异质化带宽分配下的复杂网络数据流负载问题研究

发布时间：2018-05-21 10:41

本文选题：异质化带宽分配 + 负载　；参考：《物理学报》2013年08期

【摘要】：研究了带有连接边传输容量(带宽)约束的复杂网络上如何提升网络数据流负载问题.在网络连接边带宽资源总量固定的条件下,提出了一种异质化带宽分配方案.引入"受控边"概念,通过加入适当比例的"受控边",重新分配带宽资源,并结合具有拥塞感知能力路由策略的数据流量模型,利用带宽分配调节数据流量走向,提高了带宽利用效率,最终使得网络整体的负载能力较带宽匀质化分配时有显著提升.分别在Baraba′si-Albert无标度网络和Watts-Strogtz(WS)小世界网络平台上仿真,发现按照本文的带宽分配方案,WS小世界网络中节点连接边带宽与网络负载有较强的相关性,节点连接边带宽分配最均衡的时候,网络负载能力达到最大.
[Abstract]:In this paper, the problem of how to enhance the load of network data flow on a complex network with connection side transmission capacity (bandwidth) constraints is studied. A heterogeneous bandwidth allocation scheme is proposed under the condition that the total amount of network connection bandwidth resources is fixed. This paper introduces the concept of "controlled edge", redistributes bandwidth resources by adding a proper proportion of "controlled edges", and combines with the data traffic model with congestion sensing routing strategy, adjusts the data traffic direction by bandwidth allocation. The efficiency of bandwidth utilization is improved, and the load capacity of the whole network is improved significantly compared with that of uniform bandwidth allocation. The simulation results on Baraba'si-Albert scale-free network and Watts-Strogtzn Baraba'si-Albert small-world network platform show that there is strong correlation between node connection bandwidth and network load in WS small-world network, according to the bandwidth allocation scheme proposed in this paper. When the bandwidth allocation of the node connection edge is most balanced, the network load capacity reaches the maximum.
【作者单位】：东北大学信息科学与工程学院;中国林业科学研究院木材工业研究所;青岛理工大学经济与贸易学院;
【基金】：国家自然科学基金(批准号:60774097,60774010,60274009) 中央级公益性科研院所基本科研业务费(批准号:CAFINT2009K07)资助的课题~~
【分类号】：O157.5;TP393.06

【相似文献】