| As a key role of improving the spectrum efficiency of wireless communication systems, the multiple-input multiple-output (MIMO) technique was developed quickly during the last decade. In partciluar, since nowdays the topology and the requirement of celluar networks and other wireless networks become more and more diverse, based on the researches on traditional point-to-point MIMO systems, some advanced MIMO techniques have been proposed recently to deal with these new characteristics of wireless systems, such as multi-user MIMO techniques, cognitive MIMO techniques and green MIMO techniques. Compared with the traditional point-to-point MIMO techniques, these emerging MIMO techniques can significantly improve the performance of the MIMO systems under some special topologies and meets the requirments of future wireless systems. Therefore, the aforementioned advanced MIMO techniques have drawn wide attention in recent years.This thesis focuses on the scheduling techniques of muti-user MIMO systems, the limited-feedback transmission schemes of cognitive MIMO systems and the power allocation and precoding techniques of green MIMO systems. The main contribution of our works is proposing some simple and efficient resource allocation schemes on antenna mapping, power allocation and user scheduling according to the requirements of these advanced MIMO systems, and also providing some theoretical results on the system performance.Firstly, to reduce the computation complexity during the muti-user MIMO scheduing, a low complexity mutli-user MIMO scheduling algorithm is proposed. Based on the well known semi-orthogonal user selection algorithm, the proposed algorithm is obtained via the iteration relationship of the parameters between each user selection step according to the matrix theory. It can be proved that the proposed algorithm can achieve the same throughput as the semi-orthogonal user selection algorithm while the complexity order is reduced by one.Secondly, to relieve the performance decrement during multi-user MIMO scheduling caused by imperfect channel state information (CSI), a robust multi-user MIMO scheduling scheme is proposed. The proposed algorithm can improve the system througput via mesuring the ergodic capacity under CSI errors. Based on the lower bound of measured ergodic capacity, a CSI modification factor is also proposed to modify the scheduling user CSI and make the tradional multi-user MIMO scheduling algorithm based on perfect CSI become robust under imperfect CSI scenarios. Simulation results show that the poposed CSI modification factor can improve the scheduling performance of the mutli-user MIMO systems with imperfect CSI significantly, and achieve the performance of the proposed ergodic capacity based scheduling algorithm.Thirdly, to deal with the limited-feedback transmission problem of cognitive MIMO systems, a cognitve MIMO transmission scheme based on the statistical CSI feedback is proposed, and the corresponding system performance analysis is also given. The proposed scheme gives the power allocation and the precoding algorithms to maximize the ergodic capacity based on the statistical CSI feedback and the MIMO channel fading characteristic. Simulation results show that the proposed scheme can approach the optimal system performance with perfect CSI.Finally, according to the energy efficiency requriment of green MIMO systems, an energy efficiency optimization power allocation algorithm for green cognitive multi-user MIMO systems is proposed. Since the energy efficiency optimization problem of green cognitive multi-user MIMO systems is non-convex and hard to solve directly, the proposed algorithm transforms it to an equivalent quasi-concave one-dimension problem to solve it. The convergence and the optimiality of the proposed algorithm are shown in theory. Similuation results show that compared with the traditional transmission scheme, the proposed scheme can improve the system energy efficiency significantly and meets the requirments of green communications. |
PDF Full Text Request