Analysis of TCP latency over wireless links supporting FEC/ARQ-SR for error recovery

Abstract

Abstract-We study in this paper the performance of TCP over a wireless link implementing hybrid FEC/ARQ-SR at the link layer. The study is done by simulating a large number of TCP transfers over a wireless link showing Bernoulli errors. We are motivated by how to tune link-level error recovery, e.g. amount of FEC, persistency of ARQ, so as to minimize the latency of TCP. We provide results for different physical characteristics of the wireless link (delay, error rate), different traffic loads and different file sizes. Our main finding is that the latency of TCP always improves with the persistency of ARQ, except for some extreme cases where the delay is large, files are small, and the loss rate is low. When adding FEC, the latency of TCP improves then deteriorates, and the deterioration is more pronounced in the case of large files, high loss rate and small delay. Another finding of our study is that with the hybrid mechanism, the wireless link is able to carry more traffic than when FEC and ARQ-SR are separately used.

Real-Time Communications in Large-Scale Wireless Networks

“…The network layer drops overflowed packets by queueing management, which is then detected by the transport layer. Such methods have been known to incur problems in wireless networks [27]. By the real-time unicast module, the paradigm of the classical "network queueing" could be transferred to "queueing in network."…”

Cross-Layer Design to Improve Wireless TCP Performance with Link-Layer Adaptation

“…These drawbacks present the need for careful consideration in the use of link-level ARQ and FEC, based on application type. For example, applications which are sensitive to delay may prefer not to use an ARQ mechanism, and may rely on FEC instead [11]. Hence, the aim of this paper is to find the best reliability approach for the wireless link, in order to enhance end-to-end TCP performance.…”

A Novel Adaptive SR-SW-ARQ/FEC Scheme for an Ultra Low-Latency Mobile Network

“…Let I be the LL-unit size in bits, DR the data rate after encoder, CR the coding rate. For N retransmission attempts, the average transmission latency ATL and the maximum transmission latency MTL of an IP packet are defined based on the transmission latency TL of IP packets N rr(i, j)l TL(i) = E DRXCR(i,j) ATL = meanj (TL (j) MTL = mnaxj (TL(')), j = 1: .. X (5) (6) (7) A summary of simulation parameters is listed in Table I. LL-unit size=150 bytes are depicted in Figures 6 and 7, respectively.…”

“…These two factors, usually studied at the physical or link layer independently, should be considered together for the successful deployment of effective error control strategies. Some researches, focusing on the performance results of three basic ARQ protocols (Selective Repeat ARQ [SR-ARQ], Stop-and-Wait ARQ [SW-ARQ] and Go-Back-N ARQ [GBN-ARQ]), have investigated fading mobile radio channels [1]- [5], moreover, an extended ARQ based on conventional SR-ARQ, in which the retransmission information is combined with the previously received information, has also been proposed [6]. On the other hand, to overcome the discontinuity of throughput of ARQ protocols, many adaptive ARQ protocols have been proposed [7]- [9].…”