Improving IoRT Networks: Cross-Tier Resource Allocation for Multi-Antenna UAV Relays in SAGIN

In response to reducing power consumption while reducing equipment and maintaining high data rate requirements in Space-Air-Ground Integrated Networks (SAGIN) as part of the Internet of Remote Things (IoRT) networks, this paper proposes a system model and optimization scheme. Our modified architecture features multi-antenna Unmanned Aerial Vehicle (UAV) relays, gateway selection among existing UAVs for communication with satellite, and integration of Free Space Optic (FSO) space links to enhance air-space link data rates. The optimization problem aims to minimize the weighted sum of the number of gateways and the total UAV power consumption by jointly optimizing UAV deployment, UAV power allocation, gateway selection, and channel allocation. By jointly adopting proposed size-constrained PSO-K-means clustering, Simulated Annealing (SA) method, and Successive Convex Approximation (SCA) method, a two-stage scheme is devised to solve the proposed NP-hard and non-convex problem. Our simulation results indicate that an average performance improvement of 44% is achieved by our proposed scheme compared to the case where only clustering is performed. Furthermore, our proposed solution finds the number of gateways for different scenarios. For instance, almost all UAVs are selected as gateways to minimize UAV power consumption without considering any constraint on the number of gateways. Moreover, the performance difference between considering total UAV power consumption and only transmission power in the system optimization is shown. The results indicate that taking circuit power into account influences the selection of the number of gateways, as circuit power consumption is substantially higher than transmission power.