Distinguishing the Cause of TCP Retransmission Timeouts in Multi-hop Wireless Networks

Abstract

TCP’s performance significantly degrades in multi-hop wireless networks because TCP’s retransmission timeouts (RTOs) are frequently triggered regardless of congestion due to sudden delay, and wireless transmission errors. The RTOs caused by wireless errors lead to TCP’s unnecessary behaviors such as reducing sharply its sending rate and increasing exponentially its back-off value even when the network is not congested. Since TCP has no ability to distinguish the cause of a RTO, it is unavoidable for TCP to underutilize available bandwidth by blindly reducing its sending rate for all the RTOs. In this paper, we propose an algorithm to distinguish the cause of RTOs and to respond to the triggered RTOs according to the cause in order to improve TCP’s performance. When a RTO is triggered, our algorithm estimates the queue usage in the network path during the go back-N retransmissions, and decides if the RTO is triggered due to congestion or not. If any RTO is triggered regardless of congestion, our algorithm prevents TCP from increasing unnecessarily its back-off value as well as reducing needlessly its sending rate. Throughout the extensive simulation scenarios, we evaluate our algorithm in terms of accuracy, misclassification and good put. The experiment results show that the accuracy of our algorithm reaches up to around 90% while its misclassification is less than 20%. Also it is shown that with the high accuracy our scheme improves TCP’s performance the most highly up to 70% in multi-hop wireless networks.