10.83%, the response time of 14.75%, congestion factor of 8.2%, computational overhead of 6.4%, and information gain factors of 6.86% are analyzed through a comparative study.