基于局域世界的无线传感器网络分簇演化模型

当前无线传感器网络无标度演化模型研究往往将网络视为同质网络，且未充分考虑网络在真实情形下的演化特征，导致所获网络拓扑与实际网络差异明显。因此，基于局域世界理论，考虑无线传感器网络典型分簇结构、能耗敏感与真实网络中普遍存在的节点与链路退出的动态性行为等特征，提出无线传感器网络分簇演化模型。该模型与同类研究成果相比，更为接近真实网络情形。利用平均场理论推导出此模型具有无标度特征。通过研究拓扑生长对网络容错性能影响，发现扩大局域世界规模与提升饱和度约束上限可有效提升网络容错性能。与之相反，簇头比例与节点删除概率的上升将导致网络容错性能的下降。

Evolutionary model of heterogeneous clustering wireless sensor networks based on local world theory

Current research on the scale-free evolutionary model of wireless sensor networks (WSN) treated each network as a homogenous one and does not take into account the evolutionary characteristics of the network in realistic scenarios, thus leading to significant differences between homogenous networks and realistic ones. Therefore, based on local-world theory, a heterogonous evolution model of WSN in relation to cluster-structure, energy-sensitivity and dynamic behavior of WSN (e.g., failures of nodes and links) was proposed. Compared with previous research results, the proposed model is closer to reality. By using the mean-field theory, the scale-free feature is demonstrated. Moreover, by studying the impacts of topol-ogy growth on error tolerance of the network, it is proved that enlarging the size of the local-world and enhancing the upper limit of degree is able to improve the network performance in terms of error-tolerance, while an increase in the cluster proportion and deletion probability led to the degradation of the network error tolerance.