Page 56 - ITU Journal Future and evolving technologies Volume 2 (2021), Issue 6 – Wireless communication systems in beyond 5G era
P. 56
ITU Journal on Future and Evolving Technologies, Volume 2 (2021), Issue 6
to signal processing on relay nodes. Furthermore, channel capacity. Numerical evaluations are
since the cost is higher than other relay methods, conducted to verify the effectiveness of the
it is not suitable for deploying a large number of DF proposed massive analog relay MIMO system.
nodes. Intelligent Reflecting Surfaces (IRS) is also a The remainder of this paper is organized as follows.
remarkable solution. The IRS is a reflector that can In Section 2, a system model of the proposed
switch the reflection direction of radio waves using massive analog relay MIMO is described. In
e.g., meta-surface technologies [14]. However, IRS Section 3, expansion to multi-hop relay is explained,
cannot significantly improve the coverage and and Section 4 describes numerical analysis. Finally,
capacity since reflection directions are not Section 5 concludes and summarizes the paper.
controlled instantaneously. In this paper, we focus
on the Amplifier-and-Forward (AF) relay method 2. MMWAVE MASSIVE ANALOG RELAY
with active beam-forming capability. Since the AF
method only amplifies signals, delay is almost MIMO
negligible. Moreover, its cost, size, and power This section describes architecture of the proposed
consumption of nodes are relatively small, so that it mmWave massive analog relay MIMO that can
is suitable to install a large number of AF nodes. generate MIMO channels artificially by using a large
There have been several studies on AF relay number of analog RSs.
methods for mmWave communications [15][16]. 2.1 System model of massive analog relay
The performance of mmWave communications with
a full-duplex AF relay was investigated in [15] by Fig. 1 shows the system architecture consisting of
considering the effect of beam width and self- one source (BS), one destination (UE), and a
interference. In [16], the basic characteristics of massive number of RSs. Because of the compact size,
mmWave MIMO communications with multiple AF low cost, and low power consumption of analog RSs,
relays were evaluated under a fixed topology, and it it is feasible to massively and non-invasively deploy
shows potential of the method. When applying the them in the target environments to enhance or
AF relay to mmWave mobile communications, enable mmWave communications.
adaptive beam-forming is required to improve
communication quality. The conventional AF relay
only performs amplification and frequency
conversion but adaptive beam-forming has not
been considered. Moreover, relay-node selection
from massive relay nodes is also important in the
case of mobile communications.
In this paper, we propose an mmWave massive
analog relay MIMO system for mobile
communications by extending the authors’ previous
work in [17]. A large number of dynamically
controllable AF Relay Stations (RSs) are deployed to
relay signals for NLOS User Equipment (UE). Fig. 1 – System model for mmWave massive
Adaptive beam-forming in each RS is assumed. analog relay MIMO
A dynamic RS selection algorithm based on UE
location is introduced to generate an artificial
multipath channel to improve capacity of mmWave
MIMO communications. The RSs should have a
function of adaptive beam switching for dynamic RS
selection and a function of beam tracking to follow
the UE’s mobility. We start from the formulation in
single-hop relay, and then extend it to multi-hop
relays to improve mmWave capacity. The full search
of optimized route for multi-hop relays requires a
huge computational cost for beam selection due to
the massive number of RSs. Hence, a fast-routing Fig. 2 – mmWave analog RS with
algorithm is introduced in this paper to maximize beam-forming
44 © International Telecommunication Union, 2021