Conference PaperPDF Available

Simulation based channel hardening of cell-free massive MIMO in mm-Wave

Authors:

Abstract

Channel hardening is one of the key properties of massive MIMO systems. Cell-free array antenna topology is one of the three typical massive MIMO array antenna topologies. In this paper, channel hardening in mm-Wave is investigated based on ray tracing simulations. Cell-free array antenna topology are evaluated in two different radio wave propagation channel conditions. The propagation scenarios are an indoor office room furnished with desks, chairs and shelters and the same, but emptied room. The analysis of the simulation results show that the channel hardening effect is more pronounced in the furnished room as compared with empty room. For the cell-free array antenna topology and at the locations near the antennas, the channel hardening effect is poor.
URSI GASS 2021, Rome, Italy, 28 August - 4 September 2021
Simulation based channel hardening of cell-free massive MIMO in mm-Wave
C. Qin(1,2), Y. Gao(1), J. Chen(1), A.A. Glazunov(2,3) and J. Zhang(1,4)
(1) Ranplan Wireless Inc., Cambridge, U.K.
(2) University of Twente, Enschede, the Netherlands
(3) Chalmers University of Technology, Gothenburg, Sweden
(4) University of Sheffield, Sheffield, U.K.
Abstract
Channel hardening is one of the key properties of massive
MIMO systems. Cell-free array antenna topology is one of
the three typical massive MIMO array antenna topologies.
In this paper, channel hardening in mm-Wave is
investigated based on ray tracing simulations. Cell-free
array antenna topology are evaluated in two different radio
wave propagation channel conditions. The propagation
scenarios are an indoor office room furnished with desks,
chairs and shelters and the same, but emptied room. The
analysis of the simulation results show that the channel
hardening effect is more pronounced in the furnished room
as compared with empty room. For the cell-free array
antenna topology and at the locations near the antennas, the
channel hardening effect is poor.
1 Introduction
Three key features of 5G networks are the massive
multiple-input multiple-output (MIMO) antenna systems,
the millimeter Wave (mm-Wave) frequency spectrum
allocation and the ultra-dense deployment of Small Cells
(SC) [1]. Massive MIMO has the potential to improve the
spectrum efficiency allowing several users to share the
same time-frequency resource by Multiuser MIMO (MU-
MIMO) techniques. Furthermore, the mm-Wave bands
offer larger chunks of available frequency bandwidths, but
also, due to the shorter wavelengths, it enables the use of
massive MIMO array antenna with many elements
mounted in the base station or access point contributing
larger spectrum efficiency. The larger bandwidth translates
directly to higher capacity and data rates [2]. The dense
deployment of SC improves the capacity per unit area by
reusing the frequency resource in a smaller coverage area.
In addition to higher spectrum efficiency, another key
advantage of massive MIMO system is higher reliability [3,
4], which is essential for applications such as remote
surgery, intelligent transportation systems and industry
automation. In a massive MIMO system, as the number of
the antennas increases, the variations of the channel gain
decrease over the time, the frequency and the spatial
domains, and the channel gain becomes concentrated
around its mean, i.e., the channel becomes more
deterministic. This phenomenon is known as channel
hardening. The study of channel hardening has mainly
focused on variations over frequency or time, due to its
impact on decreasing delay spread and the coherent
combination of signals from a large number of antennas,
respectively [5]. As to the spatial channel hardening, not as
many studies have investigated this phenomenon. In [6],
the spatial channel hardening is investigated based on
simulation results in real scenario.
In papers [3] and [4], the theory of channel hardening has
been discussed. In papers [5, 7, 8], the authors have
investigated the channel hardening phenomenon based on
measurement data. In paper [6], the spatial channel
hardening is investigate based on ray tracing channel
simulations considering different array antenna topologies
deployed in different radio wave propagation conditions.
The deployment of current cellular networks has been
mainly restricted to sub-6GHz frequencies at the µwave
bands. Although these frequency bands show favorable
propagation conditions, they can’t measure up with the key
features of mm-Wave 5G networks. As a matter of fact, the
channel propagation characteristics at mm-Wave and at
µwave are rather different. For example, propagation at
mm-Wave is mainly based on the line-of-sight (LOS) due
to the higher penetration and diffraction losses. Paper [2]
investigated six key differences between massive MIMO at
µwave and massive MIMO at mm-Wave. Many other
papers have studied the combination of massive MIMO
with mm-Wave frequency bands (see, e.g., [9, 10]).
