LAUSR

Tunnel construction is a dynamic excavation process that requires a variety of construction machinery and workers to cooperate according to standard procedures. However, due to insufficient information sharing and inefficient construction situation awareness, the operation efficiency cannot be largely enhanced, and the safety of workers cannot be guaranteed. To deal with the problem, it is imperative to establish a wireless communication system for construction machinery and workers. In this paper, we propose a wireless communication architecture for tunnel construction to carry multiple services including video, voice and sensor data with different quality-of-service (QoS) provision. A resource block (RB) allocation and power control scheme is developed for uplink transmission, where the long-term energy consumption is minimized and the transmission performance is optimized. As a discrete time-constrained stochastic optimization problem, it can be modeled as a constrained Markov decision process (CMDP). Furthermore, the dueling deep Q-network (Dueling-DQN) is utilized to obtain the scheduling policy based on the wireless fading characteristics of tunnels and network queues state. Simulation results demonstrate that the proposed scheme achieves better performance in a variety number of user equipment (UE) scenarios. The scheme gives priority to VoIP and sensor services in the limited bandwidth, which can provide efficient information sharing for tunnel construction.