Lyapunov Optimization Based Energy Efficient Congestion Control for MPTCP in HetNets
Huanxi Cui, XuDa Liao, Sin Xu, Bei Liu · 2018
Improving the throughput of communication systems is the ultimate goal of communication research. However, with the diversification and functions of mobile terminals, the energy consumption is also increasing, especially when using multiple interfaces to transmit data simultaneously. In this paper, the research objective is energy-efficient congestion control for concurrent transmission based on multi-path transport control protocol (MPTCP) in heterogeneous networks (HetNets). In order to ensure that the mobile terminal can last for a long time, we consider the energy consumption model at the receiving end. Because the transmission delay is mainly dominated by the transmission delay, the transmission delay is determined by the length of the transmission queue. In order to guarantee the QoS of the user, we must ensure the stability of the transmission queue, reduce the occurrence of data out of order, and reduce the retransmission. Probability, increase effective throughput. In this paper, a model of energy efficiency (EE) and throughput trade-off is established by using Lyapunov optimization method. By this method, we can obtain the split ratio of each link. In order to better complete the splitting, we use the split ratio to directly control the change of the congestion window. In this context, we use the opportunistic linked increases algorithm (OLIA) as an alternative to linked increases algorithm (LIA). The EE factor is obtained by extending the EE rate splitting vector into the window-based congestion control algorithm via a fluid model. Our contributions in this paper are two-fold: i) we introduce queue-stability to guarantee all arrived data leaving the buffer in a finite time; ii) we improve the OLIA to ensure the responsive and non-flappy, besides, introducing a control parameter of the EE factor to ensure the validation of splitting. Finally, we show how to implement the proposed method in the simulation scenario.