Experimental study of link and transport protocols in interference-prone wireless LAN environments

  • Authors:
  • Vijaynarayanan Subramanian;K. K. Ramakrishnan;Shivkumar Kalyanaraman

  • Affiliations:
  • ECSE Dept, RPI;AT&T Labs;IBM India Research Labs

  • Venue:
  • COMSNETS'09 Proceedings of the First international conference on COMmunication Systems And NETworks
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

As wireless networks are deployed widely and user expectations grow, it is important to study the trade-offs among goodput, latency and loss rates that exist in link and transport protocols. Under low residual loss rates, retransmissions can be effective in bringing down the loss rate. However, at high loss rates, current links trade-off goodput and low latency for a small loss rate which diminishes overall performance. We look at the performance at the link and transport layers in the presence of raw loss and see how current protocols make an unattractive trade-off at the link layer. We also show high ARQ on the link can cause high per-packet latency on the link which can lead to interactions with TCP mechanisms leading to spurious retransmissions and timeouts and reduced TCP-layer goodput. Our measurements results, based on experiments conducted on the ORBIT testbed, show the need for modifications to the link layer to obtain a favorable three-way trade-off. In prior work, we have developed link and transport protocols that are designed to work under high loss rates. Our link design LL-HARQ (hybrid FEC/ARQ) is designed to minimize the link delay while exporting a very small residual loss rate and high goodput while our transport protocol Loss-Tolerant TCP is designed to operate over a wide range of residual loss rates. We provide insights about the balance between error-protection functions at the two layers and examine the case for cross-layer co-operation. The measurement traces were used as inputs to our simulations on the ns-2 simulator. Our results show the favorable trade-offs among goodput/latency/loss rate obtained with our approach over traditional approaches.