802.11 Rate Adaptation

RRAA

Robust Rate Adaptation Algorithm

Overview - A brief description of the project
Papers - RRAA documents
Talks - Selected presentations
People - The people behind the project

Overview

Rate adaptation is a mechanism unspecified by the 802.11standards, yet critical to the system performance by exploiting the multi-rate capability at the physical layer. In this project, we conduct a systematic and experimental study on rate adaptation over 802.11 wireless networks.

The motivation for a new rate adaptation design comes from the fact that the current rate adaptation designs follow a few conceptually intuitive and seemingly effective guidelines. However, all these guidelines suffer problems in practice.These guidelines are
    - Decrease transmission rate upon severe packet loss
    - Use probe packets to assess possible new rates
    - Use consecutive transmission successes/losses to increase/decrease rate
    - Use PHY metrics like SNR to infer new transmission rate
    - Longterm smoothened operation produces best average performance

By understanding the problems from existing guidelines, we design a new rate adaptation, RRAA, which does not use any of the existing guidelines. RRAA bases on 2 key components:

  • Loss Estimation: It assesses the channel condition by keeping track of the frame loss ratio within a short time window (5 ~ 40 frames in our implementation), and 
  • Adaptive RTS Filter: It selectively turns on RTS/CTS exchange to suppress collision losses, and adapts the number of RTS-protected frames to the collision level.

The goals of RRAA are:

  • Robust against random channel loss, 
  • Responsive to drastic channel changes, and
  • React correctly to collision loss
We implement RRAA on Meru Networks products (AP 100 and AP 200) and conduct extensive experiments to evaluate RRAA performance in different scenarios. Experimental results show that RRAA performs better than existing designs in all cases, and the performance gain can up to 143%.

Papers

  • "Robust Rate Adaptation in 802.11 Wireless Networks". Starsky H.Y. Wong, Hao Yang, Songwu Lu, Vaduvur Bharghavan, In Proc. of ACM MOBICOM, Sep. 2006 [pdf]

Talks

  • "Robust Rate Adaptation in 802.11 Wireless Networks", presented at ACM MOBICOM, Sep. 2006 [ppt]

Project Members

Faculty

Student

Collaborators

WiNG