An intelligent bio-inspired multi-objective and scalable UAV-assisted clustering algorithm in flying ad hoc networks

Area of Science:

• Computer Science
• Network Engineering
• Artificial Intelligence

Background:

• Flying Ad Hoc Networks (FANETs) are crucial for mission-critical Unmanned Aerial Vehicle (UAV) operations.
• Efficient, scalable, and adaptive clustering is essential for managing UAV swarms in dynamic 3D environments.
• Existing clustering algorithms face challenges in high-mobility and energy-constrained FANETs.

Purpose of the Study:

• To introduce a novel clustering optimization framework for FANETs using the Secretary Bird Optimization Algorithm (SBOA).
• To enhance cluster head (CH) selection by optimizing intra-cluster distance, residual energy, and load balancing.
• To evaluate SBOA's performance against other metaheuristic algorithms in terms of efficiency, scalability, and robustness.

Main Methods:

• Developed a multi-objective optimization framework based on the bio-inspired SBOA.
• Simulated UAV networks with varying node populations (30-160) and communication ranges (100-900m).
• Compared SBOA against Fire Hawk Optimization Algorithm (FHOA), Portia Spider Optimization Algorithm (PSOA), and MOSFP.

Main Results:

• SBOA achieved up to 15% higher optimization fitness and 10% greater cluster density.
• Demonstrated a 40% reduction in load imbalance and improved convergence stability.
• Showcased over 85% optimal fitness in sparse environments, confirming scalability and robustness.

Conclusions:

• SBOA provides a superior, robust, and scalable clustering solution for real-time, energy-constrained FANETs.
• The algorithm exhibits excellent performance in dynamic and critical operational environments.
• Future work may involve integrating mobility prediction and energy-aware routing for enhanced FANET scalability.