Quantum Deep Q Network Technique for Latency Minimization in STAR-RIS assisted VRCS

The increasing demand for ultra-reliable and low-latency communication (URLLC) in vehicle road cooperation systems (VRCS) has propelled the development of intelligent and efficient optimization techniques. This paper presents a Quantum Deep Q-Network (QDQN) based approach for minimizing latency in a Simultaneously Transmitting and Reflecting Reconfigurable Intelligent Surface (STAR-RIS) enabled VRCS. STAR-RIS improves signal coverage and energy efficiency by simultaneously serving users in both transmission and reflection modes. However, latency optimization remains a critical challenge due to dynamic environments and computational complexity. The proposed QDQN technique integrates quantum computing principles with deep reinforcement learning (DRL) to accelerate decision making and optimize resource allocation in real time. Using quantum parallelism and entanglement, QDQN reduces convergence time while effectively learning the dynamic state of the communication environment. The simulation results demonstrate that the proposed method achieves a significant latency reduction compared to conventional DRL and classical Q-learning techniques. This study highlights the potential of quantum-enhanced reinforcement learning for future URLLC applications in intelligent vehicular networks.