Title:Adaptive Quality of Service Provisioning in IEEE 802.11 Wireless Local Access Network
VOLUME: 2 ISSUE: 2
Author(s):Chunming Liu, Chi Zhou, Niki Pissinou and Kia Makki
Affiliation:T-Mobile USA, 12920 Southeast 38th Street, Bellevue, WA 98006, USA.
Keywords:Call admission control, scheduling, IEEE 802.11e, packet loss ratio, Quality of Service, real-time, throughput,
packet delay, wireless LAN.
Abstract:–This paper provides a cross-layer Quality of Service (QoS) provisioning scheme for IEEE 802.11 Wireless
Local Area Network (WLAN). Specifically, we develop a new scheduling algorithm in link layer and a call admission
control mechanism in network layer to support multimedia services. Instead of using a single coordination function in link
layer, we adopt polling scheme to support real-time (RT) services while using EDCF contention-based function for nonreal-
time (NRT) services, based on wireless channel condition and service fairness. The proposed polling algorithms
search idle system time and use it to reduce average packet loss of RT variable-bit-rate (VBR) services, as well as
satisfying transmission delay and bandwidth requirements of all RT services. The admission control mechanism provides
long-term quality for both RT and NRT services by ensuring low average packet loss ratio for RT services and acceptable
minimum throughput for NRT services. Service priority and fairness are achieved by assigning VBR bursts higher priority
for resource allocation based on their weighting values. Fairness is also implemented by reserving resource for NRT to
avoid possible starvation. Rate allocation for VBR connections are adaptively adjusted with sliding window measurement.
Detailed analysis is given by taking into account several key statistical factors, such as service traffic model, the ratio
between the polling period and the contention period (CP), burst generation probabilities, and wireless propagation model.
Performance is evaluated through OPNET simulation. Both numerical and simulation results demonstrate that our
proposed joint scheme is effective in terms of both system resource utilization and QoS provisioning.
