This paper focuses on the next-generation mobile systems and fixed wireless networks that are evolving in response to the need for high-bandwidth, low-latency service support. As technologies such as the Industrial Internet of Things, Extended Reality (XR), and Human-Machine (H2M) collaboration are promoting industrial and social revolutions such as Industry 4.0/5.0 and Society 5.0, we propose a new H2M collaboration scheme to solve the delay and motion sickness problems that occur in the real-time synchronization process of XR contents. This scheme uses a high-accuracy prediction model such as a bidirectional long-short-term memory network to predict the user's head movement and adjust the machine's camera direction in advance based on this. In addition, we predict the change in XR frame size according to the user's head movement and calculate the bandwidth requirement, and propose a human-machine collaborative dynamic bandwidth allocation (HMC-DBA) scheme based on this. The simulation results show that the proposed HMC-DBA scheme satisfies the delay and jitter requirements of XR frames with lower bandwidth consumption than the existing schemes and improves the efficiency of network resource utilization.
