度特征遗传的复杂网络竞争增长模型

Complex Network’s Competitive Growth Model with Degree Inheritance

复杂网络是一种介于规则网络和随机网络之间的网络模型，受生物进化的启发，结合无标度网络的优先连接特性，将进化过程中资源竞争和基因遗传引入到网络的增长过程中，提出一种新的增长模型．仿真实验发现，竞争增长网络模型度分布不再是幂律的，而是服从指数分布．完全遗传竞争增长网络则呈负偏态，且平均度与网络规模取对数后呈线性关系，从数学推导上也证明了这一关系．新模型具有优越的抗恶意攻击能力，且遗传系数越高这种能力越强．另外，网络的平均聚集系数随遗传系数增大而减小，平均路径长度随遗传系数增大而增大．整个模型是拓扑可调的，不同的参数组合可产生具有不同性质的网络模型．

Complex network is a kind of network between regular network and stochastic network. Inspired by biological evolution, we introduced resource competition and genetic inheritance into the growth process of network, and proposed a new growth model with the priority connection of scale-free network. Emulated analysis shows that the network model of competitive growth is no longer power-law, but it obeys exponential distribution. The competitive growth model of degree-characteristic inheritance is negative skew. And it shows a linear relationship between the logarithm of average degree and the network size which is also proved by the mathematical deduction. The new model enhances the ability of resisting malicious attacks. And it is related with the genetic coefficient. In addition, the average clustering coefficient of network decreases with the increase of the genetic coefficient, while the average path length increases with the increase of the inherited coefficient. The whole model is topologically tunable. Different combinations of parameters can produce network models with different properties.