A Dynamic Spectrum Decision-Making Method for UAV Swarms in Complex Electromagnetic Environments

To address frequency conflicts and electromagnetic interference (EMI) for UAV swarms operating in complex electromagnetic environments, this paper conducts a decoupling analysis of channel allocation and power selection in the spectrum decision-making process and proposes a spectrum decision-making method for UAV swarms in such scenarios. A complex electromagnetic scenario model is established by incorporating three dominant interference characteristics, including internal friendly frequency conflicts, adversarial jamming interference, and time-varying background electromagnetic fluctuations. On this basis, performance evaluation metrics are formulated to quantify the command–communication and radar reconnaissance of UAV nodes. Then we propose an Ant Colony Optimization-Double Deep Q-Network (ACO-DDQN) algorithm, where the ACO module implements global channel assignment and the DDQN module conducts adaptive transmit power selection. This decoupled design enables the algorithm to obtain an approximately optimal spectrum decision scheme for coordinated operations. Simulation experiments under different swarm sizes and jamming intensities verify the feasibility and effectiveness of the proposed algorithm.