Enhanced Slime Mould Optimization with Deep-Learning-Based Resource Allocation in UAV-Enabled Wireless Networks.

Area of Science:

• * Wireless Communication Networks
• * Artificial Intelligence in Telecommunications
• * Optimization Algorithms

Background:

• * Unmanned Aerial Vehicle (UAV) networks are crucial for diverse applications like public safety and disaster management.
• * Providing reliable communication to mobile users (MUs) in dynamic UAV environments presents significant challenges.
• * Efficient resource allocation (subchannels, power, user serving) is vital for coverage and energy efficiency in UAV networks.

Purpose of the Study:

• * To present an Enhanced Slime Mould Optimization with Deep-Learning-based Resource Allocation Approach (ESMOML-RAA) for UAV-enabled wireless networks.
• * To achieve computationally and energy-effective resource allocation decisions.
• * To enhance coverage and energy efficiency in UAV-assisted transmission networks.

Main Methods:

• * Developed the ESMOML-RAA technique, treating the UAV as a learning agent for resource assignment.
• * Employed a highly parallelized long short-term memory (HP-LSTM) model for resource allocation.
• * Utilized the Enhanced Slime Mould Optimization (ESMO) algorithm to optimize HP-LSTM hyperparameters.

Main Results:

• * The ESMOML-RAA technique demonstrated efficient computation and energy usage.
• * The approach successfully minimized weighted resource consumption through a designed reward function.
• * Simulations confirmed superior performance of ESMOML-RAA compared to other machine learning models.

Conclusions:

• * ESMOML-RAA offers an effective solution for resource allocation challenges in UAV networks.
• * The integration of ESMO and HP-LSTM significantly enhances network performance and efficiency.
• * This approach provides a robust framework for optimizing UAV-enabled wireless communication systems.