Research On User Scheduling In MU-MIMO Broadcast System

The application of the multiple input multiple output (MIMO) technology improves the spectrum efficiency and throughput of the communication systems and satisfies the practical demand for the high data rate and reliability, owing to the spatial multiplexing and spatial diversity. So MIMO is becoming more and more popular in recently. However, in an MIMO communication system with the existence of multiple users (MU-MIMO), the multiple user interference (MUI) is inevitable. Fortunately, precoding methods are proposed to eliminate the MUI. In the precoding processing, user selection is essential in the overloaded MU-MIMO system. This thesis focuses on the user selection algorithms in MU-MIMO systems.Firstly, the discussion about the development and the evolution of the mobile communication is made, and the technologies about the mobile communication are analyzed. Mostly, the challenges of the new technology are characterized. Then, the models of MIMO system are given, including single user MIMO system and MU-MIMO system, and the analysis of the system capacity is followed. After this, some precoding schemes and power allocation algorithms in MU-MIMO systems are introduced, especially waterfilling power allocation algorithm. In this part, the block diagonalization (BD) precoding method is mainly discussed in the channel state information (CSI) aware system. Based on the BD precoding scheme, two kinds of user selection algorithms in MU-MIMO system are introduced, which are based on the operations of the channel matrices and the subspace spanned from the channel matrices respectively.The first part of the description of the user selection algorithm begins with the optimal selection algorithm, which selects users based on the whole user subsets and is established on the exhaustive searching. It has the highest computational complexity with the highest capacity. Then the user selection algorithm based on the capacity comparison is introduced. To reduce the complexity, another user selection algorithm is proposed. Last but not least, the user selection algorithm based on the condition number of channel matrix is proposed by us. Then, numerical simulations are made to analyze the performance of these algorithms.In the other part of the description of the user selection algorithm, the classic semi-orthogonal user selection algorithm (SUS) is given at first. Based on the idea of this algorithm, L2-Hausdorff algorithms are proposed with two types. One method is based on the L2-Hausdorff distance and capacity, and another one is based on L2-Hausdorff distance and the cardinality-reduced subsets. Besides, the algorithms based on the angles between subspaces are proposed. At last, the simulation results and analysis about those algorithms are given.In short, this thesis mainly studies about the user selection algorithms in MU-MIMO systems. And several user selection algorithms are introduced. The simulation results confirm that the proposed algorithms are more practical because of the good tradeoff between the throughput and the complexity.