However, to the best knowledge of the authors, no
publication has focused on the spatial channel hardening
effects of massive MIMO at the mm-Wave frequencies.
In this paper, we extend our analysis in [6] by considering
the channel hardening at mm-Wave. The simulation
analyses are based on a cell-free massive MIMO array
antennas topology deployed in two different radio wave
propagation conditions. Heat maps and the corresponding
cumulative distribution functions (CDF) of the spatial
distribution of the channel hardening effect are given based
on numerical simulations. The study considers cell-free
massive MIMO array antenna with various numbers of
antenna elements deployed in the two different propagation
scenarios. The results show the channel hardening effect
is quite different at the different locations of the room for
both the furnished room and the empty room. This is
because the high propagation loss at the mm-Wave
frequency band, which also highly differs in the non-line-
of-sight (N-LOS) and the LOS conditions.
2 Channel Hardening
Let’s assume a massive MIMO system comprising an array
antenna with M antenna elements and K single antenna
users. The channel between the M antenna elements and
the k-th user is an M×1 complex channel vector, defined by

(1)
where, and  are the large scale fading and
small scale fading channel coefficients, respectively.
denotes the matrix transpose operation. While  is
constant over multiple coherence intervals is time
varying. The channel hardening is defined as


(2)
where  is the Euclidean Norm of the channel
vector at the t time slot. var {·} and E {·} denote the
variance and the mean, respectively. The statistics are
computed over the time realizations.
The norm of equation (1) is computed as
(3)
which represents the total power received by the k-th user.
3 Simulation Scenarios
The radio propagation channel is simulated by means of the
Ranplan Professional software [12], which is a 3D ray
tracing tool. It considers the 3D building structures, the
material electric properties, as well as the radiation pattern
of the antenna [6]. It has the capability to capture the
complex channel gain considering large-scale fading and
small-scale fading of a real 3D scenario. It supports
propagation channel simulation up to 70 GHz frequency
bands, which has been tested and validated in the literature
[13]. The output of this propagation engine includes the
complex path gain and the power delay profiles (PDP),
from which the channel transfer function can be calculated
based on the channel model specification defined by 3GPP
[14].
The simulation results are used to estimate the spatial
distribution of the channel hardening effect at different
locations in two scenarios. One is an office room furnished
with desks, chairs, and shelters, the other is an empty room
with the same size walls layout as the furnished room (see
Figure 1). The cell free massive MIMO array antenna
topology with different numbers of antennas are simulated.
The parameters of the simulation scenarios are listed in
Table 1.
Table 1. Simulation scenario parameters
Simulator tool
Ranplan Professional
Propagation model
3D ray tracing
Size of scenario
50×120×3
Height of desks
90 cm
Height of chairs
90 cm
Height of shelters
1.5 m
Height of Tx antennas
2.4 m
Frequency
26 GHz
Number of Tx antennas (M)
4, 16, 32, 64
Array antenna topologies
Cell-free
Antenna type
Isotropic
Minimum antenna distance
5.1m
Maximum antenna distance
50.0m
Height of users
1.0 m
Simulation resolution
0.5 m
Number of snapshots
10,000
Cell-free massive MIMO array antenna topology with 4,
16, 32 and 64 antennas are considered. All the antennas are
distributed according to a uniform distribution along the
perimeter comprised by the outer wall of the room, which
is the same for the two scenarios. Figure 1. (a) and (b) show
a 3D view of 64 antennas deployed in the furnished and the
empty rooms, respectively.
(a) 64 cell-free antennas deployed in the furnished room
(b) 64 cell-free antennas deployed in the empty room
Figure 1. 3D view of the scenarios and antennas
4 Numerical Results
2D heat maps representing the spatial distribution of the
channel hardening effect in the two defined scenarios with
16 and 64 antennas are shown in Figure 2. As can be seen
from the heat map, increasing the number of antennas, the
channel hardening effects are enhanced in both the
furnished and the empty room. For both scenarios, the
channel hardening effect in the areas closer to walls is
much poorer than that in the middle area. The poor channel
hardening area in the empty room distributed evenly, while
that in the furnished room is highly correlated to the interior
structure of the room. In the area that near the edge of the
room the channel hardening effect is poor because the high
propagation loss of mm-Wave. The signal from the nearest
antennas has a relatively high signal strength than the
signals from the other antennas. Only a limited number of
antennas dominate in the massive MIMO system, making
it less likely to behave like a massive MIMO system.
