School of Computer, University of South China, Hengyang 421001, China.
School of Electrical Engineering, University of South China, Hengyang 421001, China.
Sensors (Basel). 2023 Jun 5;23(11):5337. doi: 10.3390/s23115337.
In practical wireless sensor networks (WSNs), cascading failures are closely related to network load distribution, which in turn strongly relies on the locations of multiple sink nodes. For such a network, understanding how the multisink placement affects its cascading robustness is essential but still largely missing in the field of complex networks. To this end, this paper puts forward an actual cascading model for WSNs based on the multisink-oriented load distribution characteristics, in which two load redistribution mechanisms (i.e., global routing and local routing) are designed to imitate the most commonly used routing schemes. On this basis, a number of topological parameters are considered to quantify the sinks' locations, and then, the relationship between these quantities with network robustness is investigated on two typical WSN topologies. Moreover, by employing the simulated annealing approach, we find the optimal multisink placement for maximizing network robustness and compare the topological quantities before and after the optimization to validate our findings. The results indicate that for the sake of enhancing the cascading robustness of a WSN, it is better to place its sinks as hubs and decentralize these sinks, which is independent of network structure and routing scheme.
在实际的无线传感器网络（WSN）中，级联故障与网络负载分布密切相关，而负载分布又强烈依赖于多个汇聚节点的位置。对于这样的网络，了解多汇聚节点放置如何影响其级联鲁棒性至关重要，但在复杂网络领域这方面的研究还很缺乏。为此，本文基于面向多汇聚节点的负载分布特性，提出了一种实际的 WSN 级联模型，其中设计了两种负载重新分配机制（即全局路由和局部路由）来模拟最常用的路由方案。在此基础上，考虑了一些拓扑参数来量化汇聚节点的位置，然后研究了这些数量与网络鲁棒性之间的关系，分析了两种典型的 WSN 拓扑。此外，通过采用模拟退火方法，我们找到了最大化网络鲁棒性的最优多汇聚节点放置，并比较了优化前后的拓扑参数，以验证我们的发现。结果表明，为了提高 WSN 的级联鲁棒性，最好将汇聚节点放置为中心节点，并分散这些汇聚节点，这与网络结构和路由方案无关。
