Delay-minimization and back-off aware Q-learning with advanced bio-inspired CH selection for multi-hop communication in vehicular ad-hoc networks

The increasing significance of Vehicular Ad-hoc Networks (VANETs) in intelligent transportation systems has introduced challenges related to high mobility, network congestion, and energy efficiency. To address these challenges, this paper proposes a new approach based on Delay-Minimization and Back-Off Aware Q-Learning with Advanced Bio-Inspired Cluster Head (CH) Selection (DBACH) to enhance multi-hop data transmission in VANETs. The DBACH framework features network formation, latency minimization, a back-off Q-learning model, and an improved dragonfly algorithm-based CH selection. This method reduces transmission delay, routing overhead, and power consumption to enhance VANET QoS. DBACH was evaluated against RCDC, DCPA, and WCAM for effectiveness. The simulated vehicle numbers and speeds (km/h) were used to assess energy efficiency, throughput, packet delivery ratio, data loss ratio, computation time, and routing overhead. The DBACH boosts energy efficiency to 85 J, throughput to 160–200 Kbps, and packet delivery ratio to 11%–13%. Data loss drops to 7%–15%, latency is 60–94 ms, and routing overhead is 170–300 packets. When DBACH is a promising option for enhancing VANET communication dependability and energy economy due to its efficiency, communication stability, and success rates.