(a)
(b)
(c)
(d)
Figure 2. Spatial distribution hardening effect (a) 16 cell-
free antennas deployed in the furnished room, (b) 16 cell-
free antennas deployed in the empty room, (c) 64 cell-free
antennas deployed in the furnished room, (d) 64 cell-free
antennas deployed in the empty room.
Figure 3 shows the CDFs of channel hardening effect for
the cell-free array antenna topology. As can be seen from
Figure 3, the channel hardening effect is relatively smaller
in an empty room than that in the room with furniture with
the same number of antennas. Take 64 antennas as an
example, in the furnished room, 90% of the area can get a
channel hardening effect better than -10dB. But in the
empty room, only 65% of the area can achieve that results.
Increasing the number of antennas enhanced the channel
hardening effect in both the empty and the furnished rooms.
Figure 3. CDF of channel hardening variation in dB for
cell- free antenna in 2 different scenarios.
The major findings of this paper are summarized as
follows:
1) Channel hardening can be expected for cell-free
massive MIMO in the mm-Wave frequency bands.
2) Increasing the number of massive MIMO antenna
elements increases the channel hardening effects as
expected.
3) The channel hardening effect in a furnished room is
larger than in the same, but empty, room.
4) For the cell-free array antenna topology the areas near
the antennas will have a smaller channel hardening
effect.
5 Conclusion
This paper presents a simulation-based evaluation of the
spatial distribution of the channel hardening in mm-Wave
frequency band. A cell-free array antenna topology was
evaluated in an office room with and without furniture. The
comparison shows that the furnished room have a larger
channel hardening effect. Channel hardening heatmaps
show a deeper sight of channel hardening effect in different
areas of the room, which demonstrate that the channel
hardening effect in mm-Wave frequency band has a strong
relationship with the interior structure and the antenna
location.
6 Acknowledgements
The authors would like to acknowledge the support of
WaveComBE project, under Horizon 2020 research and
innovation program with grant agreement No. 766231.
7 References
[1] T. E. Bogale and L. B. Le, "Massive MIMO and
mmWave for 5G Wireless HetNet: Potential Benefits and
Challenges," in IEEE Vehicular Technology Magazine,
vol. 11, no. 1, pp. 64-75, March 2016, doi:
10.1109/MVT.2015.2496240.
[2] Stefano Buzzi, Carmen D’Andrea. Massive MIMO 5G
Cellular Networks: mm-Wave vs. μ-Wave Frequencies[J].
ZTE Communications, 2017, 15(S1): 41-49.
[3] T. L. Marzetta, “Noncooperative cellular wireless with
unlimited numbers of base station antennas,” IEEE Trans.
Wireless Commun., vol. 9, pp. 35903600, Nov 2010.
[4] E. G. Larsson, O. Edfors, F. Tufvesson, and T. L.
Marzetta, “Massive MIMO for next generation wireless
systems,” IEEE Communications Magazine, vol. 52, no. 2,
pp. 186195, Feb 2014.
[5] S. Gunnarsson, J. Flordelis, L. Van Der Perre and F.
Tufvesson, "Channel Hardening in Massive MIMO: Model
Parameters and Experimental Assessment," in IEEE Open
Journal of the Communications Society, vol. 1, pp. 501-
512, 2020, doi: 0.1109/OJCOMS.2020.2987704.
[6] C. Qin, Y. Miao, Y. Gao, J. Chen, J. Zhang and A. A.
Glazunov, "Simulation-based Investigation on Spatial
Channel Hardening of Massive MIMO in Different Indoor
Scenarios and with Different Array Topologies," 2020
XXXIIIrd General Assembly and Scientific Symposium of
the International Union of Radio Science, Rome, Italy,
2020, pp. 1-4, doi:
10.23919/URSIGASS49373.2020.9232421.
[7] l. O. Martínez, E. De Carvalho, J. O. Nielsen, "Massive
MIMO properties based on measured channels: Channel
hardening user decor-relation and channel sparsity",
Signals Systems and Computers 2016 50th Asilomar
Conference on, pp. 1804-1808, 2016.
[8] À. O. Martínez, E. De Carvalho and J. Ø. Nielsen,
"Towards very large aperture massive MIMO: A
measurement-based study," 2014 IEEE Globecom
Workshops (GC Wkshps), Austin, TX, pp. 281-286, 2014.
[9] L. Swindlehurst, E. Ayanoglu, P. Heydari and F.
