Research On User Selection Schemes In Multi-cell Scenarios

The mobile business will increase exponentially in the future,also the performance requirements of the users for mobile/wireless network are getting higher and higher,but the spectrum of resources has a serious deficit.In the real environment,due to the interference source,shade,signal multipath and other factors,a common network device is usually difficult to support a large number of concurrent users online.Especially in the high-density network scenarios,due to the existence of large amounts of co-frequency overlapping area,mutual interferences between users will seriously limit the performance of the system.This paper studies the user selection algorithm in the multi-cell multi-user environment,selecting a group of users with specific channel characteristics to reduce the interference between users and realize high-speed data transmission.Compared to the single-cell communication scenarios,users that in the multi-cell multi-user transmission scenarios suffer the interferences not only from users in the same cell,but also from users outside the cell.So,at the sending and receiving side,the interference that need to be dealt with is more complicated.In order to eliminate these interferences and take into account the implementation complexity,we usually adopt linear processing scheme in the transmission(such as ZF and MMSE).However,since the performance of the linear precoding or receiver is strongly constrained by the orthogonality of channels(or the condition number of the channel matrix),it is important to employ the user selection to improve the orthogonality of the channels so as to improve the system's transmission performance.This paper proposes the user selection algorithms for multi-cell scenarios based on the orthogonality of channels.Firstly,this paper proposes centralized user selection schemes under the assumption that the perfect CSI can be obtained and there is a central processing unit to schedule the users.Secondly,the paper also considers distributed user selection schemes with the condition that the base stations are partially coordinated to share some scheduling information.In addition,the paper proposes a user selection that has low feedback overhead in the limited feedback system.In the design of the user selection algorithms that proposed in this paper,we use the tools of vector inner product and matrix determinant,and both achieve a considerable performance gain.In this paper,the user selection algorithms in multi-cell multi-user scenarios can greatly reduce the interferences between users and improve the transmission rate of the communication system,which can guide the interference management in the future 5G high density wireless network.