3rd International ICST Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities (TridentCom)

Research Article

Introducing Finer Prioritization in EDCA using Random AIFSN

Download573 downloads
  • @INPROCEEDINGS{10.1109/TRIDENTCOM.2007.4444699,
        author={Sudhanshu Gaur and Todor Cooklev},
        title={Introducing Finer Prioritization in EDCA using Random AIFSN},
        proceedings={3rd International ICST Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities (TridentCom)},
        proceedings_a={TRIDENTCOM},
        year={2010},
        month={5},
        keywords={WLAN MAC IEEE 802.11e EDCA AIFSN},
        doi={10.1109/TRIDENTCOM.2007.4444699}
    }
    
  • Sudhanshu Gaur
    Todor Cooklev
    Year: 2010
    Introducing Finer Prioritization in EDCA using Random AIFSN
    TRIDENTCOM
    IEEE
    DOI: 10.1109/TRIDENTCOM.2007.4444699
Sudhanshu Gaur1,*, Todor Cooklev2,*
  • 1: Wireless Systems Research Lab Hitachi America, Limited San Jose, CA 95135, USA
  • 2: School of Engineering San Francisco State University San Francisco, CA 94132, USA
*Contact email: sudhanshu.gaur@hal.hitachi.com, tcooklev@sfsu.edu

Abstract

The recently adopted WLAN standard IEEE 802.11e provides quality of service (QoS) differentiation by grouping the traffic into different access categories (ACs) with different priorities. It defines different arbitration inter-frame space number (AIFSN) on a per-category basis to ensure that each category has different probability of accessing the channel. Within an AC, random back-off mechanism is used to spread the traffic. However, when there are a large number of QoS stations (QSTAs) within an AC, the probability of two or more stations choosing the identical back-off values is increased leading to increased packet collisions. In this paper, we propose the use of randomized AIFSN. The random AIFSN method can be exploited to introduce finer priority values within an AC. Each AC is assigned an AIFSN interval and a probability distribution function (PDF) defined over that interval. The simulation results indicate that the proposed method leads to reduced probability for collisions and higher throughput. We also present some testbed results to verify the improvement using random AIFSN method.