The Internet was originally used to share resources through address addressing.Nowadays,with the rapid development of internet technology and the diversification of network services,users pay more attention to the data content.The traditional network architecture based on TCP/ IP has been unable to meet the needs of users gradually.Named data networking(NDN)is a new internet architecture based on data content name addressing.It transforms network semantics from packet delivery to data content retrieval,and shows good performance in the fields of content distribution,security,mobile support and application development.In any network system,there is the possibility of network congestion,and a good congestion control method can improve the network congestion effectively,so as to improve the network transmission performance,reduce the packet loss rate and improve the network throughput.In the named data networking architecture,data can be obtained from multiple content sources(i.e.producers or routing nodes)through multiple paths,which makes the traditional congestion control method based on end-to-end(i.e.TCP / IP)invalid.Therefore,congestion control has become one of the concerns of the named data networking architecture design,and it is the subject of this thesis too.Based on the explicit congestion notification mechanism and intelligent forwarding characteristics of NDN,this paper studies the explicit congestion control of single bottleneck link and multi-path aware congestion control.The main contents are as follows: This paper expounds the architecture and basic theory of NDN,including data content naming mechanism,data packet type and structure,data content forwarding process and strategy;The existing research work of congestion control in NDN is analyzed and summarized.An explicit congestion control method based on real-time adjusting the sending rate of interest packets is proposed.When the network is congested,this method uses the counter and shaping queue to limit the data flow with too fast rate,encapsulates the congestion information in the extended field of interest packet,and then forwards the interest packet according to the hop by hop forwarding characteristics of NDN;The congestion information will be copied into the data packet and returned to the consumer,so as to adjust the sending rate of interest packet accordingly.At the same time,this method divides the data stream into different priorities according to the different acceptance of the performance of the data stream to the changes of the allocated bandwidth resources,so as to ensure the quality of service of different data streams.The experimental results based on ndnsim2.0 show that this method can improve the throughput of the network on the premise of reducing the packet loss rate.A rate based multipath aware congestion control method is proposed.There has been work in NDN to realize congestion control by limiting the sending rate of interest packets,but most of the work ignores the intelligent forwarding characteristics in NDN.This method will combine the multipath forwarding characteristics of rate control protocol(RCP)and NDN to design,implement and evaluate a rate based multipath congestion control method.At the same time,this method calculates the transmission rate of each data stream and feeds it back to consumer,so as to control the sending rate of interest packet accordingly.The experimental results based on ndnsim2.0 show that compared with other congestion control methods,this method can ensure fairness and obtain higher throughput.According to the available bandwidth of the path,this method can also achieve the best traffic segmentation rate,so as to perform better in load balancing. |