Research On Resource Allocation And Scheduling Strategies In Future Mobile Communication Systems

With fast development of mobile broadband networks and rapid popularization of intelligent mobile terminals, mobile networks have invaded into every corner of social life of human. Meanwhile, users have further expectations for mobile communications. Demands for diversity application, quality of service and business experience are growing. However, because resources of future mobile communication systems are limited and various wireless access technology and multimedia data services are integrated in the systems, how to allocate and scheduling scare wireless resources reasonably and effectively in complicated and changeable communication environment so as to satisfy people’s growing needs of services has become the crucial issue to be solved in future mobile communication area.In order to improve the operational capacity of the mobile communication networks and solve the contradiction between the growing demand of multimedia data services and limited wireless resources, unstable wireless links and complicated wireless environment, resource allocation and scheduling are studied in depth in this thesis. With swarm intelligence optimization, convex optimization and game theory techniques, we propose multi-traffic QoS guaranteed cross-layer resource scheduling strategy, MOS-driven energy efficient cross-layer resource allocation strategy, potential game theory-based resource allocation strategy and utility fair-based resource allocation strategy in single-cell OFDMA system, multi-cell coordination network and two-tier Femtocell network. Specific research results and innovations are as follows:1. For single-cell OFDMA system, a QoS guaranteed cross-layer resource allocation strategy is presented. First, with cross-layer optimization techniques, we jointly consider the multi-layer service parameters and system characteristics. Through analyzing and adopting dependencies among the layers, we track differentiated QoS demands of multi-traffic and real-time channel changes dynamically to improve the multi-layer diversity gain of the system. Then, after analyzing fairness among users and conducting global unified modeling about QoS demands of multi-traffic, wireless time-frequency resource blocks, power resources, the modulation and encoding scheme of the future wireless communication system, we design a utility function aiming at ensuring QoS of multi-traffic, thus realize cross-layer resource scheduling. Besides, in order to reduce the additional communication overhead and system computational cost due to the introduction of cross-layer design, we propose a binary constrained particle swarm optimization algorithm, which can effectively reduce the complexity of the cross-layer resource scheduling model and ensure rapid convergence of iterative solution process. Under the premise of guaranteeing system fairness and QoS of multi-traffic, the strategy can improve the overall throughput of the system.2. For single-cell OFDMA system, a MOS-driven energy efficient cross-layer resource allocation strategy is proposed. The strategy focuses on wireless video services, which are the killer of the future wireless telecommunication networks. Besides, aiming at system energy consumption issue brought by the rapid development of wireless video services, the design of our strategy can ensure QoE of the service and improve effectiveness of wireless resource allocation strategy. First, we abstract related parameters of cross-layer services and wireless system according to the varying characteristics of video services and wireless networks, then ’power-distortion’model and’distortion-PSNR-MOS’mapping model are established to introduce video transmission power into the MOS mapping relationship. Second, consolidating MOS prediction information and the transmission power, we build a ’MOS-power-MOS energy efficiency prediction’resource allocation model which reflects the relationship between system energy consumption and video service parameters, wireless video channels effectively. Finally, the constrained particle swarm optimization algorithm is adopted to solve resource allocation and parameter optimization strategy. The strategy proposed in this thesis improves the MOS energy efficiency and reduces energy consumption of the wireless communications under the constraint of guaranteeing user experience.3. For multi-cell coordination network, a potential game theory-based resource allocation and scheduling strategy is presented. In order to reduce inter-cell interference, in addition to the cooperative multi-cell technology, the strategy also uses the pricing mechanism in the objective function modeling process and designed cross-layer resource scheduling model based on the pricing factor. Then based on the potential game theory we establish a cross-layer potential game model of resource allocation and scheduling, which maps cross-layer resource allocation strategy as potential game strategy. Through building a cross-layer potential game model based potential function, we prove Nash equilibrium existence and uniqueness by the convergence characteristics of the function, which reduces the complexity of the system model and objective problem solving effectively and achieves the optimal system effectiveness so as to improve the scalability of the system model. Finally, KKT conditions in convex optimization and iterative water-filling algorithm are adopted to solve the optimization problem. The strategy achieves the purpose of effective suppression of inter-cell interference and increasing overall system performance.4. For two-tire Femtocell network, a utility fair-based resource allocation strategy is proposed. First, the strategy studies interference suppression technology of shared spectrum mode in the two-tire Femtocell network, and jointly considers the cross-layer interference between macro cell and Femtocell and interference between Femtocell of the same layer. Under the premise of ensuring regular transmission of users in macro cell, we analyze the minimum SINR demand of users and present a system utility function which aims at maximizing spectrum efficiency. Besides, we map the objective optimization issue as the Nash bargaining power control model which is based on cooperative game theory. The model reflects the network spectrum utilization effectively and shows the fairness of the user utility. Finally, focusing on target Nash bargaining power control game model, we analyze and identify the bargaining solution with Pareto optimal RKS. The simulation results show that the proposed strategy not only improves the network spectrum utilization of Femtocell with the constraint of satisfying minimum SINR of users, but also guarantee users fairness well.