Research Of Key Technology On Localization In Massive MIMO System

Multiple input multiple output(MIMO)technology is a key technology of the cellular radio systems.Nowadays,along with the development of wireless cellular network system,the number of mobile users is growing fast.Wireless localization technology is also becoming more and more indispensable.The fifth generation(5G)technology will be widely applied in the future,massive MIMO base stations are equipped with antenna array that consists of a huge number of antennas.Reasonably using receiving signal parameters to locate in the 3-dimensional(3-d)space can not only provide location information for the users,but also provide mobile users with power control and handoff switching.First,the background of the massive MIMO technology and wireless localization technology,state-of-art technologies and current problems are summarized.The integration of above two technologies brings up the main research content of this article,the massive MIMO localization technology.Secondly,we introduce common localization algorithms used in the wireless localization systems and analyze their advantages and disadvantages,respectively.Then the main problems in the wireless localization technology and the existing solutions are analyzed and the common performance metrics of localization systems are introduced.Based on the above basic knowledge,multi-parameter joint localization technology is finally selected.Thirdly,traditional direction of arrival(DOA)estimation algorithms are introduced.Then we compare multiple signal classification(MUSIC)and estimation of signal parameters via rotational invariance techniques(ESPRIT)algorithms and select the ESPRIT algorithm for massive MIMO system.Then a 2-dimensional(2-d)DOA estimation algorithm of rectangular antenna array is deduced.After that,we put forward an improved 2-d DOA estimation algorithm,which has the best performance under the square array.At last,limitations of the least-square joint localization method and parameter reconstruction localization method based on maximum likelihood are investigated.We generalize the least-square joint localization method to 3-d localization system and propose a multi-parameter joint constraint localization method based on transmission model under non line-of-sight(NLOS)environment.This method first preprocesses multipath signals to get signal parameters,then gets constraint inequality group by transmission model,and finally figures out the estimated location by solving an unconstrained nonlinear least-square problem.Simulation results show that the proposed algorithm can effectively reduce the localization error under NLOS environment.