无标度网络弹性的熵优化  

Entropy optimization on scale-free networks' resilience

作　　者：曾宇[1,2] 肖人彬[1] 

机构地区：[1]华中科技大学系统工程研究所,湖北武汉430074 [2]湖北工业大学理学院,湖北武汉430068

出　　处：《系统工程学报》2013年第2期143-150,共8页Journal of Systems Engineering

基　　金：国家自然科学基金资助项目(60974076)

摘　　要：针对大规模且幂律指数位于2和3之间的无标度网络,推导给出了网络结构熵函数,进而通过与网络总承受阈值的对比分析,提出了一种网络结构熵优化模型,最后针对该模型进行了仿真实验并对实验结果做了分析.理论研究和仿真实验结果表明网络结构熵函数是度量这种无标度网络综合弹性的一个有效的新方法,同时考虑随机失效和选择性攻击时的综合弹性增强问题可以转化为熵函数最小的优化问题.For large scale-free networks with power-law indices between 2 and 3, this paper firstly deduces the network structure entropy function. Then, after taking a comparison analysis of that and the total threshold, this paper proposes a network structure entropy optimization model. Finally, some simulation experiments and result analyses are provided to the entropy model. Both theoretical analysis and numerical simulations indicate that the entropy function is a simple and effective new method to measure the overall resilience of this kind of scale-free networks, and the problem of enhancing the overall resilience of networks to both random failure and intentional attack can be converted into an optimization problem when the network structure entropy attains its minimum value.

关 键 词：无标度网络 弹性 网络结构熵函数 总承受阈值 

分 类 号：TP393[自动化与计算机技术—计算机应用技术] N94[自动化与计算机技术—计算机科学与技术]