Key Technologyies Research In Distributed MIMO Systems

The next generation mobile communication systems require large-capacity and high-quality data transmission.It is believed that MIMO technology will be one of the key technologies and has wide application prospect in the third generation (3G) and the fourth generation (4G) mobile communication systems.In traditional centralized MIMO systems,the multiple transmitted antennas are all co-located in the base station (BS).Distributed MIMO systems are different from centralized MIMO systems,in which the multiple transmitted antennas always locate in different areas, the channels between transmitted antennas and received antennas are more independent. Compared with centralized MIMO, it has the advantages of large system capacity, low transmit power, enhanced coverage and low radiation on the human body. Distributed MIMO systems have been promised as one of the candidate systems in the future wireless communication systems.In this dissertation, key techniques in the transmit side related to the distributed MIMO systems are investigated.Detailed research works include:the channel characteristic in distributed MIMO systems,the channel capacity analysis in multi-cell environment, antenna selection technology, power allocation technology, multi-user precoding technology and so on.The detailed research contents and the main contributions of this dissertation are listed as follows:For the channel characteristic in distributed MIMO systems,the correlations between multiple co-located antennas in the same distributed port are analyzed, and the Kronecker correlation channel model is given. This dissertation also investigates the correlations among different distributed ports and draws the conclusion that the correlations are monotonic decreased with the distance between base station (BS) and user equipment.For the channel capacity of distributed MIMO systems, this dissertation first considers that distributed ports only equip one antenna, and the ergodic capacity expression of multi-cell environment is derived. Furthermore, the channel capacities of distributed systems and cellular network systems are simulated.Simulation results show that compared with cellular network systems, distributed systems can improve channel capacity in most areas.It is also shown that selecting fewer antennas for transmission can decrease the interference among neighbor cells and further enhance the channel capacity. Second, when distributed ports equip multiple antennas, the ergodic capacity is also derived and simulated.For the antenna selection, the upper bound expression of channel capacity is derived in this dissertation.Then, with the capacity criterion, an adaptive antenna selection scheme based on partial channel state information (CSI) is proposed.This scheme uses the large scale fading information to select the transmitted distributed port.When the user's location varies, BS can adaptively adjust the transmitted antenna subset to obtain the maximum channel capacity according to users'position. Simulation results illustrate that compared with tradition blanket transmission scheme and selection diversity scheme, the proposed scheme can achieve more ergodic capacity.For the power allocation technology, a low feedback power allocation scheme based on partial CSI is proposed.The proposed scheme uses the large scale fading information and the rank of transmit correlation matrix to allocate the transmit power. Since the large scale fading information and the transmit correlation matrix vary slowly, the feedback period can be set long. Therefore, the feedback overhead can be reduced.Simulation results verify that the proposed scheme can greatly improve ergodic capacity compared with the average power allocation scheme, and the capacity loss compared with the optimal water-filling scheme is less.For the multiple distributed MIMO systems, a precoding technology based on channel prediction is investigated in time-varying channels. However, in the higher speed mobile environment, channel prediction will result in serious prediction errors. To solve this problem, an improved signal-to-leakage-and-noise rate(SLNR) precoding scheme based on prediction error compensation is proposed. The proposed scheme uses the normalized mean square error (NMSE) of channel prediction to modify the precoding matrix.This compensation reduces the system performance loss due to the unmatched channels.With this scheme, the interferences among multiple users are suppressed efficiently and the system capacity is further enhanced.Simulation results show that the proposed scheme can obtain better bit error rate (BER) and throughput performance in the higher speed mobile scenario.