Multiple network coded TCP sessions in disruptive wireless scenarios

Abstract

TCP over wireless networks is challenging due to random losses and DATA-ACK interference. Random linear coding schemes have been proposed to improve TCP robustness against extreme random losses, but a critical issue still remains of DATA-ACK interference. To address this problem, we use inter-flow coding between DATA and ACK at potential intersecting nodes to mitigate self-interference. In addition, a pipelined random linear coding scheme with adaptive redundancy is introduced to overcome high loss rates over unreliable links. The resulting coding scheme, ComboCoding, combines inter-flow and intra-flow coding to provide robust, fair TCP communication in multi-flow disruptive wireless networks. Simulation results show that TCP with ComboCoding delivers higher throughput than other coding options in high loss scenarios. Moreover, we study the behavior of multiple TCP flows intersecting and interfering with each other in the same ad hoc network. The results show that ComboCoding consistently provides better fairness among multiple co-existing TCP sessions when compared with TCP without coding. The main contributions of this paper are: the practical design of random linear coded packets and XOR coded packets on bi-directional streams (DATA and ACK), the adaptive redundancy control scheme, and the multiple coded TCP sessions fairness evaluation. © 2011 IEEE.