Tham khảo tài liệu 'quality of service and resource allocation in wimaxfig part 3', kỹ thuật - công nghệ, cơ khí - chế tạo máy phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | A Comprehensive Survey on WiMAX Scheduling Approaches 41 Information Module will process the queue size information and update the Scheduling Database Module. The Service Assignment Module retrieves the information from the Scheduling Database Module and generates the UL-MAP. BS broadcasts the UL-MAP to all SSs in the downlink subframe. BS s scheduler transmits packets according to the UL-MAP received from the BS. Authors in Tsu-Chieh et al 2006 present an uplink packet scheduling with call admission control mechanism using the token bucket. Their proposed CAC is based on the estimation of bandwidth usage of each traffic class while the delay requirement of rtPS flows shall be met. Each connection is controlled by token rate ri and bucket size bi. Then they find an appropriate token rate by analyzing Markov Chain state and according to delay requirements of connections. In their Uplink Packet Scheduling Algorithm they adopt Earliest Deadline First EDF mechanism proposed in Kitti Aura 2003 . There is a database that records the number of packets that need to be sent during each frame of every rtPS connection. The disadvantage of this mechanism is that depends on the estimation model that is used. In Yanlei Shiduan 2005 authors propose a hierarchical packet scheduling model for WiMAX uplink by introducing the soft-QoS and hard-QoS concepts as shown in figure 8. The rtPS and nrtPS traffic are classified as soft-QoS because their bandwidth requirement varies between the minimum and maximum bandwidth available for a connection. UGS traffic is classified as hard-QoS. The model is able to distribute bandwidth between BE and other classes of traffic efficiently and guarantees fairness among the QoS-supported traffic UGS rtPS and nrtPS . 42 Quality of Service and Resource Allocation in WiMAX The packet-scheduling algorithm comprises of four parts 1. hard-QoS server scheduling 2. soft-QoS server scheduling 3. best-effort server scheduling 4. Co-scheduling among the above .
