Scholarlink

<p>The increasing use of Unmanned Aerial Vehicles (UAVs) in mission-critical operations has intensified the need for efficient, scalable, and adaptive clustering in Flying Ad Hoc Networks (FANETs). This paper presents a clustering optimization framework based on the Secretary Bird Optimization Algorithm (SBOA), a bio-inspired metaheuristic that simulates the strategic hunting behavior of the secretary bird. Compared to the Fire Hawk Optimization Algorithm (FHOA), the Portia Spider Optimization Algorithm (PSOA), and multi-objective optimization algorithm based on sperm fertilization procedure (MOSFP), SBOA results in a balanced exploration–exploitation trade-off that facilitates dynamic and energy-efficient cluster head (CH) selection in high-mobility 3D UAV networks. It is modeled as a multi-objective optimization problem with the aim of minimizing intra-cluster distance, maximizing residual energy, and load balancing. The population of UAVs ranges from 30 to 160 nodes, the communication range from 100 to 900&#xa0;m, and the 3D grid scale. It emerges that SBOA outperforms all its counterparts in terms of up to 15% higher optimization fitness, 10% higher cluster density, and 40% reduced load imbalance. SBOA’s superiority with respect to convergence stability, cluster uniformity, and CH workload distribution is further validated using several visualization tools like heatmaps, t-SNE projections, and statistical plots. SBOA has also been able to achieve over 85% optimal fitness even in highly sparse environments to establish its scalability and robustness. Statistical validation confirmed that SBOA significantly outperformed FHOA, PSOA, and MOSFP to achieve as high as 0.15 higher fitness, with <i>p</i> &lt; 0.001, reduced the convergence time by almost four frames, with <i>p</i> = 0.003, provided 40% lower load imbalance, with <i>p</i> &lt; 0.001, with consistently tighter cluster stability distributions to validate its robustness for large-scale real-time FANET deployments. Such findings make SBOA a viable and high-performance clustering solution for next-generation, real-time, energy-constrained FANET deployments in critical and dynamic environments. SBOA may be extended to incorporate mobility prediction and energy-aware routing to enhance real-time scalability in larger and more dynamic FANET scenarios using a hybrid approach.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

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

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月8日

查看全文
  • Zaheer Aslam,
  • Taj Rahman,
  • Ghassan Husnain,
  • Atiq Ur Rehman,
  • Anandhavalli Muniasamy,
  • Hend Khalid Alkahtani

摘要

The increasing use of Unmanned Aerial Vehicles (UAVs) in mission-critical operations has intensified the need for efficient, scalable, and adaptive clustering in Flying Ad Hoc Networks (FANETs). This paper presents a clustering optimization framework based on the Secretary Bird Optimization Algorithm (SBOA), a bio-inspired metaheuristic that simulates the strategic hunting behavior of the secretary bird. Compared to the Fire Hawk Optimization Algorithm (FHOA), the Portia Spider Optimization Algorithm (PSOA), and multi-objective optimization algorithm based on sperm fertilization procedure (MOSFP), SBOA results in a balanced exploration–exploitation trade-off that facilitates dynamic and energy-efficient cluster head (CH) selection in high-mobility 3D UAV networks. It is modeled as a multi-objective optimization problem with the aim of minimizing intra-cluster distance, maximizing residual energy, and load balancing. The population of UAVs ranges from 30 to 160 nodes, the communication range from 100 to 900 m, and the 3D grid scale. It emerges that SBOA outperforms all its counterparts in terms of up to 15% higher optimization fitness, 10% higher cluster density, and 40% reduced load imbalance. SBOA’s superiority with respect to convergence stability, cluster uniformity, and CH workload distribution is further validated using several visualization tools like heatmaps, t-SNE projections, and statistical plots. SBOA has also been able to achieve over 85% optimal fitness even in highly sparse environments to establish its scalability and robustness. Statistical validation confirmed that SBOA significantly outperformed FHOA, PSOA, and MOSFP to achieve as high as 0.15 higher fitness, with p < 0.001, reduced the convergence time by almost four frames, with p = 0.003, provided 40% lower load imbalance, with p < 0.001, with consistently tighter cluster stability distributions to validate its robustness for large-scale real-time FANET deployments. Such findings make SBOA a viable and high-performance clustering solution for next-generation, real-time, energy-constrained FANET deployments in critical and dynamic environments. SBOA may be extended to incorporate mobility prediction and energy-aware routing to enhance real-time scalability in larger and more dynamic FANET scenarios using a hybrid approach.