Research On Adaptive Resource Allocation Algorithms In OFDM System

With the development of society, mobile communications users are constantlyrising, which produced a huge contradiction with the lack of bandwidth resources. Inorder to solve this problem, a lot of researches has been done and finally OFDMadaptive technology has been developed. With this technology, the anti-multipleinterferences can be realised and the spectrum utilization be improved.OFDMtechnology will be used as the key technology of the4th generation mobilecommunications.OFDM technology that transforms high-speed data stream into several parallellow-speed data stream, these subchannels are orthogonal each other, this feature caneffectively reduce the interference, and can improve the spectrum utilization. Thispaper will discuss how to dynamically allocated the carrier and power resources tousers. To ensure fairness between users, it put fairness factor into the algorithm, thesystem will be better performance and guartantee the user’s quality of service.Based on the principle of OFDM system, the paper mainly includes thefollowing aspects: the development of mobile communications, the characteristics ofthe radio channel, OFDM system theory, the advantages and disadvantages, keytechnologies, and adaptive technology of the OFDM technologies.Next, the OFDMsymbol function, time domain and frequency domain model is analyzed detailed bysome basic knowledge, some basic knowledge of Convex optimization problem.Following the premise of adaptive allocation guidelines, the paper discusses theprinciple of water injection, Hughers-Hartogs, Chow, and Fischer these types ofclassic algorithms in single-user resource allocation algorithm, also compared witheach other.Then, it discusses comprehensively the multi-user OFDM adaptive systemresource allocation algorithm. Looking up References that it introduces briefly thePSO and fish algorithm, on the basis of artificial immune algorithm, produceimproved artificial immune algorithm which consider the fairness between users,，andthen comparing the performance by the MATLAB simulation, and made an analysis,verified the improved algorithm outperforms Particle Swarm Optimization andFish-Swarm algorithm.Finally, summary has been briefly made, some suggestions for improvement of this research have been given and finally the aspects of further research on the subjectare put forward.