Visit for more related articles at International Journal of Advancements in Technology
In this research article, we have proposed an analytical model for the performance analysis of the enhanced distributed channel access (EDCA) protocol of IEEE 802.11e wireless local area networks standard, using four-dimensional Markov chain. The contemporary EDCA models support only a small subset of EDCA features with limited accuracy. Our model accurately covers all salient features of standard EDCA, like multiple numbers of simultaneously active access categories (AC) per station, internal collision handling, post-back-off after successful transmission and frame-discarding after maximum retransmission limit. The proposed model has also implemented pre-back-off carrier sensing mechanism and back-off counter freezing with deferred states in both the cases. It has incorporated different carrier sensing and back-off parameters for each active AC for access category wise service differentiation. We have computed the saturation throughput and frame access delay of each access category for both RTS/CTS and basic modes. Analytical model is validated by simulation.
Access Category (AC), Arbitration Inter-Frame Space (AIFS), Enhanced Distributed Channel Access (EDCA), Quality of Service (QoS), Short Interframe Space (SIFS).
The enhanced distributed channel access (EDCA) protocol of the emerging IEEE 802.11e standard  supports access category wise quality of service (QoS) differentiation between the real-time and the non real-time applications. In the last few years, performance analysis of EDCA has attracted the attention of several researchers.
The contemporary research articles - on EDCA are all based on simulations. The analytical models - of EDCA have considered only one priority class active AC or flow per station. But the IEEE 802.11e EDCA standard has suggested four simultaneously active access categories (ACs) per station with internal collision handling feature. Also, the models , -, -, - have not implemented the back-off counter freezing i.e. not decrementing the back-off counter during the channel sensing of pre-transmission back-off process, if the channel goes busy due to the transmissions of other access categories. The model  has not considered frame discarding after successful transmission, which reduces excessive frame access delay.
The model proposed by Tao and Panwar , Tantra et al. [ 23 ], Foh et al. [ 24 ] and Hwang et al.  have implemented only two priority class ACs per station. These models - have not considered back-off counter freezing for the higher priority class AC. Also, during the freezing process, the model  for the higher priority class AC and the models - for the lower priority class AC, have not incorporated the appropriate deferred states during the busy channel. Also, no deferred states are implemented during physical carrier sensing, which follows the back-off process.
None of the models - have implemented pre-back-off carrier sensing for any access category with the deferred states. The model proposed by Kong et al.  has also implemented only for two ACs per station. For the purpose of pre-back-off carrier sensing and channel sensing during the pre-transmission back-off process, it has considered the absolute busy probability of the channel, not the probability of sensing the channel busy by the target access category. Also, for the pre-back-off carrier sensing and back-off counter freezing, the model has considered arbitrary values for AIFS and total deferred time for all ACs.
Also, the models – have not incorporated the standard post back-off after successful transmission which reduces the starvation of lower ACs.
Scope of our proposed model
The aforesaid models are less accurate and inefficient due to their inability to capture the real situations. The differences between the above models and the model implemented by us are listed as follows:
(i) We have incorporated multiple number of simultaneously active ACs per station which is theoretically unlimited in our solution framework. (ii) We have implemented the internal collision handling feature with the extensive study of its effect on the performance of EDCA. (iii) We have incorporated pre-back-off carrier sensing for continuous AIFS[i] duration with deferred states for the ith access category ACi(iv) We have implemented the channel sensing and back-off counter freezing with deferred states during pre-transmission back-off process, followed by carrier sensing similar to case (iii). (v) We have considered the probability of sensing the channel busy (Pisb as per equation 14) by the access category ACi, for both the carrier sensing and back-off counter freezing as mentioned in (iii) and (iv), instead of considering the absolute busy probability of the channel. Also, for both the above cases, we have considered the exact AIFSii value of each ACi, as specified by the IEEE 802.11e standard, instead of choosing an arbitrary value. Again for both cases, we have computed the exact value of the total deferred time Tiof the target ACi in the deferred states, as per computation in equation (20), instead of a choosing an arbitrary value. This computation involves Ti as a function of the successful transmission probabilities of other ACs (Pj,su) and the collision probability (Pcl) which are the equation variables. (vi) We have incorporated the standard post back-off after successful transmission to reduce the starvation of lower ACs. (vii) Frame discarding after retry limit which reduces excess frame access delay, is also considered by us.
The implementation of the above mentioned features in our model has made it very accurate, fair and efficient by taking care of the real situations.
The rest of this paper is organized as follows. The proposed analytical model, performance analysis and validation of model are discussed in section 2, 3 and 4 respectively. Finally the conclusion is drawn in section 5.
Markov chain formulation
The state transition probabilities
Saturation delay computation
For the computation of saturation delay, we have followed the model  with substantial modifications.
In the analytical model of this research article, we have studied the performance features of EDCA mode of operation of WLAN based on IEEE 802.11e standard . From the analytical and simulation results discussed in section 4, we draw the following conclusions:
1. Analytical and simulation results match well for both normalized saturation throughput and saturation delay and validate our model.
2. The throughput and delay pattern of standard EDCA shows the access category wise QoS differentiation feature, with higher ACs having higher throughput and lower delay. This shows that EDCA is suitable for soft real-time application, when the latter is being run through higher ACs.
3. RTS/CTS mode is better than the basic mode for both throughput and delay because of less collision loss.
4. Contention window and AIFSN variation is very effective means of service differentiation between the ACs.
5. Internal collision handler has dominating effect on system performance enhancement for higher ACs at lower load.
In this research article, our key research contributions are:
(i) Implementation of pre-back-off carrier sensing with actual deferred states.
(ii) Implementation of back-off counter freezing during the channel sensing of pre-transmission back-off process with actual deferred states followed by carrier sensing with deferred states.
