使用混合蝙蝠-人工蜂群算法的无线传感器网络节能优化

摘要(中文译文)

本研究提出了一种新颖的混合蝙蝠-人工蜂群（BA-ABC）算法，用于无线传感器网络（WSNs）中的节能优化，以应对节点能量有限和网络生命周期退化的关键挑战。所提出的框架将蝙蝠算法的快速局部收敛性与人工蜂群的鲁棒全局探索能力相结合，实现了聚类和路由过程的统一优化。研究开发了一种自适应多目标适应度函数，以平衡能量消耗、网络生命周期和通信效率，从而在不同网络条件下实现动态、高效的资源利用。在MATLAB R2024a中进行的全面仿真表明，所提出的BA-ABC算法显著优于传统方法和近期优化方法。结果显示，与蚁群优化（ACO）等基线方法相比，总能耗降低了约22-30%，网络生命周期提升了18-25%，延迟减少了近24%。统计验证（包括置信区间和假设检验）证实了所提出框架在多次仿真运行中的鲁棒性、稳定性和一致性。与现有的混合方法和基于机器学习的方法不同，BA-ABC算法在不引入过多计算开销或复杂训练要求的情况下实现了高优化性能，使其适用于资源受限的WSN环境。此外，所提出的方法展现出强大的可扩展性和适应性，使其成为实际应用（包括智慧城市、环境监测和医疗保健系统）的实用解决方案。本研究通过提供一个可扩展、自适应且计算高效的混合框架，与下一代物联网赋能网络的新兴趋势相一致，为智能WSN优化的发展做出了贡献。

Abstract

This study presents a novel hybrid Bat-Artificial Bee Colony (BA-ABC) algorithm for energy-efficient optimization in Wireless Sensor Networks (WSNs), addressing the critical challenge of limited node energy and network lifetime degradation. The proposed framework integrates the rapid local convergence of the Bat Algorithm with the robust global exploration of the Artificial Bee Colony to achieve unified optimization of clustering and routing processes. An adaptive multi-objective fitness function is developed to balance energy consumption, network lifetime, and communication efficiency, enabling dynamic, efficient resource utilization across varying network conditions. Comprehensive simulations conducted in MATLAB R2024a demonstrate that the proposed BA-ABC algorithm significantly outperforms conventional and recent optimization approaches. The results show a reduction in total energy consumption of approximately 22-30%, an improvement in network lifetime of 18-25%, and a latency reduction of nearly 24% compared to baseline methods such as Ant Colony Optimization (ACO). Statistical validation, including confidence intervals and hypothesis testing, confirms the robustness, stability, and consistency of the proposed framework across multiple simulation runs. Unlike existing hybrid and machine-learning-based approaches, the BA-ABC algorithm achieves high optimization performance without introducing excessive computational overhead or complex training requirements, making it suitable for resource-constrained WSN environments. Furthermore, the proposed method demonstrates strong scalability and adaptability, positioning it as a practical solution for real-world applications, including smart cities, environmental monitoring, and healthcare systems. This work contributes to the advancement of intelligent WSN optimization by providing a scalable, adaptive, and computationally efficient hybrid framework aligned with emerging trends in next-generation IoT-enabled networks.

如何引用

Sensors (Basel, Switzerland) 的其他文献

• 基于嵌入式MOX的挥发性传感，用于植物乳饮料的实时分类
• 基于钻石NV中心的频率标准频率锁定方法
• 四柱谐振器耦合MIM波导传感器
• 使用光镊实时探测分子亲和力
• 开发了一种基于神经模糊的上肢康复外骨骼可变导纳控制策略
• 临床剂量学中百分深度剂量分布的测量：传统技术与新兴传感器技术
• 用于体上生物医学应用的紧凑型AMC支持的柔性UHF RFID标签天线
• 从生物标志物到生物传感器：透析护理中的共病管理转型
• 基于表面肌电图的肌肉协同分析与功能驱动比用于机器人辅助上肢康复期间的定量评估
• 柔性压力传感器矩阵的技术特征和生物医学应用：范围综述