Recent trends and future directions in millimeter wave massive MIMO channel characterization at 30/60/140 GHz: a review

Abstract

Currently, 5G technology is being implemented worldwide to meet the increasing demands of users. However, it is unable to fully achieve the projected growth in data traffic and high-end service quality for emerging applications and intelligent terminal devices. As a result, researchers are exploring other promising technologies and high-frequency bands for the next generation of cellular technology, known as beyond 5G (B5G) or 6G. Millimeter wave (mmWave) communications and Massive Multi-Input-Multi-Output (mMIMO) technology are receiving significant attention from the research community due to the availability of an extremely large frequency spectrum (30–300 GHz) and the ability of mMIMO to reduce propagation and absorption losses. This study provides a comprehensive survey of current achievements in mMIMO-mmWave communication systems, highlighting their benefits, challenges, and proposed solutions. It also describes the channel characterizations, recommended standards, and existing channel models at the 30/60/140 GHz bands, and presents a state-of-the-art comparison in terms of performance measures, parametric specifications, propagation losses, and methodologies for the measurement setup in different propagation environments. Additionally, the paper discusses the importance of beamforming and the challenges in realizing mMIMO communication systems.

Full text

Telecommunication Systems (2025) 88:45
https://doi.org/10.1007/s11235-025-01275-1
Recent trends and future directions in millimeter wave massive MIMO
channel characterization at 30/60/140 GHz: a review
Manish Sharma1,2 ·Anand Agrawal1·Chandraveer Singh1,3 ·Chetna Sharma1
Accepted: 25 February 2025
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025
Abstract
Currently, 5G technology is being implemented worldwide to meet the increasing demands of users. However, it is unable to
fully achieve the projected growth in data traffic and high-end service quality for emerging applications and intelligent terminal
devices. As a result, researchers are exploring other promising technologies and high-frequency bands for the next generation of
cellular technology, known as beyond 5G (B5G) or 6G. Millimeter wave (mmWave) communications and Massive Multi-Input-
Multi-Output (mMIMO) technology are receiving significant attention from the research community due to the availability
of an extremely large frequency spectrum (30–300 GHz) and the ability of mMIMO to reduce propagation and absorption
losses. This study provides a comprehensive survey of current achievements in mMIMO-mmWave communication systems,
highlighting their benefits, challenges, and proposed solutions. It also describes the channel characterizations, recommended
standards, and existing channel models at the 30/60/140 GHz bands, and presents a state-of-the-art comparison in terms
of performance measures, parametric specifications, propagation losses, and methodologies for the measurement setup in
different propagation environments. Additionally, the paper discusses the importance of beamforming and the challenges in
realizing mMIMO communication systems.
Keywords Massive MIMO ·Beamforming ·mmWave channel characterization ·Standards ·Channel models
1 Introduction
Millimeter wave (mmWave) massive multiple-input multiple-
output (MIMO) systems have emerged as a promising
technology for enabling high data rate wireless communi-
cation in future 5G and B5G networks [1]. The extremely
high frequency (EHF) band is a term that can be used inter-
changeably with the mmWave band, and it has a frequency
range of 30 to 300 GHz. In the electromagnetic spectrum, this
frequency range falls between the microwave and infrared
regions as shown in Fig. 1, offer a large amount of spectrum
that can be used to support high data rate communication.
BManish Sharma
2021krec2003@iiitkota.ac.in; manish.sharma@poornima.org
1Department of Electronics and Communication Engineering,
Indian Institute of Information Technology, Kota, Rajasthan,
India
2Department of Electronics and Communication Engineering,
Poornima College of Engineering, Jaipur, Rajasthan, India
3School of Automation, Banasthali Vidyapith, Tonk,
Rajasthan, India
However, the deployment of mmWave massive MIMO sys-
tems requires a thorough understanding of the characteristics
of the mmWave channels, which are significantly different
from those at lower frequencies [2].
MmWave channels are highly directional, suffer from sig-
nificant path loss, and are highly susceptible to blockage by
obstacles. These characteristics pose significant challenges
for the deployment of mmWave systems, but they also offer
opportunities for exploiting the spatial dimension of the
channel to achieve high data rates using massive MIMO
techniques [3–5]. To Unlock the full connectivity poten-
tial of mmWave technology, a lot of research activities have
been carried out around the world and identify the benefits,
challenges and potential solutions. The spectrum is further
categories into the several sub-bands shown in Table 1and
have conducted thorough investigations to understand the
propagation characteristics, opportunities, and limitations of
each sub-band [6]. Some of the commonly used subbands of
mmWave spectrum include:
•24.25–27.5 GHz: This subband is suitable for satellite
communications, remote sensing, and radar applications.
0123456789().: V,-vol 123

