Generative Chaotic Hybrid Multi-Objective Optimization Approach for Satellite-UAV Cognitive Radio Networks

In this paper, we consider the cognitive radio satellite-UAV downlink communication system, which leverages cognitive radio technology to optimize spectrum utilization. Specifically, this scenario involves a low Earth orbit (LEO) satellite sharing the spectrum with a UAV swarm, where both systems communicate with ground users simultaneously. The co-existence of satellite and UAV downlink channels introduces significant interference, leading to challenges in maintaining communication efficiency and energy efficiency. We formulate this as a multi objective optimization problem (MOP), which aims to maximize the total transmission rates of both satellite and UAV users while minimizing the energy consumption of the UAV swarm. However, this MOP is NP-hard due to its complex nature involving large scale decision variables and conflicting objectives such as interference mitigation and energy efficiency. To tackle these challenges, we propose a generative chaotic hybrid multi-objective hiking optimization algorithm (GCHMHOA). The algorithm includes several enhancements, which are chaos-based population initialization for better global exploration, a generative population evolution using diffusion models to maintain diversity, and genetic operators to handle sequentially encoded decision variables. Simulation results demonstrate that the proposed GCHMHOA outperforms various state-of-the-art benchmark algorithms and achieves superior convergence and solution diversity. Specifically, the proposed GCHMHOA achieves approximately 48% higher satellite transmission rate, 11% higher UAV swarm transmission rate, and 4% lower energy consumption compared to the best baseline algorithm.