AI-Driven Spectrum Occupancy Prediction Using Real-World Spectrum Measurements

TL;DR

The paper tackles short-horizon spectrum occupancy prediction using 61 days of real-world mid-band spectrum measurements to enable proactive dynamic spectrum sharing. It formulates a next-minute, multi-bin prediction task based on a binary occupancy representation and evaluates three learning-based methods—Random Forest, XGBoost, and LSTM—against a first-order Markov baseline under a unified framework. Results show learning-based models outperform the Markov baseline on dynamic channels, with Random Forest offering a favorable balance of accuracy and deployment simplicity. The study demonstrates the practicality of lightweight AI models for near-term spectrum forecasting in DSS systems and highlights the value of real-world measurement data for guiding deployment decisions.

Abstract

Spectrum occupancy prediction is a critical enabler for real-time and proactive dynamic spectrum sharing (DSS), as it can provide short-term channel availability information to support more efficient spectrum access decisions in wireless communication systems. Instead of relying on open-source datasets or simulated data, commonly used in the literature, this paper investigates short-horizon spectrum occupancy prediction using mid-band, 24X7 real-world spectrum measurement data collected in the United States. We construct a multi-band channel occupancy dataset through analyzing 61 days of empirical data and formulate a next-minute channel occupancy prediction task across all frequency channels. This study focuses on AI-driven prediction methods, including Random Forest, Extreme Gradient Boosting (XGBoost), and a Long Short-Term Memory (LSTM) network, and compares their performance against a conventional Markov chain-based statistical baseline. Numerical results show that learning-based methods outperform the statistical baseline on dynamic channels, particularly under fixed false-alarm constraints. These results demonstrate the effectiveness of AI-driven spectrum occupancy prediction, indicating that lightweight learning models can effectively support future deployment-oriented DSS systems.

Figures (7)

• Figure 1: 3D raw power spectrum data.
• Figure 2: Spectrum utilization prediction pipeline.
• Figure 3: Test accuracy vs channels.
• Figure 4: Dynamic and static channels.
• Figure 5: Accuracy vs channel dynamics.