Research And Implementation Of Precoding Technology In Massive MIMO

Massive multiple-input multiple-output(Massive MIMO), which uses numerous antennas in transmitters and receivers, has been the key technique of the next generation of wireless communication techniques to further maximize network capacity and conserve transmission energy. The cloud-radio access network(C-RAN) architecture has emerged as a promising solution due to its potential to reduce energy consumption, improve spectral efficiency and network capacity, with lower implementation and operation cost. However, one practical hurdle for its large scale implementation is the stringent requirement of high capacity and low latency fronthaul connecting the BBU(Baseband Processing Unit) to RRU(Remote Radio Unit). Based on this background, the massive MIMO precoding technique in C-RAN are analyzed in this paper.Firstly, we describe the MIMO system model and MIMO channel capacity. Detailed analyses of the channel characteristics and the system capacity about point-to-point massive MIMO system, multiuser massive MIMO system and multicell massive MIMO system with infinite number of antennas are also given in this paper.Secondly, we study the linear precoding techniques in massive MIMO such as MF(Matched Filter), ZF(Zero Forcing) and MMSE(Minimum Mean Square Error). The performance about the spectral efficiency and energy efficiency under the different precoding schemes is analyzed. And we compare the performance of the different precoding schemes under the perfect and imperfect channel state information. Also we introduce the basic principles of compressive sensing. And several main signal reconstruction algorithms are dicussed and simulated.Thirdly, we make an in-depth study of the massive MIMO precoding technique in C-RAN. Under the limited capacity of the fronthaul, the system capacity of the fully centralized precoding and partially centralized precoding is analyzed. In order to further reduce the fronthaul loading, the partially centralized precoding is improved combined with the compressive sensing based on the channel correlation properties in massive MIMO system. The channel sparse representation and system performance are analyzed and compared. Compared with the first two precoding methods, the proposed algorithm greatly improves the system capacity under the limited capacity of the fronthaul.Finally, considering the implementation complexity and the actual conditions of hardware platform, we study the implementation of the key ZF precoding module. We propose an optimized structure to implement the module by using the parallel multiplication. Then, we design the hardware module based on Altera Stratix V. The functional simulation and the board test can verify the reasonableness of the circuit design.