(iii) The contemporary EDCA models have implemented only one or at best two priority class active ACs per station whereas we have incorporated multiple number of simultaneously active ACs within each station (theoretically unlimited) denoted by acm, in the same solution framework.
 Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications, IEEE Std 802.11™-2007-Revision of IEEE Std 802.11-1999.
 Mangold S, Choi S, May P, Klein O, Hiertz G, and Stibor L, “IEEE 802.11e Wireless LAN for Quality of Service”, In Proc. European Wireless, Florence, Italy, Feb.2002, pp. 32-39.
 Choi S, “Emerging IEEE 802.11e WLAN for Quality-of-Service (QoS) provisioning,” SK Telecomm. Rev., Vol. 12, no.6, 2002, pp. 894–906.
 Grilo A, and Nunes M, “Performance evaluation of IEEE 802.11e,” In Proc. PIMRC, Sep.2002, pp. 511–517.
 Choi S, Prado J D, Shankar S and Mangold S, “IEEE 802.11e contention-based channel access (EDCF) performance evaluation,” In Proc. IEEE ICC, vol. 2, May 2003, pp. 1151–1156.
 He D and Shen C Q, “Simulation study of IEEE 802.11e EDCF,” In Proc. 57th IEEE Semiannu. VTC, Apr. 2003, pp. 685–689.
 Mangold S, Choi S, Hiertz G R, Klein O and Walke B, “Analysis of IEEE 802.11e for QoS support in wireless LANs,” IEEE Wireless Communications, December 2003, pp.40-50.
 Lindgren A, Almquist A, and Schelen O, “Quality of service schemes for IEEE 802.11 wireless LANs—an evaluation,” Mobile Networks, Vol. 8, 2003, pp. 223–235.
 Ni Q, “Performance analysis and enhancements for IEEE 802.11e wireless networks” IEEE Network, July/August 2005, pp.21-27.
 Xiao Y, “IEEE 802.11E: QoS provisioning at the MAC layer”, IEEE Wireless Communications, June 2004, pp.72-79
 Robinson J W, Randhawa TS, “Saturation throughput analysis of IEEE 802.11e enhanced distributed coordination function”, IEEE Journal of Selected Areas in Communication, Vol. 22, no. 5, pp. 917–928.
 Xiao Y, “Performance analysis of priority schemes for IEEE 802.11 and IEEE 802.11e wireless LANs”, IEEE Transactions in Wireless Communications ,Vol. 4, no. 4, 2005, pp. 1506–1515.
 Zhu H, Chlamtac I, “Performance analysis for IEEE 802.11e EDCF service differentiation”, IEEE Transactions on Wireless Communications, Vol.4, no.4, 2005, pp. 1779–1788.
 Hui J and Devetsikiotis M, “A unified model for the performance analysis of IEEE 802.11e EDCA”, IEEE Transactions in Communications, Vol.53, no. 9, 2005, pp. 1498–1510.
 Hwang G H, and Cho D H, “Performance analysis on coexistence of EDCA and legacy DCF stations in IEEE 802.11 wireless LANs”, IEEE Transactions on Wireless Communications, Vol.5, no.12, 2006, pp.3355-3359.
 Chen X, Zhai H, Tian X, and Fang Y, “Supporting QoS in IEEE 802.11e wireless LANS”, IEEE Transactions on Wireless Communications, Vol.5, no. 8, 2006, pp. 2217-2227.
 Tantra J W, Foh C H, Tinnirello I and Bianchi G, “Analysis of the IEEE 802.11e EDCA under statistical traffic,” In Proc. IEEE ICC ’06, June 2006, pp. 546-551.
 Zhao L, Wu J Y, Zhang H and Zhang, J “Integrated quality-of-service differentiation over IEEE802.11 wireless LANs”, IET Communications, Vol.2, no.2, 2008, pp. 329–335.
 Gao D, Cai J, Foh C H , Lau C and Ngan K N, “Improving WLAN VoIP capacity through service differentiation”, IEEE Transactions on Vehicular Technology, Vol. 57, no.1, 2008, pp. 465-474.
 Inan I, Keceli F and Ayanoglu E, “Analysis of the 802.11e enhanced distributed channel access function”, IEEE Transactions on Communications, Vol. 57, no.6 , 2009, pp. 1753-1764.
 Peng F, Peng B and Qian D, “Performance analysis of IEEE 802.11e enhanced distributed channel access”, IET Communications, Vol.4, no.6, 2010, pp. 728–738.
 Tao Z, and Panwar S, “Throughput and delay analysis for the IEEE 802.11e enhanced distributed channel access”, IEEE Transactions on Communications, Vol.54, no. 4, 2006, pp. 596–603.
 Tantra J W, Foh C H and Mnaouer A B, “Throughput and delay analysis of the IEEE 802.11e EDCA saturation”, In Proc. IEEE International Conference on Communications (ICC), May 2005, pp. 3450-3454.
 Foh C H, Zhang Y, Ni Z, Cai J, and Ngan K N, “Optimised cross-layer design for scalable video transmission over the IEEE 802.11e networks”, IEEE Transactions on Circuits and Systems for video Technology, Vol. X, no. XX XXX, 2007, pp. 1-14.
 Hwang H Y, Kim S J , Sung D K, Song N O, “Performance analysis of IEEE 802.11e EDCA with a virtual collision handler” , IEEE Transactions on Vehicular Technology, Vol. 57, no.2, 2008, pp. 1293-1297.
 Kong Z , Tsang D H K, Bensaou B and Gao D, “Performance analysis of the IEEE 802.11e contention-based channel access,” IEEE Journal of Selected Areas in Communications, Vol. 22, no. 10, 2004, pp. 2095–2106.
 The Network Simulator2—NS2: http://www.isi.edu/nsnam/ns.