Hidden Markov Model-Based Spectrum Prediction and Dynamic Access for Space-Air-Ground Integrated Networks

To enhance spectrum utilization and alleviate interference in complex electromagnetic environments, this paper proposes a dynamic spectrum access method based on spectrum prediction. Real spectrum energy data in the 2.4GHz band is collected using a software-defined radio platform, and a multi-subband dataset is constructed. A two-state Gaussian Hidden Markov Model (HMM) is employed to capture the temporal correlation of spectrum occupancy, and a sliding window is introduced for one-step prediction. Based on the prediction results, an access strategy is developed by selecting subbands predicted to be low-energy and idle, thereby reducing interference. Experimental results show that the proposed method outperforms random and fixed access schemes in terms of bit error rate (BER) under various signal-to-noise ratio conditions. The proposed method can be applied to spectrum management in space-air-ground integration and other complex wireless environments.