Parameter Optimization Approach for MPTCP Performance Enhancement

Abstract

This study investigates the issues of energy usage in multipath wireless networks utilizing the Multipath Transport Control Protocol (MPTCP), which allows numerous TCP connections via different pathways.  Due to route heterogeneity, MPTCP consumes more energy.  Previous research provided several techniques to optimizing energy efficiency; however, they focused on individual systems rather than total performance.  We offer a unique stochastic multipath scheduling technique that takes into account fluctuations in data and path capacity.  The scheduling problem is presented as an optimization problem in order to maximize throughput, avoid congestion, and improve stability.  The algorithm is designed to solve problems over several time intervals utilizing drift-based constraints.  Extensive simulations were performed to compare proposed optimized MPTCP to baseline and protocols.  The results show considerable improvements in throughput and end-to-end latency, demonstrating the efficacy of the suggested strategy.