Resource Allocation in UCAN: A Multi-Objective Optimization Method Based on Heterogeneous Graph Decision Space

The satellite-terrestrial integrated networks (STINs) bring a wider range of communication connections to 6G. Due to the reduction in the base station coverage and the number of users continues to increase, user-centric access network (UCAN) has become a more common access method. However, more access options and more complex co-frequency interference make communication resource allocation more difficult. To address this issue, this paper proposes a multi-objective optimization method for communication resources based on heterogeneous graph representation in UCAN. Firstly, this method utilizes heterogeneous graphs to represent network topology and communication resources, extracting a mixture of discrete and continuous decision space from the attributes of nodes and edges. Secondly, a multi-objective optimization model is established to simultaneously optimize the interference, throughput, and energy consumption in UCAN through user access selection and power control, and solved using algorithm. Finally, compared with the benchmark methods, the superiority of the proposed method is demonstrated.