Mobility and Link Reliability Estimation Using CSS And ML In CR VANET

Abstract

Intelligent transportation system (ITS) applications may rely on the reliable infrastructure that vehicular ad hoc networks (VANETs) offer. Roadside units (RSUs) are the main component of vehicle-to-vehicle and vehicle-to-infrastructure links in VANET communication. When cognitive radio (CR) VANET investigations were analysed, two major performance problems were found: low connection stability because of the high vehicle movement and excessive energy consumption. We suggest a fresh strategy to deal with these issues: Mobility and Link Reliability Estimation Using CSS AND ML CR-VANETs, known as MLRECSS CR-VANET. MLRECSS CR-VANET consists of four main components: CR-VANET construction, Cooperative spectrum Sensing model, Speed-based mobility prediction, link-based multipath routing. First, we create CSS-based CR-VANETs to examine and address issues with power consumption and spectrum scarcity in VANETs. Vehicle speed forecasts and fluctuations are assessed via mobility prediction. Lastly, routing is made reliable and effective by utilizing the machine learning and ad hoc on-demand multipath distance vector (AOMDV) routing protocol in combination with link stability based multipath routing (LSMR). MLRECSSCR-VANET technique outperforms the previous methods by 3.69% in terms of packet delivery ratio,7.21% in terms of residual energy, 6.09% in terms of throughput, 8.33% in terms of residual node speed and 13.97% in terms of energy efficiency. Comparing it to more contemporary efforts like LMCCR-VANET, SCCR-VANET, CFCR-VANET, and MMCR-VANET, it shows improved energy efficiency, delivery rates, decreased energy consumption, end-to-end latency, and routing overhead.