Dynamic Downlink Interference Management in LEO Satellite Networks Without Direct Communications

We investigate effective interference management for Low Earth Orbit (LEO) satellite networks that provide downlink services to ground users and share the same frequency spectrum range. Since there are multi-group LEO satellites with different constellation orbits, will experience time-varying due overlapping of main/side lobes beams, which becomes even challenging when interfering cannot communicate directly. To address problem, we consider two groups communication service in area, while competing resources. develop solutions maximize throughput manage under a certain level, without explicit message exchanges between groups. By exploiting statistical learning deep reinforcement techniques, learning-based resource allocation schemes evaluate their performance through extensive simulations. show effectiveness reward settings managements, demonstrate Deep Q-Network (DQN)-based scheme can achieve close-to-optimal performance.