Scheduling in Massive MIMO: User clustering and pilot assignment

Abstract

In this paper, we consider the problem of user scheduling and pilot assignment in TDD multicell multiuser Massive MIMO systems. While in TDD systems the channel is acquired using uplink pilots, we propose a scheme that utilizes additional downlink probing in order to improve the spectral efficiency. The idea is to dynamically assign mobile users to different clusters based on the directions of their channels through the use of downlink reference beams. This will result in forcing the interference to be centered in semiorthogonal subspaces without the need for important feedback and therefore enabling reduction of the pilot contamination effect. The scheduled users in each cluster employs orthogonal training sequences with a pilot reuse factor of 1 among the clusters. We then propose a graphical framework for pilot assignment. We show that this problem can be modeled as a max cut problem and we provide an approximation algorithm that optimizes the pilot allocation.

Covariance-Aided CSI Acquisition With Non-Orthogonal Pilots in Massive MIMO: A Large-System Performance Analysis

“…Finally, we contributed to random matrix analysis by extending the trace-lemma from [24] to block matrices. 16 32…”

“…Indeed, [14] demonstrates that sharing (perfect) covariance information across different cells results in unbounded spectral efficiencies (as the number of BS antennas increases without bound) under a fairly mild assumption on the linear independence between the user covariance matrices. In the context of CSI acquisition, covariance information has been mainly exploited to propose orthogonal pilot reuse strategies [12], [13], [15], [16] or non-orthogonal pilot designs [17]- [19] to mitigate pilot contamination [20]- [22], that is, the undesired effect of obtaining a channel estimate that is contaminated by the channels of other users. All these methods rely on the intuition that users can be (partially) separated in space using their individual covariance matrices during the CSI acquisition process and perfectly orthogonal pilot sequences are no longer needed.…”

On Optimal Scheduling for Joint Spatial Division and Multiplexing Approach in FDD Massive MIMO

“…where x uv is a binary variable which is null when (u, v) ∈ V c × V c are assigned to the same cluster and is 1 otherwise. The constraints found in (7) account for the symmetry of x uv and the triangular inequality 2 satisfied by these binary variables. The interesting aspect of this partitioning formulation is that there is no need to give the target number of clusters as input.…”

“…This comes from the fact that the downlink channel estimation overhead scales linearly with the number of antennas [6]. This is mitigated in TDD systems by exploiting the channel reciprocity since the channel estimate of the uplink direction can be directly utilized for the downlink direction [6] [7] which is not feasible in FDD systems.…”

On the machine learning–based smart beamforming for wireless virtualization with large‐scale MIMO system

“…1,2 Large-scale multiple-input-multiple-output (MIMO) is regarded as an emerging technology to enhance data rate of future wireless networks 2 and the wireless virtualization is regarded as an efficient paradigm to enhance the radio frequency (RF) spectrum utilization by subleasing RF slices of wireless infrastructure providers (WIPs) to mobile virtual network operators (MVNOs). 5,6 In order to support heterogeneous wireless services, it is costly and impractical to deploy a separate dedicated network optimized to provide the required QoS for each category of service. 2,4 To meet variety of Quality of Service (QoS) requirements of diverse users, future wireless networks must be extremely flexible and adaptable to the operating environment.…”