Research And Simulation On Resource Allocation Methods In OFDMA System

With rapid development of wireless communication technology and fast growth of various traffics, limited wireless resources, such as frequency and power, appear precious. The aim of wireless communication system is to make efficient use of such precious resources under different QoS (Quality of Service) requirements. Thus, the design of effective and flexible resource allocation plays an important role in system performance.OFDM (Orthogonal Frequency Division Modulation) has become more and more popular for its high frequency efficiency and strong ability to anti-jamming, and has become one key physical-layer technology in the next generation of wireless communication system. Resource allocation in OFDM system is to allocate subcarrier and transmit power with the knowledge of instantaneous channel conditions, which is crucial to guarantee high frequency efficiency for system and excellent QoS performance for user in the future wireless system. With this in mind, this thesis focuses on the study of various resource allocation schemes, and gives their performance simulation results in OFDMA systems.In this thesis, we start with an introduction of wireless resource allocation as well as some essential OFDM knowledge, and then put emphasis on the study of power allocation and subcarrier scheduling in OFDMA system. The study scenario includes both single-cell OFDM system and multi-cell OFDMA system. In case of single-cell OFDMA system, the study ranges from single user power allocation, multi-user subcarrier scheduling to joint subcarrier and power allocation. In multi-cell case, we focus on power allocation scheme and interference coordination technology.The main contribution of this work is to employ the PSO (Particle Swarm Optimization) technique to resource allocation, and propose a new resource allocation scheme in multi-cell OFDMA system. The simulation results show that the proposed method outperforms two other traditional methods in spectrum efficiency. This result also exhibits that, without extra information exchange between base stations, the proposed method is able to provide an excellent performance by coordinating transmitting power among neighboring cells, and combating intra-cell interference.Meanwhile, we present and simulate different kinds of resource allocation algorithms in single-cell and multi-cell OFDMA systems. In single-cell case:the following four methods have been exploited:uniform power allocation, water-filling, greedy bit loading and greedy bit loading based on water-filling method, and the simulation results show that subchannels with good condition transmit more power and bits, while subchannels with poor condition transmit less bits or even not transmit. When comes to multiuser subcarrier scheduling methods, the corresponding simulation exhibits that the Max C/I algorithm achieves maximum system throughput, the Round Robin algorithm achieves best fairness, and the Proportional Fairness algorithm obtains the trade-off performance in-between the max-C/I algorithm and the round-robin algorithm, that is try to achieve maximum system throughput as much as possible while still satisfying some degree of fairness between users.In multi-cell case:we sum up interference suppression technique. Besides, through studying and simulating some downlink interference coordination proposals, we have confirmed that frequency reuse can coordinate the inter-cell co-channel interference, which in turn influences system performance and cell-edge users' capability.Furthermore, this thesis also investigates interference pricing of the distributed network resource allocation to make a more comprehensive understanding of wireless resource allocation.