45 Page 2 of 22 M. Sharma et al.
Fig. 1 Millimeter Wave Spectrum with relevant regions [3]
Table 1 mmWave frequency bands [7]
Frequency bands Frequency range (GHz)
Q-band 30–50
U-band 40–60
V-band 50–75
E-band 60–90
W-band 75–110
F-band 90–140
D-band 110–170
G-band 110–300
•28–29.5 GHz: This subband is often used for wireless
backhaul, fixed wireless access, and 5 G cellular net-
works.
•37–40 GHz: This subband is suitable for 5 G cellular
networks, wireless backhaul, and point-to-point commu-
nication applications.
•60 GHz: This subband is often used for high-speed, short-
range communication applications such as WiGig.
•110–170 GHz: This subband has a range of applications
across various industries due to its high bandwidth and
short wavelengths such as used for high-capacity wire-
less communication links, particularly in backhaul and
fronthaul networks for mobile operators.
In addition to spectrum characterization, various measure-
ment setups and parametric specifications have been pub-
lished in reputed articles to enable accurate and reliable
mmWave channel measurements. These measurement setups
typically involve the use of advanced channel sounders, such
as vector network analyzers (VNAs) and channel sounders
based on synthetic aperture radar (SAR), and are designed to
capture critical channel parameters such as path loss, delay
spread (DS), and angular spread (AS).
The article [8] describes an experimental study of the per-
formance of a virtual antenna array at frequencies between 27
and 29 GHz. The study used a VNA to measure the channel
behavior and performance in various environmental scenar-
ios. The authors computed several performance measures,
including the power delay profile (PDP), angular spread, and
delay spread. The results of the study showed that the size
of indoor scenarios had a significant impact on the channel
characteristics. The authors observed that larger indoor envi-
ronments produced a more complex and variable channel
behavior, with greater angular spread and delay spread.
The article [9] describes a study in which the authors
estimate the parameters of a large-scale antenna array sys-
tem operating at frequencies between 28 and 30 GHz. The
study uses a Maximum Likelihood Estimator (MLE) algo-
rithm to estimate the channel parameters in both Line of Sight
(LOS) and Obstructed Line of Sight (OLOS) scenarios. The
authors compare the estimated Power Azimuth Delay Pro-
file (PADP) obtained from the uniform circular array (UCA)
and rotational horn antenna measurements, and found that
the majority of the channel components were successfully
captured. The article [10] presents a 3D non-stationary 5G
massive MIMO small-scale fading channel model tailored
for various high-speed train (HST) scenarios. The model
incorporates scattering effects from overhead line poles and
categorizes environmental clusters into five types, enabling
the analysis of environmental conditions (ECs) on channel
statistical properties. The article [11] presents a novel A-S-T-
F non-stationary multi-UAV cooperative channel model for
6G massive MIMO mmWave systems, accounting for 3D
arbitrary Unmanned Aerial Vehicle (UAV) trajectories and
self-rotations. Additionally, Simulated results align closely
with ray-tracing data, validating the model’s effectiveness.
The article [12] presents a 3-D statistical channel impulse
response (CIR) model for urban LOS and Non-Line of Sight
(N-LOS) channel scenarios with arbitrary carrier frequency,
123

Recent trends and future directions in millimeter wave massive MIMO... Page 3 of 22 45
signal bandwidth, and antenna beam-width in the 28-73 GHz
bands. The study uses this measurement setup to demonstrate
the 3GPP model. The 3-D statistical CIR model presented in
the study can be useful for the design and optimization of
mmWave communication systems in urban environments.
The study in [13,14] highlights the potential of unlicensed
60 GHz spectrum in mmWave technology to achieve signif-
icantly higher data rates due to the large bandwidth when
employing an air interface. The authors compare this with
existing Wireless Local Area Networks (WLANs).
The article [15] reports on a study of the 60 GHz mmWave
channel measurements in both LOS and NLOS scenarios in
an office environment. The study investigated the angular
characteristics of the channel in both azimuth and elevation
domains. The authors observed that the estimated azimuth
spread (AS) was larger than the elevation spread (ES) for both
cluster and global levels. This means that the angular spread
of the received signal is larger in the horizontal direction than
in the vertical direction.
The article [16] describes a study of the power decay
characteristics of mmWave channels in an indoor office envi-
ronment at a frequency of 60 GHz. The study found that the
power decay rate of clusters was consistent with the Extended
Saleh-Valenzuela (S-V) model, which suggests that powers
may shift significantly inside a cluster. The Extended S-V
model is a widely used statistical channel model for mmWave
channels that takes into account the clustering of multipath
components. The authors also observed that the power decay
rate of rays within a cluster was faster than that of the cluster
as a whole. This indicates that the power is concentrated in
a smaller angular region within the cluster.
The article [17] presents a study of the diffraction effects in
mmWave channel propagation. The authors performed mea-
surements of diffraction at an edge, at a wedge, and at a
cylinder using ray tracing simulation and found that the use
of an antenna array can improve the accuracy of channel
measurements compared to a single antenna. Additionally,
the study found that the diffraction effects can be reduced
by using transparent dielectric material. This is because the
dielectric material helps to smooth out the edges and cor-
ners of objects, reducing the scattering and diffraction of the
mmWave signal.
In [18], the authors have reported that the optimum MIMO
polarimetrics filtering helps to reduce the shadowing and
heavy clutter due to human activity for mmWave radar. This is
an important observation, as human activity can cause signif-
icant interference and impair the performance of mmWave
radar systems. The 164-200 GHz frequency band is often
referred to as the water vapor absorption band. This is because
water vapor in the Earth’s atmosphere can absorb electro-
magnetic waves in this frequency range, which can limit the
propagation of signals through the atmosphere. To overcome
the limitations imposed by the water vapor absorption band,
various techniques have been developed.
In [19,20], authors have conducted a radio channel mea-
surement using channel sounding in the D-band frequency
range (110 to 170 GHz) and analyzed the effects of different
materials on the signal propagation in indoor environments.
In [21], authors have measured the wide band spatial channel
between 130 GHz and 143 GHz in an indoor environment and
found that the use of MIMO technology has the potential to
improve the data rates and connection range of wireless com-
munication systems operating in this frequency band. The
mmWave spectrum experiences significant path and environ-
ment losses due to its high carrier frequency, which makes it
susceptible to attenuation, absorption, and scattering. These
losses can be mitigated by employing MIMO techniques.
The use of a large number of antennas in mmWave-MIMO
systems can also improve the beamforming capabilities and
reduce the impact of blockages. Recent advancements in
mmWave MIMO systems include [22–25].
•Antenna design The design of the antenna array plays a
crucial role in mmWave MIMO systems. Antenna arrays
with a large number of elements and high gain have been
proposed to achieve high data rates and spectral effi-
ciency. The use of hybrid analog-digital beamforming
can also reduce the hardware complexity and power con-
sumption of fully digital beamforming [26].
•Channel modeling Channel modeling is essential for
designing and evaluating mmWave MIMO systems. Sev-
eral channel models have been proposed, such as the
3GPP spatial channel model (SCM), the geometry-based
stochastic channel model (GSCM), and the ray tracing-
based channel model. These models can be used to
simulate the wireless channel and optimize the system
performance [27].
•Beamforming and precoding Beamforming and precod-
ing techniques can improve the system performance by
exploiting the spatial diversity of the wireless chan-
nel. Several low-complexity beamforming and precoding
techniques have been proposed, such as zero-forcing
(ZF) precoding and singular value decomposition (SVD)-
based precoding [28].
•Intelligent reflecting surfaces (IRS) IRS is a promis-
ing technology that can enhance the performance of
mmWave MIMO systems by manipulating the propaga-
tion environment [29]. IRS can reflect the incident signal
and change its direction, amplitude, and phase to enhance
the signal quality and reduce the path loss [30,31]. The
performance of the system is greatly impacted by various
RIS deployment options because of the complexity of the
indoor environment [32].
•Multiuser MIMO Multiuser MIMO is a key technique
for improving the spectral efficiency of mmWave MIMO
123