Capolino, "Millimeter-wave massive MIMO: the next
wireless revolution?," in IEEE Communications Magazine,
vol. 52, no. 9, pp. 56-62, Sep. 2014.
[10] S. Buzzi and C. D’Andrea, Doubly massive
mmWave MIMO systems: Using very large antenna arrays
at both transmitter and receiver”, 2016 IEEE Global
Telecommunications Conference (GLOBECOM 2016),
Washington DC, USA, 5-7 Dec. 2016.
[11] Y. MIAO, S. Pollin, A.A. Glazunov, “Simulation-
based Investigation on Massive Multi-Antenna System as
to Spatial Channel Hardening for Mobile Single User in a
Controlled Multipath Environment,” in Proceedings of the
14th European Conference on Antennas and Propagation,
Copenhagen, Denmark, Mar. 2020.
[12] https://ranplanwireless.com/products/ranplan-
professional/
[13] W. Yang, J. Huang, J. Zhang, S. Salous and J. Zhang,
"Indoor Measurement Based Verification of Ray
Launching Algorithm at the Ka-Band," 2020 XXXIIIrd
General Assembly and Scientific Symposium of the
International Union of Radio Science, Rome, Italy, 2020,
pp. 1-4, doi: 10.23919/URSIGASS49373.2020.9232150.
[14] 3GPP TR 38.901 V16.0.0, Technical Specification
Group Radio Access Network; Study on channel model for
frequencies from 0.5 to 100 GHz (Release 16), 2019-10.
ResearchGate has not been able to resolve any citations for this publication.
Conference Paper
Full-text available
Spatial channel hardening is one of the key properties of a massive MIMO system. In this paper, we investigate the spatial channel hardening based on ray tracing channel simulations with different array antenna topologies deployed in different radio wave propagation conditions. Investigated are 9 different indoor deployment scenarios of massive MIMO. Considered are the empty room, the rich-furniture room and the room-to-room scenarios, where massive multi-antennas are deployed in the co-located, the split and the cell-free array topologies. The simulation results show that the cell-free topology results in the most-spatially-hardened channel in target area, while the co-located antennas the least-spatially-hardened. It is also found that the convergence of channel spatial variations is consistent with the number of paths estimated in multipath scenarios, i.e., the more multipath components are estimated in a propagation scenario, the smaller the number of antennas is required for the channel to reach the spatially hardened state. In our simulations, the rom-to-room scenario leads to the most-spatially-hardened channel, while for the empty room scenario it is the opposite.
Article
Full-text available
The irruption of the cell-free (CF) massive multiple-input multiple-output (MIMO) network topology has meant taking one step further the concept of massive MIMO as a means to provide uniform service in large coverage areas. A key property of massive MIMO networks is channel hardening, by which the channel becomes deterministic when the number of antennas grows large enough relative to the number of serviced users, easing the signal processing and boosting the performance of simple precoders. However, in CF massive MIMO, the fulfillment of this condition depends on several aspects that are not considered in classical massive MIMO systems. In this work, we address the presence of channel hardening in both CF massive MIMO and the recently appeared user-centric (UC) approach, under a spatially correlated Ricean fading channel using distributed and cooperative precoding and combining schemes and different power control strategies for both the downlink (DL) and uplink (UL) segments. We show that the line-of-sight (LOS) component, spatially correlated antennas and UC schemes have an impact on how the channel hardens. In addition, we examine the existent gap between the estimated achievable rate and the true network performance when channel hardening is compromised. Exact closed-form expressions for both the hardening metric and achievable DL/UL rates are given as well.
Article
Full-text available
Reliability is becoming increasingly important for many applications envisioned for future wireless systems. A technology that could improve reliability in these systems is massive MIMO (Multiple-Input Multiple-Output). One reason for this is a phenomenon called channel hardening, which means that as the number of antennas in the system increases, the variations of channel gain decrease in both the time-and frequency domain. Our analysis of channel hardening is based on a joint comparison of theory, measurements and simulations. Data from measurement campaigns including both indoor and outdoor scenarios, as well as cylindrical and planar base station arrays, are analyzed. The simulation analysis includes a comparison with the COST 2100 channel model with its massive MIMO extension. The conclusion is that the COST 2100 model is well suited to represent real scenarios, and provides a reasonable match to actual measurements up to the uncertainty of antenna patterns and user interaction. Also, the channel hardening effect in practical massive MIMO channels is less pronounced than in complex independent and identically distributed (i.i.d.) Gaussian channels, which are often considered in theoretical work.
