Design And Simulation Of Sub-carrier Allocation Algorithms In OFDMA Relaying Networks

Due to the characteristics of combating the frequency selective fading and supporting high data rate effectively, Orthogonal Frequency Division Multiplexing (OFDM) has been widely acknowledged as a key technology for next-generation mobile communication system and wireless broadband access system. Meanwhile, relay technology can improve the quality of service of edge users, upgrade system capacity and enlarge cellular coverage at low cost, therefore it is becoming a research hotpot in the future wireless communication system.This thesis focus on sub-carrier allocation and usage in the OFDMA relaying network, and proposes two efficient resource allocation and optimization schemes which can maximize network throughput and resources utilization. The first scheme finds the links which are suitable for data transmission from a base station to terminals at the begining. Then, for the purpose of increasing resource utilization, it need to group links so that interfere-free links are in the same group to realize resource reuse. To real-time, dynamically and efficiently use network resources, it is necessary to group sub-carriers to guarentte that link groups can complete data flows transmission by better resources. Due to the link grouping strategy, some links in a link group have surplus resources which bring chances for further reuse. According to this information, sub-carrier allocation can be accomplished. However, with the increasment of the number of the terminals, the number of links and sub-carrier groups should increase and some link groups can not choose better subcarrier groups for communication. In order to optimize the link groups and sub-carrier groups, the Tabu Search (TS) algorithm is introduced in the second sub-carrier allocation scheme.For the purpose of performance evaluation, we construct the simulation environment by MATLAB, and make comparisons with other algorithms. Simulation results demonstrate that the algorithms proposed in this paper outperform other algorithms as expected.