Investigation On Precoding Technology In Cooperative Mimo Multiple Relay System

With the rapid development of modern wireless communications business, the contradiction between the limited radio spectrum resources and the requirements for higher speeds and better service quality is growing lager. Multiple-input multiple-output (MIMO) technology has been shown to effectively improve the system spectral efficiency and error performance without increasing the bandwidth and transmit power. However, due to the limitations of the actual terminal, it is difficult to apply in the actual communication system. Generation of cooperative communication finds the breakthrough. Its main idea is by increasing the relay link to form a virtual MIMO, so as to achieve distributed transmission and the diversity gain. For consideration of the fairness of the user and terminal power constraints, it is more feasible to form a cooperative relay system through the deployment of relay stations than collaboration between users. This thesis studies the pre-coding in cooperative multi-relay system.First of all, on the basis of accessing large amounts of literature, the research status of cooperative MIMO relay systems and pre-coding techniques are reviewed. Pre-coding and detection techniques of traditional MIMO point-to-point system are briefly introduced and analyzed in the second chapter.Secondly, the model of multi-relay cooperative MIMO transmission is led. In the multi-relay system of single-source single destination, the advantages of multi-relay technology is distributed transmission structure by each relay link. For coherently combined of the signals from all the relay nodes in the destination node, relay nodes can use bi-directional zero-forcing, minimum mean square error and the QR decomposition approach. A bi-directional QR decomposition based on MMSE relay approach is analyzed. It combines the advantages of the minimum mean square error method and the QR decomposition method, taking into account the impact of noise, making the equivalent system channel triangle, supplemented by successive interference cancellation detection. Numerical examples show that the method enhances the system error performance.Finally, the transmission structures for multi-relay system of single-source multi-destination nodes and multi-sources multi-destination nodes are analyzed. For the former model, a block diagonalization with singular value decomposition or geometric mean decomposition joint pre-coding method is introduced, which eliminates the interference between users, transforming the system into multiple parallel sub-channels, In order to obtain distributed array gain, a zero forcing combined with block diagonalization relay approach is analyzed. For the latter model, it can also eliminates the interference between different users by the block diagonalization method supplemented by additional processing while obtain the distributed array gain. The performance characteristics of the different processing methods are compared through simulation.