基于多专家学习的MPTCP调度优化算法  

Multi-expert learning-based MPTCP scheduling optimization algorithm

在线阅读下载全文

作  者:陈成兵 张琳琳 江凌云[1] CHEN Chengbing;ZHANG Linlin;JIANG Lingyun(School of Communications and Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;College of Science,Nanjing University of Posts and Telecommunications,Nanjing 210023,China)

机构地区:[1]南京邮电大学通信与信息工程学院,南京210003 [2]南京邮电大学理学院,南京210023

出  处:《智能计算机与应用》2024年第5期19-26,共8页Intelligent Computer and Applications

基  金:国家自然科学基金重点项目(92067202)。

摘  要:在多径传输控制协议(Multipath TCP,MPTCP)的支持下,多网络接入设备能够聚合网络带宽,提高网络利用率。然而,利用MPTCP协议在多接入异构网络环境中传输数据时,网络环境的不对称以及网络性能(延迟、抖动)之间的差异,必然会导致接收端缓冲区阻塞的问题,这将会造成网络传输性能的下降。针对上述问题,现有的解决方案大都利用前向延迟展开对MPTCP数据包的调度,然而这些对于延迟的计算并不能准确地反映真实的情况。因此,本文提出了一种基于多专家学习前向延迟预测的MPTCP调度优化算法,该算法通过建立的多专家学习模型对前向延迟做出预测,基于预测到的前向延迟,将数据包通过延迟补偿的方式调度到各个路径。仿真结果表明,基于该方法的MPTCP调度优化算法相比于其他方法,有效地降低了接收端缓冲区无序数据包的数量,提高了吞吐量。With the rapid development of modern network technology,multi-network access devices have gradually become the mainstream.With the support of Multipath TCP(MPTCP),these multi-access devices can aggregate network bandwidth and improve network utilization.However,when using MPTCP protocol to transmit data in a multi-access heterogeneous network environment,the asymmetry of the network environment and the differences in network performance(latency,jitter)will inevitably lead to the problem of receiver buffer blocking,which will result in the degradation of network transmission performance.For these problems,most of the existing solutions utilize forward delay to expand the scheduling of MPTCP packets,but these calculations for latency do not accurately reflect the real situation.Therefore,this paper presents an MPTCP scheduling optimization algorithm based on multi-expert learning forward delay prediction,which predicts forward delay by establishing a multi-expert learning model,and schedules packets to each path through delay compensation based on the predicted forward delay.The simulation results show that the MPTCP scheduling optimization algorithm based on this method can effectively reduce the number of unordered packets in the receiver buffer and improve the throughput compared with other methods.

关 键 词:多专家学习 多径传输控制协议 路径管理 数据包调度 前向延迟 

分 类 号:TP18[自动化与计算机技术—控制理论与控制工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象