45 Page 4 of 22 M. Sharma et al.
systems. The use of beamforming and precoding can
manage the interference and improve the system capacity
[33–35].
A concise summary of the pertinent survey is available
in Table 2, which includes the key parameters for the
30/60/140 GHz frequency bands. This table summarizes a
literature survey on parametric data related to various central
frequencies in the range of 30/60/140 GHz. The parame-
ters include Path Loss (PL), Power Delay Profile (PDP),
Signal-to-Noise Ratio (SNR), Angle of Arrival/Departure
(AoA/AoD), Channel Impulse Response (CIR), and Delay
Spread (DS). The majority of studies focus on the 60 GHz
band, reflecting its relevance in wireless communications.
Path loss widely measured across studies, especially at 28
GHz, 60 GHz, and the 60–300 GHz range. PDP frequently
addressed at 60 GHz and certain studies at 28 GHz and higher
bands (e.g., 73 GHz). SNR analysis is sparse, with fewer stud-
ies covering this parameter at 60 GHz. AoA/AoD Addressed
in a subset of works, primarily at 28 GHz and 60 GHz. CIR
studies are consistent at lower bands like 28 GHz but sparser
at higher frequencies. DS Commonly studied in the 60 GHz
range, with limited attention in other bands. References [13,
16], and [36] cover the majority of parameters, providing
holistic insights. References [37,38] focus on specific param-
eters, leaving gaps in other areas. Research on frequencies
above 60 GHz remains limited, with sporadic data across key
parameters.
1.1 Contribution of the survey
This survey paper covers the latest research findings in
channel modeling, measurements, and analysis in mmWave
MIMO systems at different frequencies, including 30 GHz,
60 GHz, and 140 GHz. The paper also identifies the chal-
lenges and research directions in mmWave MIMO channel
characterization and provides insights for future research in
this field. The survey takes into account the regulatory issues,
severe propagation and absorption losses, and inherent prop-
agation characteristics that are associated with mmWave
communication. The contribution of this survey paper are
summarized here:
•This survey offers a comprehensive review and compar-
ative analysis of various IEEE standards that are suitable
for the 30/60/140 GHz frequency bands.
•This survey highlights the limitations of mmWave com-
munication, including severe propagation, penetration
losses, poor coverage, inherent propagation character-
istics, and the need for parametric specifications that
corresponds to various indoor environments.
•This survey presents various measurement methodolo-
gies and compares different performance measures,
including delay spread, angular spread, PDP, and more.
Additionally, the survey examines different channel mod-
eling schemes, their benefits, and limitations.
•This survey includes simulation results of mmWave prop-
agation characteristics at the 60 GHz band for both
residential and office indoor environments.
•This survey presents various beamforming schemes and
their role in implementing MIMO mmWave communi-
cation systems.
•Finally, this paper identifies the benefits, challenges, and
potential solutions for implementing mMIMO mmWave
communication systems.
1.2 Organization of the paper
The organizing structure for this survey paper is presented
in Fig. 2. The organization of this paper is as follows:
Sect. 2provides an overview of various IEEE standards
and channel modeling schemes. Section 3discusses chan-
nel propagation characteristics and measurement setups, as
well as highlights beamforming architecture and paramet-
ric specifications. Section 4explores different beamforming
applications for MIMO systems. Section 5presents simulated
results and compares various performance outcomes. Sec-
tion 6briefly discusses challenges and promising solutions.
Finally, the paper concludes with a summary of potential
solutions for channel issues and future research directions in
Sect. 7.
1.3 Notation
Boldface upper (A) and lower (a) case letters denote matrices
and vectors respectively. (.)∗,(.)Tand tr(.)are denote the
Hermitian transpose, transpose and trace operators, respec-
tively. |.|and .denote the determinant operator and norm
of a vector, respectively. Cr∗tdenote the size of matrix r×t.
2 Standardization and channel modelling
The architecture of mmWave communication is being devel-
oped to support various bandwidth-intensive applications,
such as D2D communication, imaging, sensing, and terres-
trial communication [46].
Figure 3illustrates the application scenarios of mmWave
communication networks. Each of these application services
has its own set of technology trends, challenges, and frame-
works. In this, a multi-tier cellular heterogeneous network
(HetNet) with massive MIMO, microwave, and mmWave
communication capabilities in both macrocell and small cell
base stations. A split control and data plane framework
ensures that efficient mobility and other control-related sig-
nalling are handled by the long-range microwave massive
123

Recent trends and future directions in millimeter wave massive MIMO... Page 5 of 22 45
Table 2 Summary of literatures survey
References Central frequency (GHz) Path loss (PL) Power delay profile (PDP) Signal-to- Noise Ratio (SNR) AoA/ AoD Channel impulse response (CIR) Delayspread(DS)
[3]60 ××× ×
[8]28  × 
[9] 28–30  ××× 
[12] 28 and 73 ×× 
[13] 28 and 38    
[15]60  ××
[16]60   ×
[17] 60 and 300 ×× ×
[18]60 ×× ×  ×
[19] 110–170 GHz  ×××
[39]60  ××× 
[40]60 ×× × ×
[41] 60-75 ×× 
[37] 71-86 ×× ×× ×
[42]60 ×××
[36] 60 and 300    
[43]60 ××× ×
[38]60 ×× ×× ×
[44]60 ×× ×
[45]73 ×× ××
—Data shown by author, ×—Data is missing
123

