Abstract
We present an analytical packet/floow-level modelling approach for the performance analysis of IEEE 802.11e WLAN, where we explicitly take into account QoS differentiation mech- anisms based on minimum contention window size values and Arbitration InterFrame Space (AIFS) values, as included in the Enhanced Distributed Channel Access (EDCA) protocol of the 802.11e standard. We first enhance the packet-level approach previously used for best-effort WLANs to include traffic classes with different QoS requirements. The packet-level model approach yields service weights that discriminate among traffic classes. From these observations, the packet/flow-level model for 802.11e is the generalized discriminatory processor-sharing (GDPS) queueing model where the state-dependent system capacity is distributed among active traffic classes according to state-dependent priority weights. Extensive simulations show that the discriminatory processor-sharing model closely represents the flow behavior of 802.11e.
Original language | English |
---|---|
Title of host publication | Proceedings 19th International Teletraffic Congress (ITC19) |
Subtitle of host publication | Performance Challenges for Efficient Next Generation Networks |
Place of Publication | Beijing, China |
Publisher | Beijing University Post and Telecommunications Press |
Pages | 1651-1662 |
Number of pages | 12 |
Volume | 6B |
Publication status | Published - 2005 |
Event | 19th International Teletraffic Congress 2005: Performance Challenges for Efficient Next Generation Networks - Beijing, China Duration: 29 Aug 2005 → 2 Sep 2005 Conference number: 19 http://www.itc19.com/ |
Conference
Conference | 19th International Teletraffic Congress 2005 |
---|---|
Abbreviated title | ITC 2005 |
Country | China |
City | Beijing |
Period | 29/08/05 → 2/09/05 |
Internet address |
Keywords
- MSC-60K25
- MSC-90B18
- MSC-90B22
- IEEE 802.11e Wireless LAN
- Quality of Service
- EDCA
- Flow-level performance
- File transfer times
- Generalized Discriminatory Processor-Sharing (GDPS)