On the impact of link layer retransmission schemes on TCP over 4G satellite links

Abstract

SUMMARYWe study the impact of reliability mechanisms introduced at the link layer on the performance of transport protocols in the context of 4G satellite links. Specifically, we design a software module that performs realistic analysis of the network performance, by utilizing real physical layer traces of a 4G satellite service. Based on these traces, our software module produces equivalent link layer traces, as a function of the chosen link layer reliability mechanism. We further utilize the link layer traces within the ns‐2 network simulator to evaluate the impact of link layer schemes on the performance of selected Transmission Control Protocol (TCP) variants. We consider erasure coding, selective‐repeat automatic request (ARQ) and hybrid‐ARQ link layer mechanisms, and TCP Cubic, Compound, Hybla, New Reno and Westwood. We show that, for all target TCP variants, when the throughput of the transport protocol is close to the channel capacity, using the ARQ mechanism is most beneficial for TCP performance improvement. In conditions where the physical channel error rate is high, hybrid‐ARQ results in the best performance for all TCP variants considered, with up to 22% improvements compared to other schemes. Copyright © 2014 John Wiley & Sons, Ltd.

Refining TCP's RTT dependent mechanism by utilizing link retransmission delay measurement in Wireless LAN

“…It also updates the TCP timestamp by subtracting the time spent in the AP. Unfortunately, all aformentioned approaches ignore the need for correcting RTTdependent mechanisms (i.e., loss detection and flow control), in spite of being affected adversely by link retransmissions [34]. TCP data and TCPack) to travel, and it works only with TCP timestamp option.…”

“…They revise RTO computation to house delay variability on account of non-congestion events and accordingly modify retransmission policy at TCP sender. Unfortunately, all aformentioned approaches ignore the need for correcting RTTdependent mechanisms (i.e., loss detection and flow control), in spite of being affected adversely by link retransmissions [34]. TCP schemes like semi-TCP [14], TCP-AP [29], and one proposed in [35] employ hop-by-hop congestion control to minimize impact of end-to-end RTT variations on the efficiency of TCP congestion control mechanisms.…”

Scheduling flows over LEO constellations on LMS channels

“…To cope with the high channel impairments experienced by the LMS channel, reliability mechanisms have been introduced [29,12,24]. One of the most efficient is Hybrid Automatic Repeat reQuest (HARQ), which combines forward error-correction codes and link layer retransmission.…”

Making H-ARQ Suitable for a Mobile TCP Receiver over LEO Satellite Constellations

“…However, LEO constellations are characterized by important delay variations [2,3,4], where propagation impairments are mostly localized on the Land Mobile Satellite (LMS) channel [5]. In particular, as link errors often occur on the last mile and strongly impact on the end-to-end transmission, link and physical layers error correcting codes are deployed to minimize these errors [6,7,8]. However, reliable transport protocols remain necessary to ensure end-to-end reliability but the inadequacy of the interactions between both transport and link layers retransmission schemes may result in high end-to-end latency.…”

“…For instance, reliable sensor data transfer or aeronautical message services might require that 95 % of data messages respect a certain delay threshold [9]. To ensure this kind of service, and to cope with losses that might occur on LMS, several ARQ and Hybrid ARQ schemes have been proposed [10,8]. Basically, these solutions make more robust this last link by adding error correcting, erasure coding, or enhanced Automatic Repeat reQuest (ARQ) schemes.…”