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

TL;DR

This paper evaluates AI-based methods for short-term spectrum occupancy prediction using real-world measurements, demonstrating their superiority over traditional statistical models for dynamic spectrum access.

Contribution

It introduces a new dataset from real-world measurements and compares multiple AI models against a baseline, highlighting their effectiveness in spectrum prediction.

Findings

Learning-based models outperform statistical baseline on dynamic channels.

AI methods achieve higher accuracy under fixed false-alarm constraints.

Real-world data enhances the relevance of spectrum prediction results.

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.