![](https://csdnimg.cn/release/download_crawler_static/15511151/bg1.jpg)
Asynchronous Pilots Scheduling in Massive
MIMO Systems
Zhichao Zhou
School of Computer and
Information Technology
Beijing Jiaotong University, China
Email: zczhou@bjtu.edu.cn
Dong Wang
School of Computer and
Information Technology
Beijing Jiaotong University, China
Email: wangdong@bjtu.edu.cn
Zhenchao Wang
School of electronic and
communication engineering
Hebei University, China
Email: wangzhenchaohbdx@163.com
Abstract—Pilot contamination has been confirmed as
a major bottleneck to performance improvement of
time division duplexing (TDD) massive multiple-input-
multiple-output (MIMO) systems. An asynchronous pi-
lots scheduling (APS) scheme is proposed in this paper
for pilot decontamination. According to interfering lev-
els, users are divided into cell-center users, which are
generally interfered slightly, and cell-edge users, which
suffer from severe interference. Base stations (BSs) in
all cells transmit forward-link symbols to cell-center
users, meanwhile the cell-edge users in different cells
send reverse-link pilots in non-overlapped time periods.
After the cell-edge users in all cells transmit pilots,
cell-center users in all cells send pilots synchronous-
ly to their corresponding BSs, whereas the cell-edge
users are served with forward-link data. In this case,
the BS can easily recover the estimation of the pilots
as it knows the forward-link data it is transmitting.
Additionally, the cell-edge users in adjacent cells send
pilots in non-overlapped time periods, so severe pilot
contamination of cell-edge users can be considerably
mitigated. Simulations verify the proposed APS scheme
in pilot decontamination.
Index Terms—Massive MIMO, channel estimation,
asynchronous pilots scheduling, pilot decontamination.
I. INTRODUCTION
Massive multiple-input-multiple-output (MIMO) is
a promising technique in which the base station (BS)
is equipped with a large number of antennas serv-
ing a smaller number of users on the same time-
frequency resource [1]-[3]. So massive MIMO is able
to improve spectral and energy efficiency significantly
[4]. However, pilot contamination, which results from
non-orthogonal pilots reuse in adjacent cells, becomes
a bottleneck to the performance of massive MIMO
systems [5].
Extensive research about pilot decontamination has
been conducted in [6]-[13]. Smart pilot scheduling
scheme [6] can significantly mitigate pilot contamina-
tion, but suffers from high computational complexity
as the number of terminals increases. [8] proposed a
precoding scheme based on minimum mean square
error (MMSE) for pilot decontamination, but design-
ing an optimal precoding scheme is a tough job. Joint
precoding among BSs are proposed in [9], which is
effective to reduce pilot contamination. However, extra
control overheads are required to exchange between
BSs leading to low time-frequency efficiency. The
approach in [10] use angle-of-arrival (AoA) to distin-
guish pilots that come from different users. However,
it is not always true that each user locates in a partic-
ular AoA in real scenario. Cells grouping and time-
shift pilots transmission [11] can reduce pilot con-
tamination considerably, but encounters interference
between non-orthogonal pilots from different cells in
one group. [12], [13] proposed a fractional pilots
reuse (FPR) scheme in which cell-center users in all
cells use the same pilots and cell-edge users in all
cells apply orthogonal pilots. The scheme is effective
to mitigate pilots contamination intuitively, but the
number of serving users is limited by orthogonal
pilots.
In this paper, we propose an asynchronous pi-
lots scheduling (APS) scheme in which the cell-
center users in all cells transmit forward-link symbols,
whereas the cell-edge users in different cells send
pilots in non-overlapped time periods. After that, the
same thing: cell-edge users are served in the forward-
link with data, while the cell-center users are detected
by the BS with their pilots in the reverse-link di-
rection. In this case, the BS can easily recover the
estimation of the pilots all the way as it always knows
the forward-link data it is transmitting. Additionally,
the cell-edge users in adjacent cells send pilots in non-
overlapped time periods, so severe pilot contamina-
tion of cell-edge users can be considerably mitigated.
Simulations verify the proposed APS scheme in pilot
decontamination.
The rest of this paper is organized as follows. In
Section II, we represent the system model and pilot
contamination is stated in III. The proposed APS
scheme will be described in Section IV. Numerical
analysis is discussed in Section V. The conclusions
are drawn in Section VI.
II.
SYSTEM MODEL OF MASSIVE MIMO
The scenario entails L hexagonal adjacent cells
and each cell consists of a BS with M antennas as
well as K single-antenna users. The channel includes
small-scale fading that follows superimposed geomet-
IEEE ICC 2017 Wireless Communications Symposium
978-1-4673-8999-0/17/$31.00 ©2017 IEEE