45 Page 6 of 22 M. Sharma et al.
Fig. 2 Organizing structure of
this survey
Fig. 3 Application scenarios for
mmWave wireless network [4]
MIMO macrocell base station, while high-capacity mmWave
massive MIMO small cells handle data signals. When users
are closer together, dual-mode or dual-band small cells pro-
vide mmWave access lines, while those who are farther away
receive microWave frequency band. By using this method,
the "cell breathing" idea present in conventional networks is
successfully emulated by creating a dynamic cell structure.
In-band backhauling minimizes costs and maximizes spec-
trum utilization by use the same mmWave frequency band for
both access and backhaul lines. Microwave macrocell base
stations are used in vehicular communication to accommo-
date users driving at high speeds and long distances. Users of
virtual cells are not limited to using the closest base station;
instead, they can choose their preferred base station, espe-
cially if they are close to the boundary of a coverage region.
This is in contrast to the conventional method of defining
coverage zones for user association and handover. There are
a lot of possible situations, use cases, and applications that
can be made with this architecture [4].
In recent years, The IEEE has proposed standards for
mmWave communication systems, including 802.11ad,
802.11ay, 802.15.3c, and 802.16 to achieve interoperability
among the services across different regions. These stan-
dards have different channel characteristics, which can affect
the performance of MIMO systems. To accurately model
these channels, different channel models have been proposed,
including the geometric channel model, the S-V model, and
the 3GPP spatial channel model [47]. Each model has its own
benefits and limitations, and researchers have used them to
study different aspects of mmWave communication, such as
delay spread, angular spread, and path loss [48–51].
123

Recent trends and future directions in millimeter wave massive MIMO... Page 7 of 22 45
2.1 IEEE standards
Here are some of the popular IEEE standards for mmWave
systems that are discussed in this survey [51]:
2.1.1 IEEE802.15.3c
IEEE 802.15.3c is a wireless standard for high data rate com-
munications in personal area networks (PANs). It operates in
the 60 GHz frequency band and provides data rates up to 5
Gbps over short distances, typically within a room. This stan-
dard is designed to support multimedia applications, such as
high-definition video streaming, wireless docking, and wire-
less display. It uses beamforming and directional antennas to
overcome the challenges of mmWave communication, such
as severe path loss and interference from other devices oper-
ating in the same frequency band [52–54].
2.1.2 IEEE802.11ad
IEEE 802.11ad is a wireless communication standard for
multi-gigabit data transfer in the 60 GHz band. It was
designed to provide high-speed wireless connectivity for
devices such as laptops, smartphones, and tablets. The stan-
dard utilizes beamforming technology to improve the SNR
and provide a more stable connection. It can support data
rates of up to 7 Gbps, making it ideal for applications that
require high bandwidth such as streaming video, virtual
reality, and gaming. However, it has limited range and is
susceptible to interference from objects such as walls and
furniture, making it more suitable for indoor environments
[55].
2.1.3 IEEE802.11ay
This is a new standard under development for WLANs oper-
ating in the 60 GHz frequency band, and it aims to support
data rates of up to 100 Gbps. The standard supports both
line-of-sight (LOS) and non-line-of-sight (NLOS) commu-
nication and offers a range of up to 300ms in outdoor
environments. It also offers a backward compatibility mode
with the previous IEEE 802.11ad standard [56–60].
2.1.4 WiGig
WiGig is a wireless technology standard developed by the
WiGig Alliance, which was later merged with the Wi-Fi
Alliance. It operates in the unlicensed 60 GHz frequency
band and supports high-speed wireless data transfer rates up
to 7 Gbps, making it suitable for a wide range of applications
[61,62].
2.1.5 WirelessHD
WirelessHD is a wireless high-definition multimedia inter-
face (HDMI) standard that operates in the unlicensed 60
GHz frequency band. It was introduced by the WirelessHD
Consortium in 2008 as a solution for wirelessly transmit-
ting uncompressed high-definition audio and video signals
between devices. The standard supports data rates up to 28
Gbps [63].
2.1.6 IEEE 802.15.4g
This standard is used for wireless sensor networks (WSNs)
operating in the 60 GHz frequency band, and it supports data
rates of up to 500 Mbps.
2.1.7 IEEE 802.16
This standard is used for broadband wireless access (BWA)
systems, and it includes a variant for the 60 GHz frequency
band called IEEE 802.16d.
2.1.8 IEEE 802.19.1
This standard provides coexistence mechanisms for wireless
networks operating in the same frequency band, including
mmWave systems.
2.1.9 5GNR (new radio)
5G NR is the worldwide standard for 5G networks created
by 3GPP to handle a wide range of applications, including
Enhanced Mobile Broadband (eMBB), Ultra-Reliable Low-
Latency Communication (URLLC), and Massive Machine-
Type Communication (mMTC). It provides strong connec-
tion, ultra-low latency, and large data rates while operating
in the mmWave and Sub-6 GHz bands. Scalable numerol-
ogy, massive MIMO, advanced beamforming, and network
slicing are important aspects that allow for efficiency and
flexibility in a variety of use cases.
The brief comparison among the various IEEE standards
for the high data rate indoor wireless communication is given
Table 3.
2.2 Losses and attenuation at mmWave
At mmWave frequencies, electromagnetic waves behave dif-
ferently than at lower frequencies, which has implications for
losses and attenuation. Here are some key factors that affect
losses and attenuation at mmWave.
123

