Cluster channel equalization using adaptive sensing and reinforcement learning for UAV communication

Area of Science:

• * Electrical Engineering
• * Computer Science
• * Artificial Intelligence

Background:

• * Uncrewed aerial vehicle (UAV) cluster communications face challenges in dynamic sensing and channel equalization.
• * Existing methods struggle with real-time adaptation to complex channel environments and noise reduction.

Purpose of the Study:

• * To develop an advanced equalization algorithm for UAV communication systems.
• * To enhance dynamic sensing, channel equalization, and resource utilization in UAV networks.

Main Methods:

• * Development of a U-Net-based signal processing algorithm for noise reduction and real-time channel state perception.
• * Enhancement of fuzzy reinforcement Q-learning with a fuzzy neural network for improved Q-value approximation and adaptability.
• * Integration of the U-Net model and enhanced fuzzy reinforcement Q-learning into the U-FRQL-EA algorithm.

Main Results:

• * The U-FRQL-EA algorithm significantly reduces the bit error rate (BER) in UAV communication systems.
• * Demonstrated enhancement in overall communication quality and network resource optimization.
• * Effective real-time channel state sensing and intelligent data forwarding strategies.

Conclusions:

• * The U-FRQL-EA algorithm offers a novel and effective solution for improving UAV communication performance.
• * The integration of U-Net and fuzzy reinforcement Q-learning addresses key challenges in dynamic sensing and equalization.
• * The proposed method optimizes network resource utilization and enhances communication reliability.