Open Access

ARTICLE

VHO Algorithm for Heterogeneous Networks of UAV-Hangar Cluster Based on GA Optimization and Edge Computing

Siliang Chen¹, Dongri Shan^2,*, Yansheng Niu³

1 School of Information and Automation Engineering, Qilu University of Technology (Shandong Academy of Sciences), Jinan, 250300, China
2 School of Electronics and Information, Aerospace Information Technology University, Jinan, 250200, China
3 School of Mechanical Engineering, Qilu University of Technology (Shandong Academy of Sciences), Jinan, 250300, China

* Corresponding Author: Dongri Shan. Email:

(This article belongs to the Special Issue: Collaborative Edge Intelligence and Its Emerging Applications)

Computers, Materials & Continua 2025, 85(3), 5263-5286. https://doi.org/10.32604/cmc.2025.067892

Received 15 May 2025; Accepted 04 July 2025; Issue published 23 October 2025

Abstract

With the increasing deployment of Unmanned Aerial Vehicle-Hangar (UAV-H) clusters in dynamic environments such as disaster response and precision agriculture, existing networking schemes often struggle with adaptability to complex scenarios, while traditional Vertical Handoff (VHO) algorithms fail to fully address the unique challenges of UAV-H systems, including high-speed mobility and limited computational resources. To bridge this gap, this paper proposes a heterogeneous network architecture integrating 5th Generation Mobile Communication Technology (5G) cellular networks and self-organizing mesh networks for UAV-H clusters, accompanied by a novel VHO algorithm. The proposed algorithm leverages Multi-Attribute Decision-Making (MADM) theory combined with Genetic Algorithm (GA) optimization, incorporating edge computing to enable real-time decision-making and offload computational tasks efficiently. By constructing a utility function through attribute and weight matrices, the algorithm ensures UAV-H clusters dynamically select the optimal network access with the highest utility value. Simulation results demonstrate that the proposed method reduces network handoff times by 26.13% compared to the Decision Tree VHO (DT-VHO), effectively mitigating the ping-pong effect, and enhancing total system throughput by 19.99% under the same conditions. In terms of handoff delay, it outperforms the Artificial Neural Network VHO (ANN-VHO), significantly improving the Quality of Service (QoS). Finally, real-world hardware platform experiments validate the algorithm’s feasibility and superior performance in practical UAV-H cluster operations. This work provides a robust solution for seamless network connectivity in high-mobility UAV clusters, offering critical support for emerging applications requiring reliable and efficient wireless communication.

---

Keywords

---