Enhanced HCCA mechanism for multimedia traffics with QoS support in IEEE 802.11e networks

Capacity enhancement for a rate-variance-envelop-based admission control in IEEE 802.11e HCCA WLANs

“…The performance of IEEE 802.11e HCCA has been addressed in [2,3], and a Markov decision process approach has been considered to improve the system throughput in [4]. Furthermore, several scheduling algorithms have recently been proposed for efficient delivery of variable bit rate (VBR) traffic over the IEEE 802.11e HCCA [5][6][7][8][9][10][11][12][13][14]. In [5], an alternative to IEEE 802.11e HCCA is modeled and analyzed.…”

“…In [10], the authors proposed a VBR scheduler which uses the admission control of the reference scheme in [1] to negotiate a minimum bandwidth; the excess bandwidth higher than the negotiated minimum one would be redirected to the EDCA function. In [11], the authors proposed a beacon based polling scheme to reduce the polling overhead and improve the transmission performance. In [12], a HCCA scheduling algorithm has been proposed for VBR traffic among multiple access points (APs) by leveraging the rate variability and spatial reuse.…”

“…In general, most scheduling algorithms for VBR traffic, such as [7,8,10,11], adopt the admission control of the reference scheme in IEEE 802.11e. Although channel utilization can be improved and/or mean packet loss rate can be reduced by those schemes mentioned above, QoS cannot be guaranteed during admission control.…”

“…In our work, traffic shaping is utilized to ensure that the output of the traffic shaper is upper bounded; in addition, after integrating perflow polling with a D/G/1 queue model, system capacity can be accurately evaluated and QoS can be assured. On the other hand, per-flow polling is performed in [5][6][7][8][9][10][11]15].…”

“…Among them, no multiplexing gain can be obtained in [8][9][10][11]15], because unused bandwidth cannot be recycled. Since a feedback channel is used to notify the QoS scheduler of the backlog of an uplink flow in [5][6][7], bandwidth allocation to a flow can thus be accurate, resulting in no bandwidth waste.…”