Research On Adaptive Transmission Schemes In Multi-user Downlink System

Multiple-input multiple-output (MIMO), which is considered as one of effective techniques for the broadband high-speed data services in future, has been proven its potential to offer high spectral efficiency as well as link reliability without additional power and frequency expenses. The fourth generation mobile communication technology needs to support multimedia services such as video, images and others, which are often the up / down asymmetric data businesses, concentrated in the downlink transmission bottlenecks. The adaptive transmission techniques matching the load channel information by dynamically adjusting the transmission parameters, transmission programs and services, can make an improvement in the system throughput. So, research on the MIMO downlink systems with adaptive transmission techniques is very significant.Based on the depth analysis of the existing adaptive transmission strategy and multi-user scheduling algorithm about Beamforming, the paper proposes a mixed-mode orthogonal transmission scheme based on beamforming to solve the problem that the capability is low when the number of users is small; further more, to make a better trade-off between the capability of system and the fairness of users. The paper improves two multi-user scheduling algorithms. We completed Specific work of the paper as follow:Firstly, the paper designs a cost function to optimize the technical performance of beam-forming transmission. There is a problem that the capability is low when the number of users is small in Orthonormal Random Beamforming algorithm. The improved correlation algorithms improved system throughput, but requires an additional channel information feedback and an increase of overhead. In response to these problems, on the base of ORBF model, the paper designed a cost function to consider the "income" and "cost" relationship of system to minmum the expense, and to improve the overall system performance. The results show that with Small increase of feedback, we get better performance of the system when the number of system user is less.Secondly, the paper proposed an adjustable scheduling algorithm based on SINR. On the base of the study of the max(C/I) scheduling algorithm and Round Robin algorithm, we proposed an adjustable maximum fair scheduling algorithm. New algorithm designs the signal to interference ratio which the user back to the base station differently. If the user is scheduled in the last slot, according to certain rules, then reduce their feedback of signal in the next time slot; if the user has not been scheduled in the last slot, according to certain rules, then compensate the user in the next slot. For this, we can give users in poor channel conditions more opportunities to communicate and improve the fairness among users. Also the new scheduling algorithm can be applied to diffirent systems by adjusting the fairness variable.Thirdly, the paper proposed an improved Proportional Fairness algorithm with packet characteristic. We completed the design from mathematical average to exponential moving average, and proved its importance in communication system. By taking into account the communication features of the data series and time of serving, we dynamicly designed statistics window, making the size of the time window changes with the characteristics of the user. Futher more, we discussed the time fairness and the service fairness of the new algorithm by new methods.The simulation results show that the proposed new algorithm has a lower system delay and better system fairness than the traditional method.