45 Page 8 of 22 M. Sharma et al.
Table 3 Comparison among the various IEEE standards for the high data indoor communication [56–60]
Parameters IEEE802.15.3c IEEE802.11ad IEEE802.11ay
Technology Gi-Fi Wi-Fi Wi-Fi
Operating frequency 57–64 GHz 57–70 GHz 60 GHz
Data transfer speed 5 Gbps 11 Mbps–7 Gbps 100 Gbps
Signal range 10 m–100 m 100 m 300–500 m
Antenna technology Beamforming, MIMO Beamforming MIMO, Beamforming
Modulation techniques OFDM OFDM, BPSK, DPSK 256-QAM BPSK, QPSk, OFDM
2.2.1 Path loss
Path loss refers to the reduction in signal power as it propa-
gates through the air over a given distance. The path loss at
mmWave frequencies can be modeled using the free space
path loss model, which assumes that the signal propagates
through free space without obstacles. However, in real-
ity, mmWave signals can be affected by obstacles such as
buildings, trees, and even human bodies, which can cause
additional attenuation due to reflection, diffraction, and scat-
tering.
PL[dB]=Al og10 (d/1m)+B+Xσ.(1)
where Aand Bare the slope and the interception, respec-
tively, dis the distance between transmitter and receiver, Xσ
is denotes the shadow fading which can be expressed as Gaus-
sian random variable with zero mean and standard deviation
σSF. Path loss at sub THz, 60 GHz and 30 GHz are 80 dB/m,
54 dB/m and 6 dB/m, respectively [64,65].
2.2.2 Atmospheric attenuation
Atmospheric attenuation refers to the loss of signal strength
due to the absorption, scattering, and reflection of electro-
magnetic waves by the atmosphere as they propagate through
it. Atmospheric attenuation is an important factor to con-
sider in wireless communication systems, especially at higher
frequencies such as mmWave frequencies, where the atten-
uation can be more significant [66]. The attenuation due to
the atmosphere can be divided into two main components:
gaseous attenuation and rain attenuation. Gaseous attenua-
tion is caused by the absorption of the electromagnetic waves
by atmospheric gases, mainly oxygen and water vapor. The
absorption is strongest at frequencies around 60 GHz, where
oxygen molecules have a resonant absorption peak shown in
Fig. 4. At higher frequencies, water vapor becomes the dom-
inant contributor to gaseous attenuation. Rain attenuation is
caused by the scattering and absorption of the electromag-
netic waves by raindrops [67].
Fig. 4 Atmospheric attenuation at different frequency bands 30/60/140
GHz [66]
2.2.3 Penetration loss and blockage effects
Penetration loss refers to the reduction in signal strength
when the signal passes through a material, such as a wall
or a building. The amount of penetration loss depends on
the frequency of the signal and the properties of the mate-
rial. At mmWave frequencies, the penetration loss can be
significant due to the high absorption and scattering of the
signals by building materials such as concrete, metal, and
glass. Blockage effects refer to the obstruction of the signal
path by obstacles such as buildings, trees, and even human
bodies. The blockage effect is more significant at mmWave
frequencies due to the shorter wavelengths of the signals and
the higher sensitivity of the communication systems to small
changes in the signal power and can be expressed as [5,68].
PLoS (θ)=1
1+a(exp (−b(θ−a))) .(2)
Where θis the Elevation angle, aand bare the Modeling
parameters.
123

Recent trends and future directions in millimeter wave massive MIMO... Page 9 of 22 45
2.2.4 Polarization loss
This refers to the loss of signal strength that occurs when
the signal is transmitted and received with different polar-
izations. Polarization loss can be significant at mmWave
frequencies due to the short wavelengths of the signals.
2.3 Channel models
Channel measurement and characterization models are essen-
tial for mmWave communication because they enable us to
understand the behavior of the channel, optimize antenna
arrays, mitigate blockages, and develop signal processing
techniques. Accurate channel measurement and characteri-
zation are critical to address the unique challenges posed by
the mmWave frequency band and optimize the performance
of mmWave communication systems [69]. There are several
proposed channel models for mmWave communication, each
with its advantages and disadvantages. Here are some of the
commonly used channel models:
2.3.1 Path loss channel measurement model
Path loss channel measurement models are used to estimate
the path loss in a given environment. These models are based
on measurements of the received signal strength at various
distances from the transmitter. The measured data is then
used to derive a mathematical expression that describes the
path loss is given below [6,70].
PL(f,d)[dB]=PL (f,d0)+10n log10 (d/d0)+Xσ.
(3)
Where PL(f,d)is denotes the path loss at frequency f,
PL(f,d0)is the reference path loss at frequency fand a
reference distance d0,nis the path loss exponent, which is
a measure of how quickly the signal power decays with dis-
tance. Apart from the several benefits, path loss channel mea-
surement models for mmWave communication have some
additional limitations compared to traditional RF systems
due to the unique characteristics of mmWave propagation.
These limitations include highly directional propagation,
sensitivity to antenna orientation, dynamic nature of the chan-
nel, limited penetration through objects, and limited range. It
is important to consider these factors when selecting a chan-
nel measurement and characterization model for mmWave
communication systems. The summary of path loss parame-
ters for different environments is shown in Table 4.
2.3.2 Delay and angular dispersion model
The Delay and Angular Dispersion (DAD) model is a channel
characterization model that is commonly used for mmWave
communication systems. The model takes into account the
spatial and temporal dispersion of the channel, which is
important for accurately characterizing the propagation of
mmWave signals. The DAD model represents the channel as
a series of clusters, each of which contains multiple propa-
gation paths with similar delay and angular characteristics.
Each cluster is characterized by a delay spread parameter
and an angular spread parameter, which represent the range
of delays (στ), and angles (σθ), over which the paths in the
cluster are distributed. The DAD model can be represented
mathematically as follows:
h(t,θ)=
M

