Abstract:
With the increasing popularity of using Wireless Local Area Networks (WLANs) in a sense of cost-effectiveness and easiness in deployment, the controlled channel access mechanism in IEEE 802.11e WLANs, i.e., Hybrid Coordination Function Controlled Channel Access (HCCA) has received much more attention. Its inherent centralized mechanism is more efficient to enhance the Quality of Service (QoS) provision for time-bounded services in the current IEEE 802.11 Medium Access Control (MAC) protocol. In HCCA, the station is allocated a fixed Transmission Opportunity (TXOP) based on its Traffic Specification (TSPEC) parameters so that it is efficient for Constant Bit Rate (CBR) streams. However, it does not efficiently cope with the fluctuation of the Variable Bit Rate (VBR) video streams since its reference scheduler generates a schedule based on mean characteristics of the traffic. Provide scheduling according to these mean characteristics is not always accurate because of traffics irregularity nature over time. In this paper, we propose a new scheduling algorithm called fair scheduling based HCCA for video stream that supports QoS over WLAN. The proposed scheduler guarantees QoS by using adaptive service intervals, transmission opportunities, polling order, the deadline of packet and packet loss statistics. The service schedule is based not only on TSPEC parameters but also on the packet size, next packet arrival time and deadline constraints, which efficiently schedule the TS based on their actual needs. Our research work aims to achieve fair scheduling among VBR video packets with deadline constraints over WLAN and to maximize the system throughput. Simulation of the proposed mechanism has been examined on several publicly available video traces to show efficiency using network simulator Optical Modeling Network using C++ programming Language (OMNET++). Finally, the simulation results exhibit the efficiency of the proposed mechanism in providing less delay with similar throughput of reference scheduler and minimizing the number of Null-Frames caused by wasted polls.
