Research And Improvement On TCP Optimization In Heterogeneous Networks

With the development of communication technologies,the types of network access devices are increasing,and multi-homed mobile terminals equipped with multiple wireless interfaces have the ability to access different wireless networ-ks simultaneously.Driven by the remarkable advancements in wireless communication systems,mobile video traffic has undergone exponential growth over the past few years.In order to meet the requirements of the 5th Generation of Mobile Communications System(5G),and to cope with the burden of rapid traffic growth,the optimization of Transmission Control Protocol(TCP)has become the research focus in recent years.Concurrent multipath transfer has the advantages of high rate,high throughput,high bandwidth utilization.etc.,so it is more adaptable to the characteristics of 5G heterogeneous networks than single-path transmission.As a classic multipath transmission protocol,Multipath TCP(MPTCP)is the future resear-ch direction of TCP optimization.Under this background,the thesis proposes multipath solutions to solve the problems existing in the MPTCP protocol.The main contributions are described as follows.Firstly,the research status of TCP protocol is investigated in this thesis.the TCP optimization schemes in heterogeneous networks are sorted out,which are mainly classified into single-path TCP optimization and multipath TCP optimization,simultaneously the comparative analysis of various optimization schemes is summarized.Then,this thesis proposes a delay-based multipath congestion control algorithm for the out-of-order problem in multipath transmission.The proposed algorithm optimizes the adjustment mechanism of congestion window by limiting the delay difference among paths and evaluating the network congestion level,which effectively alleviates the packet reordering problem and achieves a better load balancing strategy of multipath transmission.The simulation results show that the proposed algorithm shifts part of traffic from the higher loss paths to the lower ones.significantly improving the throughput.Finally,aiming at the energy-constrained characteristics of mobile terminals,this thesis proposes a multipath congestion control algorithm based on energy efficiency.The proposed algorithm optimizes mobile traffic transmission in heterogeneous networking environment by maximizing the energy efficiency of data transmission and estimating the network bandwidth.The experimental simulation is conducted,indicating;that the proposed algorithm shifts part of traffic from the hisgher energypaths to the lower ones,which greatly improves the energy efficiency and the bandwidth utilization of mobile terminals.