Research On Massive MIMO Technology In Next Generation Mobile Communication System

With the increase of user requirements, high data-rate, high-quality communicationsystem become the research object of the new generation mobile communicationtechnology. Multiple antenna technology characterized by multiple input multipleoutput (MIMO) is recognized as the best approach to achieve the above goals. In thepractical system, however, due to the demand of high data rate and serious inter-cellinterference, the research work focus on the massive MIMO technology, whichemploy large-scale antenna array in base station (BS). The antenna number in thistechnology is usually more than ten, or even hundred, and can serve multiple userswithin the same time-frequency resources. Compared with the traditional MIMO,massive MIMO technology can not only significantly increase the system spectrumefficiency, improve the communication quality, but also reduce the complexity of theinterference elimination and transmitting power of base station effectively.The spectral and energy efficiencies of massive MIMO, as well as its performancein heterogeneous networks are presented in this paper. Firstly, using the randommatrix theory, the maximum ratio transmission (MRT), performance of zero-forcing(ZF) normalized pre-coding, and the closed expression of spectrum efficiency areanalyzed in both single-cell and multi-cell large-scale MIMO systems when thenumber of BS antenna are increased. Then, through the theoretical analysis andMATLAB simulation, the relationships between energy efficiency of these twopre-coding methods and antenna numbers as well as transmitting power are given.Finally, the performance analyses of heterogeneous networks with massive MIMOtechnology are concluded by comparing three different duplex modes.The results show that:(1) Under the conditions of complete channel stateinformation and the reduction of BS transmitting power to1/M of single antennasystem, the spectral efficiency tends to a constant and the limit expression of MRTnormalized pre-coding is equivalent to that of ZF normalized pre-coding, when theBS antennas number M increases indefinitely. So the interferences of inter-user and inter-cell as well as the fast fading influence can be eliminated by simple MRTpre-processing.(2) Under the condition of limited CSI, the inter-cell interferencecannot be completely eliminated because of the influence of CSI estimation. But thesystem status will transform from interference-limited to noise-limited with thereduction of transmitting power.(3) The system energy efficiency increases with theincrease of BS antennas, whereas it decreases with the increase of transmitting power.As a result the massive MIMO technology can improve the system energy efficiencyby either increasing the antennas number or reducing the transmitting power.(4) Inheterogeneous networks where the macro base station employ traditional multipleantennas as well as the micro base station is equipped with the large-scale antennas,the system (operating in the separated-frequency bands TDD mode) will not onlymeet the high-speed data transmission requirements of next generation mobilecommunication system, but also effectively avoid the influence of the inter-cellinterference.