m=1
R

r=1
αm,rδt−τm,raθ−θm,r.(4)
Where h(t,θ)is denotes the channel response at time tand
angle θ,Mis the number of clusters in the channel, Ris the
number of rays in each cluster, αm,ris the complex gain of
ray rin cluster m,τm,ris the delay of ray rin cluster mand
aθ−θm,ris the steering vector for the angle of arrival θ
of ray rin cluster m.
While the DAD model is a powerful channel characteriza-
tion tool for mmWave communication systems, it does have
some limitations includes Limited spatial resolution, Nar-
rowband assumption, Limited applicability to mobile envi-
ronments, Limited applicability to non-line-of-sight (NLOS)
scenarios and Inaccurate modeling of polarization. These
limitations may impact the ability of the DAD model to
accurately and reliably characterize the mmWave channel in
certain scenarios. Therefore, these limitations must be taken
into account when applying it to real-world systems.
2.3.3 Saleh-Valenzuela (S-V) model
The Saleh-Valenzuela (S-V) model is a mathematical model
used in wireless communications to analyze the behavior
of a wireless communication channel. The model is named
after its inventors, Professor M. Saleh and R. A. Valenzuela.
The S-V model takes into account the unique propagation
characteristics of mmWave frequencies, which include high
path loss, attenuation, and blockage due to obstacles in
the environment. The model also considers the effects of
antenna directivity and beamforming, which are essential
for mmWave communication due to the directional nature of
these frequencies. In the S-V model, the channel is divided
into clusters, which represent groups of paths that have simi-
lar characteristics, and rays, which represent individual paths
within each cluster. The mathematical expression for the S-V
model can be represented as follows [71]:
123

45 Page 10 of 22 M. Sharma et al.
Table 4 Summarized the parameters of the Path Loss model across three distinct indoor scenarios [6,70]
Scenarios Intercept point (PL0)[dB] Path loss exponent (n) Standard deviation(Xσ)
Indoor 1 Scenario (Laboratory ) 79.6 1.45 0.47
Indoor 2 Scenario (Conference room ) 77.4 1.93 0.6
Indoor 3 Scenario (Office ) 77 1.91 0.25
h(t,θ,φ)
M

m=1
R

r=1αm,rδt−τm,re−j2π(fCt+φm,r)
Gtθm,rGrφm,r.
(5)
Where h(t,θ,φ) is the complex impulse response of the
mmWave channel at time t and angles θand φ.Cis the speed
of light, φm,ris the phase shift of the rth rayinthemth cluster,
Gtθm,rand Grφm,rare the antenna gain pattern of the
transmitter and receiver respectively.
2.3.4 IEEE 802.15.3c channel model
The IEEE 802.15.3c channel model is designed to provide
a realistic representation of the mmWave channel for com-
munication system design and evaluation. The model is the
modified version of S-V indoor channel model and it is
based on empirical measurements and statistical modeling.
The model includes the complex propagation characteristics
of mmWave signals, effects of multipath fading, polariza-
tion, and clustering. This model allows the mmWave channel
to be represented as a series of clusters and paths, each
with its own delay, gain, and polarization characteristics as
depicts in Fig. 5. The model can be used to simulate mmWave
communication channels and evaluate the performance of
communication systems operating in these frequencies.
The mathematical expression for the this model can be
represented as follows [72]:
h(t,θ,φ
)
=βδ (t)δθδφ +
M

m=1
R

r=1αr,mδt−Tm −τr,m
×δθ−θm−ωr,mδφ−φm−ψr,m.
(6)
Where βis the Gain coefficient of the LOS path, which is
dependent on the antenna gains, the antenna heights, the Tx-
Rx separation and the path loss of the first impulse. θmand φm
are the AoA and AoD of the mth cluster, respectively. Simi-
larly, ωr,mand r,mare the AoA and AoD for the rth ray of
the mth cluster, respectively. Table 5present the comparison
of mmWave channel models for indoor environments.
Fig. 5 Graphical representation of the IEEE802.15.3c channel model
[72]
2.3.5 Existing channel simulators for mmWave MIMO
systems model
Channel simulators for mmWave MIMO systems are essen-
tial tools for evaluating wireless communication systems in
diverse scenarios. These simulators replicate the propagation
characteristics of wireless channels, such as path loss, multi-
path fading, beamforming, and spatial correlation, providing
a realistic environment for system design and performance
analysis. Table 6present the existing channel simulators for
mmWave MIMO systems. These simulators form the back-
bone of mmWave and MIMO research, enabling reliable
evaluations and optimizations for next-generation commu-
nication systems.
3 Channel characterization methods and
measurement set-up
Characterization and measurement set-up for mmWave refers
to the process of accurately measuring and characterizing
the mmWave wireless channel. This is important because
mmWave frequencies have unique propagation characteris-
tics that can impact the performance of wireless communica-
123

