Research On Resource Allocation Algorithm For Heterogeneous Traffic In OFDMA Downlink

With the increasingly demand of QoS and data transmission rate, there will be higher requirements of performance of wireless communication system in the future. And, it has been a hot topic in the study of wireless communication system that how to support users’ QoS requirements as much as possible with limited resources. Unfortunately, only one part of resource used efficiency is concerned in most of the existing researches, while how to balance the energy and spectrum efficiency is rarely involved. However, as more and more importance of energy saving and environmental protection, it should be the focus of attention that how to improve the energy efficiency in future research. Consequently, the resource allocation algorithm in downlink of OFDMA system, which can meet the QoS requirements of heterogeneous traffic, is investigated comprehensively in this thesis.Firstly, the background and status of this topic research are analyzed, and the basic principle of OFDMA, characteristics of the wireless channel, QoS and mathematical model of user’s data are also introduced. Besides, the basic assumption and general mathematical model of the resource allocation in this topic research is detailed, as well as some typical method to evaluate the system performance, which lay the foundation for subsequent discussion.Secondly, the adaptive resource assignment is studied in downlink of OFDMA system. Three instantaneous and one ergodic algorithms under Rate Adaptive (RA) rule are analyzed and compared by numerical simulation. And then, one optimal (Wong) and one near-optimal (MPDF) algorithms under Margin Adaptive (MA) rule are analyzed too. A new near-optimal algorithm calling Max-GNR-First (MGF) of multi-user OFDM is proposed, which is on the basis of the Fast-Optimal-Minimum-Power-Allocation (FOMPA) of single-user OFDM that is also proposed in this thesis. In the same way, MPDF and MGF are compared by numerical simulation.Finally, cross-layer design of adaptive resource allocation in downlink of OFDMA system is considered. To highlight the Queue Status Information (QSI), joint queue-and-channel aware adaptive subcarrier assignment algorithms are studied. And three throughput-optimal algorithms of single-carrier system are concerned, extended to the OFDMA system and compared by numerical simulation. And then, the problem of resource allocation meeting the QoS requirements of users in OFDMA system is discussed. On the basis of transforming the QSI into the minimum guaranteed rate, it is modeled to maximize the energy efficiency for heterogeneous traffics to get the optimal point of the energy efficiency and spectrum efficiency. Because of the limited total power, the MGF algorithm of MA rule aforementioned need to be modified to fulfill the minimum guaranteed rate of all users, and the modified algorithm is called Modified-MGF (MMGF) in this thesis. On the basis of MMGF, MAX C/I, Waterfilling and BSAA algorithms, the algorithm which is called Balancing-Energy-Efficiency and Spectrum-Efficiency (BEESE) is proposed to meet the QoS of heterogeneous traffics for all users in downlink of OFDMA system and verified later by numerical simulation.