Concurrent multipath transfer（CMT） using stream control transmission protocol（SCTP） multihoming has become an appealing option to increase the throughput and improve the performance of increasingly bandwidth-hungr...Concurrent multipath transfer（CMT） using stream control transmission protocol（SCTP） multihoming has become an appealing option to increase the throughput and improve the performance of increasingly bandwidth-hungry applications.To investigate the rate allocation for applications in CMT,this paper analyzes the capacities of paths shared by competing sources,then proposes the rate allocation model for elastic flows based on the framework of network utility maximization（NUM）.In order to obtain the global optimum of the model,a distributed algorithm is presented which depends only on local available information.Simulation results confirm that the proposed algorithm can achieve the global optimum within reasonable convergence times.展开更多
基金supported by the National Natural Science Foundation of China (60833002)the National Basic Research Program of China (973 Program) (2007CB307100)+2 种基金the National High Technology Research and Development Program of China (863 Program) (2007AA01Z202)the Program for Changjiang Scholars and Innovative Research Team in University (IRT0707)the Program of Introducing Talents of Discipline to Universities (111 Project) (B08002)
文摘Concurrent multipath transfer（CMT） using stream control transmission protocol（SCTP） multihoming has become an appealing option to increase the throughput and improve the performance of increasingly bandwidth-hungry applications.To investigate the rate allocation for applications in CMT,this paper analyzes the capacities of paths shared by competing sources,then proposes the rate allocation model for elastic flows based on the framework of network utility maximization（NUM）.In order to obtain the global optimum of the model,a distributed algorithm is presented which depends only on local available information.Simulation results confirm that the proposed algorithm can achieve the global optimum within reasonable convergence times.