Recent trends and future directions in millimeter wave massive MIMO... Page 11 of 22 45
Table 5 Comparison of
mmWave channel models for
indoor environments [73,74]
Channel model Hardware complexity Cost implications Scalability
Path loss model Low Moderate High
Delay & Angular dispersion model Medium to high High Moderate
Saleh-Valenzuela (S-V) Model High High Moderate
IEEE 802.15.3c Model High High High
Table 6 Comparison of existing channel simulators for mmWave MIMO systems
Simulator Frequency range MIMO support Mobility Environment support
3GPP TR 38.901 Up to 100 GHz Yes Yes Urban, rural, indoor
NYUSIM Up to 73 GHz Yes Limited Urban, indoor, outdoor
WINNER II Sub-6 GHz to mmWave Yes Yes Urban, rural, suburban
Wireless InSite Up to 300 GHz Yes Limited Urban, vehicular, indoor
MATLAB 5G Toolbox Sub-6 GHz to mmWave Yes Yes Customizable, environments
Remcom InSite Sub-THz Yes Yes Complex urban and dense
tion systems. The characterization and measurement set-up
for mmWave signals are discussed here:
3.1 Channel characterization
The characterization of mmWave signals involves the analy-
sis of the CIR, PDP, and spatial correlation function. These
parameters provide important information about the propaga-
tion characteristics of mmWave signals, including the path
loss, delay spread, and spatial correlation. The CIR is the
response of the channel to a short pulse signal and represents
the time-varying signal amplitude and phase. The PDP rep-
resents the power distribution of the channel as a function of
time delay and provides information about the delay spread
and multipath components. The spatial correlation function
represents the spatial dependence of the channel and provides
information about the angular spread and spatial diversity.
3.1.1 Statistical analysis of 28 GHz channel
The statistical analysis plays a crucial role in understanding
the behavior of the 28 GHz channel and designing wire-
less communication systems that can operate effectively in
this frequency band. Some common statistical parameters
used in 28 GHz channel characterization are Root Mean
Square Delay Spread (RMS-DS), high spatial correlation,
shadowing and Rician K-factor. To determine the narrow-
band impulse response of each sub-channel in the tunnel, all
the multipath components are added together. [75].
hnarr (t,l,k)=
τ
h(t,τ,l,k).(7)
3.1.2 Azimuth angular characterization
Azimuth angular characterization involves the measurement
and analysis of the signal strength and propagation charac-
teristics of a mmWave signal as it is transmitted and received
at different azimuthal angles. To perform azimuth angular
characterization, one typically uses an array of antennas at
both the transmitter and receiver. Angular Spread (AS) is
a measure of the angular variation of the wireless channel
in a particular environment. It is defined as the difference
between the maximum and minimum AoA or AoD of the
signal paths arriving at the receiver or transmitted from the
transmitter, respectively and it is calculated as [76]:
σθ=P(θ)θ2/P(θ)−P(θ)θ/P(θ)2
.
(8)
where θis Angle and P(θ)is the Power of the correspond-
ing angle. A large AS indicates that the signal paths are more
spread out, and the received signal has multiple, uncorre-
lated components. A small AS, on the other hand, indicates
that the signal paths are more directional, and the received
signal has fewer, correlated components. Accurate estima-
tion of the AS is critical for designing efficient beamforming
algorithms, optimizing system performance, and enhancing
channel capacity.
3.2 Measurement set up
Channel measurement is an important aspect of mmWave
communication, as it enables the characterization of the chan-
nel and the estimation of channel parameters such as path
loss, DS, patial correlation and so on. Depending on the
123

45 Page 12 of 22 M. Sharma et al.
Fig. 6 Schematic diagram of Channel Sounder [19]
particular demands of the application, various measurement
setups can be employed for mmWave communication. Here
are some common types of measurement setups:
3.2.1 Channel sounding
Channel sounding involves transmitting known signals and
measuring their received characteristics to determine the
channel properties. This information can be used to design
and optimize communication systems, including beamform-
ing and beam tracking algorithms, to ensure reliable and
efficient communication in mmWave frequency bands. Chan-
nel sounding can be performed using various types of
measurement setups, such as directional antennas and fre-
quency sweep generators. Directional antennas approach
focused radiation pattern, meaning they radiate or receive
energy preferentially in a specific direction [19]. They are
useful for characterizing the directional properties of the
channel, such as the angular spread of the received signals
whereas Frequency sweep generators, generate a continuous
waveform that varies in frequency over time. They are used
to transmit a wideband signal that covers a range of frequen-
cies, which can be used to determine the channel’s frequency
response. Figure 6displays the Channel Sounder’s schematic
diagram.
Channel sounding is an experimental technique used
to characterize the wireless channel between a transmit-
ter (Tx) and receiver (Rx). The sliding correlation-based
channel sounding approach allows for the examination of
wireless signal propagation in the time domain by utilizing
the auto-correlation between a faster pseudo-random noise
(PN) sequence with a chip-rate equivalent to a fast-clock fre-
quency and a slower replica sequence with a rate equivalent to
the slow-clock (β) frequency [77,78]. The continuous-time
operation can be expressed as matrix multiplications between
two row vectors, ¯sand ¯r. Each vector contains discrete sam-
ples of s(t−t0)and r(t), respectively. The resulting vector,
¯
R, has a sample size (n) equal to that of s for each sample
delay, k. This is the representation for the continuous-time
operation in matrix form.
γ=α
α−β.(9)
Rs(τ)=T
0
r(t)s(t−t0−τ)dt.(10)
Rk=¯sk¯rT;k=1ton.(11)
Here we write the expression of Path loss:
PL(d)[dB]=Ptx [dBm]−Prx (d)[dBm]
+Gtx [dBi]Grx [dBi].(12)
In addition to experiencing losses, the signal’s polarization
may be impacted by the different signal propagation paths
within the channel, and the transmitted energy may appear
with the Rx-orthogonal antenna orientation [79,80]. When
the Tx and Rx antennas are co-polarized, the power received
ratio (measured in dB) can identify the propagation feature
as cross-polarization discrimination (XPD).
XPD(d)[dB]=PLV−V(d)[dB]−PLV−H(d)[dB].
(13)
3.2.2 Measurement setup of sub THz
The sub-THz band, also known as the sub-terahertz band,
refers to the range of frequencies that lie just below the tera-
hertz (THz) frequency range. Generally, this band spans from
about 100 GHz to 300 GHz. This frequency range is char-
acterized by its unique properties, such as high data rates,
large bandwidths, low latency, shorter wavelengths (3 mm to
1 mm). Illustrates the arrangement utilized to measure the
sub-THz band in Fig. 7. The measurement setup consists of
a pair of antennas, one serving as the transmitter (Tx) and
the other as the receiver (Rx). Additionally, two Rohde and
Schwarz mmWave converters ZC170 were connected to a
4-port VNA. To enable antenna steering, two antenna posi-
tioners were employed, with a translational precision of 0.01
mm and an angular precision of 0.1◦. The Rx antennas were
mounted on 3-axis (x-y-z) positioners, while the Tx antennas
were positioned on 1-axis (azimuth) positioners. During the
measurement campaign, the Rx positioner executed an x-y-z
spatial grid for each Tx position. Two 20 dBi gain, linearly
polarized horn antennas were used in the experiment [81].
An external laptop and ethernet cables were used to control
the two positions and VNA acquisitions. The sounding band
covered 126 to 156 GHz, with a range of 3001 points or 10
MHz. The output power was configured to 12 dBm, and the
123

