Cross-Layer Optimization For QoS-Oriented Resoruce Management In WiMAX

With explosive growth in the data service of Internet and the market of multimedia applications, high-speed and high-quality broadband wireless access is required guarantee the quality of service (QoS) for heterogeneous services in future mobile multimedia communications. However, the mobility effect, unreliable wireless channel condition and limited available wireless spectrum resources are the bottleneck for large traffic transmission. So, based on the understanding about the effects of these factors playing on communication quality and system efficiency, it's badly needed to design an effective resource management mechanism through cross-layer optimization, which can not only ensure heterogeneous QoS requirements, but also balance the tradeoff between the spectrum efficiency and fairness requirements.WiMAX broadband wireless access technology based on IEEE 802.16 standard family has been adopted as the fourth 3G standard by ITU because of its perfect standardization, high-speed data transmission, wide service coverage, fast network deployment, easy system scalability, flexible network construction, etc. In its PHY layer, both WirelessMAN-OFDM and WirelessMAN-OFDMA specifications are combined with adaptive modulation and coding to improve the transmission performance. Meanwhile, a flexible connection-oriented QoS provision framework is defined in its MAC layer. Based on these characteristics, this thesis widely and deeply investigated QoS provision mechanism for the above-mentioned PHY specification under different fading environments through cross-layer optimization. Our major concerns are about connection-level QoS provision through dynamic resource reservation and admission control, as well as packet-level QoS provision through joint resource allocation and packet scheduling. The main contributions of this thesis are listed as follows:(1) We proposed an efficient mechanism to guarantee the connection-level QoS for the services in WirelessMAN-OFDM system. First, considering the influence of user quantity, chanel status, service distribution, various QoS restrictions and resource allocation algorithm played on the system throughput, a reasonable mathematical model is proposed first to characterize the average system capacity and instantaneous capacity, which is the basis for resource reservation and admission control. On one hand, resource reservation is performed considering the average resource consumption and the handoff probability simultaneously, which well ensures the continuity of handoff connections as well as promises high spectrum efficiency. On the other hand, admission control is performed considering both average reousce consumption and instantaneous resource consumption to balance the tradeoff among the connction-level QoS metrics.(2) The uplink transmission performance of WirelessMAN-OFDM system is optimized. First, the term "QoS rate" is defined to reflect the time-variant QoS requirement of the service due to its bursty characteristics. Based on this definition, a joint resource allocation and scheduling algorithm is designed to provide QoS guarantee based on "QoS rate" as well as fairness for the services with identical priority level in case of bandwidth shortage. Specifically, an enhanced bandwidth request mechanism is proposed, which reduces the number of bandwith request messages by aggregating the nrtPS/BE connections in the same subscribe station as one basic bandwidth request unit, as well as replaces the reactive unicast polling and multicast/broadcast polling with proactive unicast polling to reduce the bandwidth request delay and signaling overhead.(3) Based on the analysis of the dynamic resource allocation algorithm for OFDMA system in literature, this thesis discussed how to design dynamic resource allocation algorithm for muticell OFDMA networks based on WirelessMAN-OFDMA specification through MAC-PHY cross-layer optimization, which can effectively take advantage of various diversity technique to improve the system performance. To meet the QoS and fairness requirements of various services, the proposed resource allocation algorimth follows strict priority to process the "QoS rate" requirements of the services with different priority level. Meanwhile, it ensures the fairness among the SSs with different moving speeds through resource preallocation, as well as provides weight fairness and proportional fairness for the services with the same priority level by resource reallocation in case of bandwith shortage. The simulation results showed that proportional fairness provision in resource re-allocation can well balance the tradeoff between resource efficiency and system fairness under the condition that packet-level QoS constraints are well guaranteed.