A homotopy estimation based temporal-spatial spectrum prediction for UAV communications with arbitrary flight paths.

Area of Science:

• Electrical Engineering
• Computer Science
• Aerospace Engineering

Background:

• Rapid growth of Unmanned Aerial Vehicles (UAVs) leads to spectrum scarcity.
• Existing spectrum prediction methods struggle with real-time data limitations for moving UAVs.
• Need for advanced prediction models that can handle dynamic environments and limited historical data.

Purpose of the Study:

• To introduce a novel temporal-spatial spectrum prediction approach for UAVs during arbitrary flights.
• To address the challenge of estimating historical spectrum data at future locations.
• To develop a method that overcomes non-stationarity and correlation issues in spectrum data.

Main Methods:

• Extension of homotopy theory (HT) from two to multiple objects.
• Estimation of historical data using homotopy mapping derived from HT boundary conditions and model parameters.
• Spectrum prediction via a hidden Markov model (HMM) utilizing HT-estimated data, termed Multiple Objects HT-HMM (MOHT-HMM).

Main Results:

• The MOHT-HMM method effectively estimates historical spectrum data for dynamic UAV trajectories.
• The approach successfully predicts spectrum states at the next location without prior historical data.
• Experimental validation using real civil aviation data demonstrates high prediction accuracy.

Conclusions:

• The MOHT-HMM provides an effective solution for real-time UAV spectrum prediction in arbitrary flight scenarios.
• Homotopy theory is a viable tool for estimating missing or future data in dynamic systems.
• The proposed method enhances spectrum sharing capabilities for the growing UAV ecosystem.