多飞行器通信网络的抗毁性优化研究

发布时间：2017-12-26 20:22

本文关键词：多飞行器通信网络的抗毁性优化研究　出处：《南昌航空大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着网络信息技术的飞速发展,复杂网络下的多飞行器网络化作战的优势和重要性不断提高。现实中的多飞行器往往处于各种各样的网络环境当中,这些多飞行器通信网络在面对随机失效的情况时,常常表现出较强的抗毁性,而在面对恶意的攻击时,却常常表现出较弱的抗毁性。如何改造现有复杂的多飞行器通信网络和设计优化多飞行器通信网络的抗毁性,使它的抗毁性更高,越来越被人们所关注。本文要解决的问题是如何提高和优化多飞行器通信网络的抗毁性。为此,可以对多飞行器通信网络的抗毁性进行优化设计,建立抗毁性模型,使用度分布熵和平均二步度优化策略。多飞行器通信网络在面对随机失效和恶意攻击时,无标度网络的异构性对它的抗毁性有着非常重要的影响。而多飞行器通信网络是无标度网络,因此可以发现多飞行器通信网络的非均匀性越大,则多飞行器通信网络的抗毁性也越强。通过把优化多飞行器通信网络的抗毁性问题转化成优化多飞行器通信网络的度分布熵和平均二步度的问题。在优化多飞行器通信网络的度分布熵的过程中,通过遗传算法来解决遇到的有约束条件下非线性混合整数规划问题。而在优化多飞行器通信网络的平均二步度的过程中,为了更好地解决非线性混合整数规划问题,采用的是改进的蚁群算法。通过优化仿真可以分别找到最优化度分布熵和最优化平均二步度。另外,还对其他能提高多飞行器通信网络抗毁性的方法进行了探讨,如:隐藏多飞行器通信网络关键的节点;增加多飞行器通信网络关键节点之间存在的链路冗余等。通过计算机仿真和理论推导,能找到度分布熵和平均二步度的最优值,可以发现度分布熵和平均二步度不仅可以作为多飞行器通信网络抗毁性的量度而且还能帮助优化多飞行器通信网络的抗毁性。进一步推导分析可以得到如下结论:度分布熵和平均二步度都能很好地帮助优化多飞行器通信网络的抗毁性。但是,在与度分布熵相比时,平均二步度不仅可以包含节点度分布多样性的信息而且还能包含多飞行器通信网络拓扑信息。因此,在度量多飞行器通信网络异质性方面,平均二步度是一个更好的度量。
[Abstract]:With the rapid development of network information technology, the advantages and importance of multi aircraft network operations under complex networks have been improved. The reality of multi aircraft is often in a variety of network environment. These multi aircraft communication networks often exhibit strong invulnerability in the face of random failure, but often show weak invulnerability when facing malicious attacks. How to transform the existing complex multi aircraft communication network and design and optimize the invulnerability of multi aircraft communication network, making it more invulnerable, has attracted more and more attention. The problem to be solved is how to improve and optimize the survivability of the multi vehicle communication network. For this reason, the destruction resistance of the multi vehicle communication network can be optimized, the model of the destruction resistance is established, the entropy of degree distribution and the average two step optimization strategy can be used. In the face of random and malicious attacks, the heterogeneity of the scale-free networks has a very important impact on its survivability. The multi aircraft communication network is scale-free network, so it can be found that the greater the heterogeneity of multi aircraft communication network is, the stronger the survivability of multi aircraft communication network is. The problem of optimizing the damage resistance of the multi vehicle communication network is transformed into the optimization of the degree distribution entropy and the average two step of the multi vehicle communication network. In the process of optimizing the degree distribution entropy of multi aircraft communication network, genetic algorithm is applied to solve the nonlinear mixed integer programming problem with constraints. In the process of optimizing the average two steps of multi aircraft communication network, in order to solve the nonlinear mixed integer programming problem, the improved ant colony algorithm is adopted. The optimal degree distribution entropy and the optimal average two step degree can be found by the optimization simulation. Besides, we also discuss other ways to improve the invulnerability of multi aircraft communication network, such as hiding key nodes of multi aircraft communication network, increasing link redundancy between key nodes of multi aircraft communication network. Through computer simulation and theoretical derivation, which can find the optimal entropy of degree distribution and average value of two degrees, can find the entropy of degree distribution and average two degrees can be used not only as many aircraft communication network survivability measure but also help invulnerability optimization of multi aircraft communication network. The following conclusions can be obtained as follows: the degree distribution entropy and the average two step can help to optimize the survivability of the multi vehicle communication network. However, when compared with the degree distribution entropy, the average two step degree can not only contain information about the diversity of node degree distribution, but also contain the topology information of multi aircraft communication network. Therefore, the average two step is a better measure in measuring the heterogeneity of the multi vehicle communication network.
【学位授予单位】：南昌航空大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN915.0;O157.5

【相似文献】