Article
Full-text available
Cell-Free (CF) Massive MIMO is an alternative topology for Massive MIMO networks, where a large number of single-antenna access points (APs) are distributed over the coverage area. There are no cells but all users are jointly served by the APs using network MIMO methods. In prior work, it has been claimed that CF Massive MIMO inherits the basic properties of cellular Massive MIMO, namely channel hardening and favorable propagation. In this paper, we evaluate if one can rely on these properties when having a realistic stochastic AP deployment. Our results show that the level of channel hardening depends strongly on the propagation environment and there is generally little hardening, except when the pathloss exponent is small. We further show that using 5--10 antennas per AP, instead of one, we can greatly improve the hardening. The level of favorable propagation is affected by the propagation environment and by the distance between the users, where only spatially well separated users exhibit favorable propagation. The conclusion is that we cannot rely on channel hardening and favorable propagation when analyzing and designing CF Massive MIMO networks, but we need to use achievable rate expressions and resource allocation schemes that work well in the absence of these properties. Some options are described in this paper.
Article
Full-text available
Enhanced mobile broadband (eMBB) is one of the key use-cases for the development of the new standard 5G New Radio for the next generation of mobile wireless networks. Large-scale antenna arrays, a.k.a. Massive MIMO, the usage of carrier frequencies in the range 10-100 GHz, the so-called millimeter wave (mm-wave) band, and the network densification with the introduction of small-sized cells are the three technologies that will permit implementing eMBB services and realizing the Gbit/s mobile wireless experience. This paper is focused on the massive MIMO technology; initially conceived for conventional cellular frequencies in the sub-6 GHz range (\mu-wave), the massive MIMO concept has been then progressively extended to the case in which mm-wave frequencies are used. However, due to different propagation mechanisms in urban scenarios, the resulting MIMO channel models at \mu-wave and mm-wave are radically different. Six key basic differences are pinpointed in this paper, along with the implications that they have on the architecture and algorithms of the communication transceivers and on the attainable performance in terms of reliability and multiplexing capabilities.
Conference Paper
Three prominent features of massive MIMO are studied using channel measurements. Those features are extensively exploited in signal processing methods for massive MIMO and have been only partially, or not at all, validated. First, channel hardening is characterized as a function of the number of antennas. Second, user decorrelation is evaluated as a function of the distance between users. At last, the channel angular spread, proposed as a basis for pilot contamination and frequency division duplexing operation, is assessed. The whole study is based on two measurement campaigns involving a base station with 64 antennas and 128 antennas.
Article
A Cell-Free Massive MIMO (multiple-input multiple-output) system comprises a very large number of distributed access points (APs)which simultaneously serve a much smaller number of users over the same time/frequency resources based on directly measured channel characteristics. The APs and users have only one antenna each. The APs acquire channel state information through time-division duplex operation and the reception of uplink pilot signals transmitted by the users. The APs perform multiplexing/de-multiplexing through conjugate beamforming on the downlink and matched filtering on the uplink. Closed-form expressions for individual user uplink and downlink throughputs lead to max-min power control algorithms. Max-min power control ensures uniformly good service throughout the area of coverage. A pilot assignment algorithm helps to mitigate the effects of pilot contamination, but power control is far more important in that regard. Cell-Free Massive MIMO has considerably improved performance with respect to a conventional small-cell scheme, whereby each user is served by a dedicated AP, in terms of both 95%-likely per-user throughput and immunity to shadow fading spatial correlation. Under uncorrelated shadow fading conditions, the cell-free scheme provides nearly 5-fold improvement in 95%-likely per-user throughput over the small-cell scheme, and 10-fold improvement when shadow fading is correlated.
Article
There has been active research worldwide to develop the next-generation, i.e., fifth-generation (5G), wireless network. The 5G network is expected to support a significantly large amount of mobile data traffic and a huge number of wireless connections and achieve better costand energy-efficiency as well as quality of service (QoS) in terms of communication delay, reliability, and security. To this end, the 5G wireless network should exploit the potential of new developments, including superdense and heterogeneous deployment of cells and massive antenna arrays [i.e., massive multiple-input, multiple-output (MIMO) technologies] and utilization of higher frequencies, particularly millimeter-wave (mmWave) frequencies. This article discusses the potential benefits and challenges of the 5G wireless heterogeneous network (HetNet) that incorporates massive MIMO and mmWave technologies.