Improving congestion control for Concurrent Multipath Transfer through bandwidth estimation based resource pooling

TL;DR

This paper introduces a novel congestion control algorithm for CMT-SCTP that leverages bandwidth estimation for resource pooling, aiming to optimize multi-path data transfer efficiency.

Contribution

It proposes a bandwidth estimation based resource pooling approach to improve congestion control in CMT-SCTP, which is validated through ns-2 simulations.

Findings

Improved congestion window adjustment based on bandwidth estimates

Enhanced resource utilization across multiple paths

Better performance compared to traditional CMT-SCTP

Abstract

Stream Control Transmission Protocol (SCTP) was introduced in 2001 as a multipath variant to traditional transport protocols, i.e. Transmission Control Protocol (TCP) and User Datagram Protocol (UDP). Concurrent Multipath Transfer (CMT) has been proposed as an extension for SCTP to support concurrent usage of available multiple paths. In this paper, we propose a new congestion control algorithm for CMT-SCTP based on the principle of resource pooling. We use the connection bandwidth estimates to obtain the collection of the network resources being used by different flows on multiple paths. Based on these bandwidth estimates, we have used the bandwidth estimation based resource pooling approach to adjust the congestion window of the respective paths. We compare our proposed scheme with CMT-SCTP through ns-2 based simulations.