Research On Virtual MIMO Techniques In Wireless Communication Systems

The Multiple-Input Multiple-Output (MIMO) is one of the key techniques for the next generation of wireless communication systems. But as the limitation of space and cost, it is difficult to equip multi-antenna in the user equipment. To solve this problem, 3GPP LTE contributes a technique called virtual MIMO, which groups a number of single-antenna users to form a virtual MIMO channel in the same resource block and transmit users'data to the base station with multiple antennas. It can indeed achieve the benefits of MIMO technique, but there are several new problems remaining to be solved, such as: how to allocate power to different cooperative diversity users, and how to pair users for spatial multiplexing. So, it is of important value to do research on virtual MIMO techniques both theoretically and practically.This dissertation firstly gives some discussion on the cooperative virtual MIMO technique. In this technique, users share their data and transmit the data to the base station cooperatively. By the reason of limited system power, we need to allocate the total power to each cooperating user properly. The existing power allocating methods haven't taken much consideration of the realistic channel state, so they do not perform well when the channel state is bad. In this dissertation, a new power allocation method is proposed to overcome the drawback of existing methods. Based on the system model in cooperative communication, the new method defines a new symbol of MCPA (Mean Channel Power Attenuation), and uses MCPA to set the selection threshold for cooperative nodes. Simulation results show that, compared to those existing power allocation methods such as equal allocation to all nodes and equal allocation to selected nodes with threshold based on outage probability, the new method exhibits better performance in various kinds of channel condition, and is more pronounced in the situation of bad channel condition.This dissertation gives deep research on the non-cooperative virtual MIMO technique. In this technique, after being scheduled, the users transmit their data streams independently to the base station. The base station separates different data streams from different users by multiple antennas and receiver algorithms. Because the number of different users that the receiver can resolve is limited by the number of receiving antennas that the base station has, there is limited number of users that can participate virtual MIMO in one time slot. As a result, the user pairing algorithm is a key issue to the system. In this dissertation, a novel user pairing algorithm is proposed, which is based on semi-orthogonal user selection (SUS) with partial inverse selection. In the traditional SUS algorithm, users with bad channel condition will suffer low throughput. To solve the problem, we propose a mechanism of'inverse selection', which derives from the analysis of the MMSE receiver. Simulation results show that the new algorithm largely improves the throughput of users with poor SINR at a reasonable cost of the overall throughput decrease. Thus, it achieves a profitable trade-off between system throughput and user fairness.This dissertation gives some analysis on the frequency-selective fading in wideband systems, and describes the model of the non-cooperative virtual MIMO systems combined with Single-Carrier Frequency Devision Multiple Access, and also provides some basic simulations for user pairing problems.In the future, there are several subjects worth further studying, including the novel user pairing algorithms for wideband virtual MIMO systems, the algorithm design when synchronizations, channel estimations, and coherent receiving are in an unideal enviroment, the virtual MIMO combined with macro deversity at the cell edge due to multiple base stations, the analysis about the performence of virtual MIMO with channel correlations, and so on.