Resource Allocation Scheme Using Proportional Fairness For Cooperative Cognitive Radio Networks

In recent years, with the rapid development of wireless communicationtechnology, it is required for a communication system to have a higher dataprocessing capability and transmission capacity. Cognitive radio technology (CR) is anew concept proposed for improving the overall utilization of spectrum bands. Thistechnology allows the unlicensed secondary users (SUs, also referred to as CR usersor CRUs) to access those frequency bands which are not currently being used bylicensed primary users (PUs) in a given geographical area. Owing to the sufficientflexibility to support high transmition data rate, high spectral efficiency and capabilityto ensure system stability, CR technology has become a hot research topic.The cooperative communication technology supports wider network coverage andhigher system transmition rate through resource multiplexing. Using the diversity ofradio users, cooperative relaying technique can further improve the transmition datarate, fairness and spectrum efficient. Therefore, cooperative communicationtechnology has become another hot research topic in cognitive radio system recently.In the thesis, we firstly introduce the concept of cognitive radio and cooperativecommunication technology. Then we mainly research the multi PUs multi SUssubcarrier-pair based resource allocation scheme using proportional fairness fromboth AF (amplify-and-forward protocol) and DF (decode-and-forward protocol),respectively. It is considered that the relay users need to send their own data. Lastly,we extend the resource allocation and subcarrier pairing algorithm under the scenariothat each cooperative user has no channel state information (CSI) about non-adjacentlink. Three fairness factors, proportional (max-pro), max-sum and max-min werecompared and analyzed in the thesis.The tradeoff between throughput and fairness in CR system is analyzed and adetailed analysis of the performance of the proposed algorithm is introduced with thesimulation results. Simulation results have shown that max-pro is a well-balancedcriterion between efficiency and fairness compared with max-sum fairness andmax-min fairness. Simulation results have shown that, either in improving systemthroughput or in improving fairness of resource allocation, the proposedsubcarrier-pair based resource allocation algorithm offers the best performanceconditions among several existing compared resource allocation algorithms.