Precoding And Power Allocation In Wireless Communication Systems

With the rapid development of wireless communications, limited wireless resource will not satisfy the requirement of various wireless applications. The challenges of wireless communication system are high throughput, high reliability and various appli-cations. How to improve the utilization efficiency of wireless resource will have great impact on the development of wireless communications. Precoding and power alloca-tion are two effective approaches to combat channel fading and multiuser interference to enhance the utilization efficiency of wireless resource and the system performance. This dissertation focuses on the linear precoding and power allocation techniques, in-cluding eigenvector, zero forcing (ZF), minimum mean square error (MMSE) and max-imum ratio transmission (MRT) precoding. Precoding and power allocation are jointly optimized to optimize certain performance metrics to combat channel fading and mul-tiuser interference and to improve system performance. This dissertation covers the covariance based precoding in cooperative relay networks, joint precoding and power allocation in relay networks as well as in cognitive radio networks. The mentioned key problems are well analyzed and some possible solutions are proposed to give some aca-demic references to future wireless communication systems. The main contributions of this dissertation are as follows:Precoding based on covariance channel state information (CSI) in three-node co-operative networks and multiuser two-hop relay networks are studied respectively. In the three-node cooperative network, precoding is designed based on the covariance CSI of the direct channel and relay channel. We derive the signal model and get the MMSE precoding matrix. In the multiuser two-hop relay network, we study precoding schemes in AF and DF relay modes. Multiple users are served in code division multiple access (CDMA). For the AF relay network, precoding is done at the source. While in the DF network, precoding is performed at the source and the relay, respectively. The pre-coders at the source are designed based on the covariance CSI from the source to the relay, and the precoders at the relay are designed based on the covariance CSI from the relay to users. We get the MMSE precoding matrices and compare the performance of precoders in AF and DF.We study the problem of joint precoding and power allocation in multiuser two-hop relay networks. Firstly, this problem is studied in two-hop AF relay networks. Joint precoding and power allocation is performed at the base station. The relay sta-tion only amplifies the received signals and forwards them to users. Three precoding schemes are obtained based on ZF, MMSE and MRT. And the optimal power allo-cation strategies corresponding to different precoding schemes are derived. For the system with ZF precoding, the power optimization problem is a convex optimization problem, while it is a nonconvex one for the system with MMSE or MRT. We convert this nonconvex optimization problem to a convex one and get the solution by using convex optimization theory. Then, to improve the spectrum efficiency, two-way relay networks are considered. An iterative algorithm to get the optimum relay processing matrix and power allocation strategy is proposed for this system model. In this work, considering fairness among users, the optimization problem is formulated to minimize the total transmit power subject to signal to interference plus noise ratio (SINR) of each user. The optimal relay processing matrix and power allocation strategy are derived respectively and then the global optimal schemes are get by iteration.Joint precoding and power allocation in multiuser cognitive radio (CR) networks is investigated. We first study this problem in ordinary CR networks and propose two precoding schemes to remove or suppress interference from secondary users (SUs) to primary users (PUs). Besides, optimal power allocation strategy is employed for SUs to improve the system performance of secondary network. This nonlinear nonconvex optimization problem is converted into a convex one and solved by convex optimization theory. Then, to improve spectrum efficiency of SUs, we apply the two-way relay protocol in CR networks, with which SUs exchange their information through a relay station. Three relay processing schemes are derived based on ZF, ZF-MRT and MMSE to suppress multiuser interference among SUs, interference from the relay station to PUs and to exchange information between SUs. For different relay processing schemes, optimal power allocation is employed to optimize the performance of SUs and to limit the interference from SUs to PUs.