Overcoming performance pitfalls in rate-diverse high speed WLANs

  • Authors:
  • Martín Zubeldía;Andrés Ferragut;Fernando Paganini

  • Affiliations:
  • -;-;-

  • Venue:
  • Computer Networks: The International Journal of Computer and Telecommunications Networking
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Recent developments on the IEEE 802.11 family of standards promise significant increases in speed by incorporating multiple enhancements at the physical layer. These high modulation speeds apply to the data portion of the transmitted frames, while headers must remain at lower speeds; this has motivated the use of frame aggregation to increase data payloads in the newer standards. However, this simple method may still utterly fail to deliver the promised speeds, due to a series of cross-layer effects involving the transport and multiple access layers: the downward equalization of throughputs imposed by TCP under physical rate diversity, the excessive impact of the TCP ACK stream, or the unreasonable fraction of access opportunities taken by uplink flows when competing with the more numerous downlink connections. A first contribution of this paper is to demonstrate these impediments and isolate their causes through a series of experiments with the ns3 packet simulator, on the 802.11n and 802.11ac protocol versions. Our analysis leads us to propose a desirable resource allocation for situations of rate-diverse competition, and an architecture for control at the access-point to achieve it. Our implementation is compatible with the standard, involving a combination of known techniques: packet aggregation, multiple queues with TCP-ACK isolation, and control of the MAC contention window. The main contribution here is to provide a practical, comprehensive solution that imposes the desired efficiency and fairness model addressing all the previously indicated limitations. We demonstrate analytically and through extensive simulation that our method is able to provide significant enhancements in performance under a variety of traffic conditions.