Su Guodong, Zhang Boliang
Faculty of Engineering, School of Computer Science, University of Sydney, Sydney, Australia.
Faculty of Applied Sciences, Macao Polytechnic University, Macao, Macau SAR, 999078, China.
Sci Rep. 2025 May 25;15(1):18196. doi: 10.1038/s41598-025-00474-9.
Wireless Sensor Networks (WSNs) are increasingly deployed in critical applications from healthcare to military systems, yet they face significant security challenges due to their resource constraints and distributed nature. This paper addresses two fundamental security problems in WSNs, i.e., vulnerable key distribution and Sybil attacks. Specifically, we propose a synergized security framework, a groundbreaking approach tailored to enhance the security performance of WSNs. First, a bidirectional hash-based key pre-distribution scheme that increases key generation complexity by 43% while maintaining low memory requirements. Second, a multi-trust layered detection mechanism that combines energy consumption pattern analysis with fuzzy logic for Sybil attack detection, achieving a 91.7% detection rate. Finally, comprehensive evaluations using OMNeT++ with networks of 1-500 nodes demonstrate that our framework outperforms existing protocols (LEAP, SPINS, ESK) by 17-32% in network throughput and 12-26% in node connectivity, while maintaining comparable latency to quantum-based methods (QKD, BB84). Real-world validation in a 32-node test environment confirms the framework's practical effectiveness with only 4.3% performance deviation from simulation results. Our framework particularly excels in dynamic networks, maintaining 89% effectiveness even with 15% node mobility rates, significantly advancing the state-of-the-art in WSN security.
无线传感器网络(WSNs)越来越多地部署在从医疗保健到军事系统的关键应用中,但由于其资源限制和分布式性质,它们面临着重大的安全挑战。本文解决了无线传感器网络中的两个基本安全问题,即易受攻击的密钥分发和女巫攻击。具体而言,我们提出了一种协同安全框架,这是一种为提高无线传感器网络安全性能量身定制的开创性方法。首先,一种基于双向哈希的密钥预分发方案,该方案在保持低内存需求的同时,将密钥生成复杂度提高了43%。其次,一种多信任分层检测机制,该机制将能耗模式分析与模糊逻辑相结合用于女巫攻击检测,实现了91.7%的检测率。最后,使用OMNeT++对1至500个节点的网络进行的综合评估表明,我们的框架在网络吞吐量方面比现有协议(LEAP、SPINS、ESK)高出17%至32%,在节点连通性方面高出12%至26%,同时保持与基于量子的方法(QKD、BB84)相当的延迟。在32节点测试环境中的实际验证证实了该框架的实际有效性,与模拟结果的性能偏差仅为4.3%。我们的框架在动态网络中表现尤为出色,即使节点移动率为15%,也能保持89%的有效性,显著推动了无线传感器网络安全领域的技术发展。