Communication-Aware Dynamic Speed Control for Interference Mitigation in 6G Mobile Subnetworks

6G subnetworks may be installed in mobile entities such as robots navigating the factory floor in proximity. The potential high density of the subnetworks may result in interference limitation. Conventional interference mitigation approaches such as transmit power control achieve limited gain due to strong mutual interference when the subnetworks are nearby. Hence, an approach to mitigate interference by controlling the proximity of the mobile subnetworks may be required. In this paper, we study the potential of communication-aware dynamic speed control for interference mitigation in mobile subnetworks. The adjustment of the target speed of the mobile robots to minimize travel time is constrained by the minimum signal-to-interference plus noise ratio (SINR) requirement of their subnetwork. We propose a communication-aware speed control policy trained with reinforcement learning based on proximal policy optimization. With thorough simulation, we show that the method can effectively mitigate interference by controlling the proximity of the mobile subnetworks while incurring only a minor increase in average travel time compared to when the control is unaware of the communication requirement. It significantly improves the probability of achieving the stringent minimum SINR requirement from 75% to more than 95%.