Analysis of the IEEE 802.11e EDCA Under Statistical Traffic Juki Wirawan Tantra, Chuan Heng Foh Centre for Multimedia and Network Technology School of Computer Engineering Nanyang Technological University Singapore Email: jw.tantra@ieee.org, aschfoh@ntu.edu.sg Ilenia Tinnirello Dep. of Electrical Engineering Universit` a di Palermo Palermo, Italy Email: ilenia.tinnirello@tti.unipa.it Giuseppe Bianchi Dep. of Electrical Engineering Universit` a degli Studi di Roma Tor Vergata Roma, Italy Email: bianchi@elet.polimi.it Abstract— Many models have been proposed to analyze the performance of the IEEE 802.11 distributed coordination function (DCF) and the IEEE 802.11e enhanced distributed coordination function (EDCA) under saturation condition. To analyze DCF under statistical traffic, Foh and Zukerman introduce a model that uses Markovian Framework to compute the throughput and delay performance. In this paper, we analyze the protocol service time of EDCA mechanism and introduce a model to analyze EDCA under statistical traffic using Markovian Framework. Us- ing this model, we analyze the throughput and delay performance of EDCA mechanism under statistical traffic. I. I NTRODUCTION Widespread usage of high speed wireless local area net- works (WLANs) has generated much interests on quality of service (QoS) support in WLANs. High speed WLANs allow users to run many bandwidth intensive applications without the hassle of wires. Nevertheless, supporting delay sensitive applications is difficult in WLANs as WLANs use unreliable medium. Most of the current WLAN devices im- plement the IEEE 802.11 standard [1], which uses contention based Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) as the main MAC protocol. To support QoS demanding applications, IEEE is working on a new standard, the IEEE 802.11e [2], which specifies differentiated service classes in the MAC layer to support the delivery of priority packets such as multimedia data packets. IEEE 802.11e defines hybrid coordination function (HCF) for access mechanism. HCF itself uses two mechanisms for transmissions, which are enhanced distributed channel access (EDCA) and HCF controlled channel access (HCCA). EDCA is a contention based access mechanism, whereby HCCA is an optional polling based access mechanism. EDCA is basically an extension of the IEEE 802.11 distributed coordination function (DCF) mechanism, which implementes CSMA/CA, to support differentiated service classes. Many models have been proposed to analyze the throughput and delay of IEEE 802.11 DCF mechanism and IEEE 802.11e EDCA mechanism. Bianchi [3] developed an analytical model to compute the saturation throughput of the stations that im- plement DCF mechanism. Using similar model, Chatzimisios et al. [4] derived the delay of packet transmissions. Robinson and Randhawa [5] extended Bianchi’s model to analyze the saturation throughput performance of the EDCA mechanism. Tantra et al. [6] analyzed the throughput and delay perfor- mance of EDCA saturation. Foh and Zukerman [7] proposed a Markovian Framework model to analyze the performance of DCF under statistical traffic. In [8], Patil and Apte uses this Markovian Framework model to find the maximum number of users that can be supported by an Access Point. In this paper, we analyze the protocol service time of EDCA mechanism and extend the model in [7] to analyze the EDCA mechanism. We then verify the model using simulations. Using this model, we analyze the throughput and delay performance of EDCA under statistical traffic. Our model is useful to predict the performance of a WLAN that implements EDCA. It can also be used for network sizing by extending the method in [8]. This paper is organized as follow. Section II provides a brief summary of the IEEE 802.11 DCF and IEEE 802.11e EDCA. In section III, we describe the saturation model that we use for the Markovian Framework. Section IV explains the variance analysis of EDCA that we developed for EDCA protocol analysis and for selection of parameters for our Markovian Framework model. Section V describes the Marko- vian Framework model that we developed to analyze the EDCF mechanism. We discuss and verify the results of our analytical model in section VI. II. MAC PROTOCOLS This section briefly summarizes the operations of the IEEE 802.11 DCF and the IEEE 802.11e EDCA. For a detailed description, readers may refer to [1] and [2]. A. Distributed Coordination Function DCF employs CSMA/CA MAC protocol with binary expo- nential backoff. DCF does not use collision detection function as stations cannot detect collisions by listening to their own transmissions; thus, it employs handshaking method, which makes use of positive acknowledgment. When a station generates a new frame for transmission, it will first monitor the channel activity. If the channel is detected idle for a period of time called DCF interframe space (DIFS), the station can transmit immediately. If the channel is This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the IEEE ICC 2006 proceedings. 1-4244-0355-3/06/$20.00 (c) 2006 IEEE 546