The Design And Development Of Campus Network Intelligent Gateway Optimization Algorithm

Recently, the campus network has developed into a heterogeneous network composed of wired networks, wireless networks, etc. The rapid development of the campus network and network services have greatly improved the efficiency of the teaching and learning in colleges and universities, but also presents the high level requirement to the transmission performance of the campus network.Firstly，according to the characteristics of campus network, NS2-based campus network topology was implemented, based on the analysis of the campus network topology features and network traffic. And then the campus network simulation model was established in NS2by extracting a specific period of the network traffic from the campus network and added it to the campus network links.Secondly, in order to improve the throughput of the network gateway in campus network and overcome the disadvantage that the original intelligent gateway can not be flexible for the characteristics of wired and wireless links, based on the analysis and study of the existing working principles of different kinds of intelligent gateways and the improvement of the original ASA intelligent gateway, an algorithm called SC (Split Control) intelligent gateway that was based on the independent control of the characteristics of wired and wireless links and the related algorithms on the base of TCP/IP protocol in intelligent gateway are implemented.Thirdly, in order to improve of the fairness between wireless network uplink and downlink bandwidth in wireless networks, for the problem that downlink traffic is much greater than the upward flow in wireless network, a optimization algorithm called FOUD(Fairless over upflow and downflow) algorithm was proposed and implemented to guarantee uplink and downlink bandwidth fairness. In view of the influence on the data frame by the attempt limit and the bit error rate of wireless, the algorithm modeling gateways and wireless stations separately by using two- dimensional markov chain and derived the intelligent gateway sent probability under the condition of different ber and different number of upflow and downflow. Then the algorithm adjust the minimum contention window and sent probability of the intelligent gateway to guarantee the fairness.Finally, the intelligent gateway and intelligent algorithm which was purposed in the thesis were applied in the Henan University of Science and Technology campus network model. Large numbers of experiments indicated that,, under different network environment, SC intelligent gateway optimization algorithm can effectively improve the network throughput and FOUD optimization algorithm can greatly improve the fairness of the uplink and downlink bandwidth.