Recent trends and future directions in millimeter wave massive MIMO... Page 13 of 22 45
Fig. 7 Measurement Set up of
sub THz Wireless
Communication [81]
Table 7 Compare the Sub THz measurement setup of different indoor scenarios based on necessary parameters [81]
Parameters Classroom Office Hall High speed Train
Wagon
Dimension (m3) 8.54*6.70*2.71 4.78*3.44*2.85 39*20*10 22.92*2.49*2.20
Tx antenna Biconical antenna A Biconical Antenna B Biconical Antenna A Horn Antennas
Propagation Zone LOS, NLOS LOS, NLOS LOS, NLOS LOS, L-NLOS,
D-NLOS
Rx antenna Biconical Antenna B Biconical Antenna A Biconical Antenna B Horn Antennas
(Omni-directive)
Center frequency 28 GHz 28 GHz 28 GHz 60 GHz
Antenna gain 6 dB and 4.8 dB 6 dB and 4.8 dB 6 dB and 4.8 dB 0 dBi
Transmit power 15 dBm 15 dBm 15 dBm 0 dBm
UCA radius 0.24 m 0.24 m 0.24 m NA
Frequency sweep points 360 360 360 3200
Tx/Rx antenna height 1.50 m 1.50 m 1.50 m 1.51 m
Details 20 positions Remove contents and shelves 20 positions Windows, tables and
chairs are fixed
intermediate frequency bandwidth (IFBW) was set to 100
Hz. The typical dynamic range of the sounder was 105 dB,
which corresponds to the free space path loss at 30 m at the
highest frequency.
In this survey, We are considered the indoor scenarios
including Office, Classrooms, Hall and High speed Train
wagon for mmWave propagation channels by using the same
measurement system. For better explanation, we draw a Table
7with all necessary parameter’s name and specifications.
This table shows the comparison between indoor scenarios
based on parameters for sub THz measurement set up.
3.2.3 RDA method
In the Rotated directional antenna (RDA)-based measure-
ment approach, a highly directional antenna is steered in
both the azimuth and elevation domains. The Tx antenna
is mounted on an antenna positioner and rotated from −180◦
to +180◦in azimuth angles and from 30◦to 150◦in eleva-
tion angles, with a step of 5◦in each snapshot of the channel
measurements [15]. A schematic diagram of the RDA mea-
surement method is presented in Fig. 8to provide a better
understanding.
123

45 Page 14 of 22 M. Sharma et al.
Fig. 8 RDA-based
measurement method [15]
Table 8 Compare the channel measurement set-ups of mmWave communication: measurement set-up of sub THz, RDA Method, UVA Method
and Channel Sounding [15,81,83]
Parameters Measurement set-up of sub THZ RDA method UVA method Channel sounding
Bandwidth 30 GHz 2-GHz 2-GHz 4-GHz
Center frequency 141-GHz 60-GHz 60-GHz 142-GHz
Sampling points – 401 401 –
Outputpower 12dBm +13dBm +13dBm –
Intermediate frequency
bandwidth
100 Hz 1 kHz 1 kHz –
Tx locations and scan points
in azimuth and elevation
– Tx1 and Tx3, 72*25 – –
Tx locations and shift
points in cube array
– Tx1-Tx-12, 15*15*6 –
Rx antenna gain and
beamwidth
20 dBi 25 dBi, 10◦25 dBi, 10◦27 dBi, 8◦
Rotation step angle – 5◦––
Position spacing – – 2.5 mm –
Implementation Easy Hard Easy Hard
Resolution High High Low High
Measurement time Less Long Small Small
Hardware complexity Very high Medium Medium Medium to High
Cost implications Extremely high Moderate to high Moderate Moderate to high
Scalability Low Moderate High High
“–”Information is unavailable in cited references
3.2.4 UVA method
The RDA method and the Uniform virtual array (UVA) chan-
nel measurement approach employ a same setup. However,
in the UVA approach, a hollow UVA cube is used in place
of the Tx antenna. The UVA cube consists of an omnidirec-
tional monopole antenna, and both the vertical and horizontal
virtual rectangular arrays are incapable of distinguishing
between waves arriving from the front and back. The mea-
sured frequency range was between 59 and 61 GHz, and 401
frequency samples were obtained in the study [15]. Here we
compare the channel measurement set-ups of mmWave com-
munication in Table 8[16,82].
4 Massive MIMO antennas architecture and
beamforming
Massive MIMO technology in the mmWave bands is a
highly promising candidate for the next generation of cellular
systems, starting with 5 G networks. By combining the sub-
stantial available bandwidth with the high antenna gains of a
large MIMO antenna array, mmWave massive MIMO offers
significant benefits such as enhanced energy and spectral
efficiencies, reliability, flexibility, compactness, and overall
system capacity. This technology is expected to overcome
the limitations of current technology, address the challenges
123

Recent trends and future directions in millimeter wave massive MIMO... Page 15 of 22 45
posed by the exponentially growing demand for mobile data,
and unlock new possibilities for future applications.
4.1 Geometrical Configuration
The antenna array can have two geometrical configurations,
namely linear and planar. Typically, identical radiators are
arranged in lattices with periodic spacing between them, in
linear, rectangular, triangular, or circular patterns, to achieve
the desired radiation pattern. A linear phase taper can be
used to steer the main antenna lobes of a linear array in a sin-
gle plane, creating a beamforming effect. On the other hand,
planar arrays are required to steer beams in both the azimuth
and elevation planes. However, planar arrays have a more
complex structure, which incurs additional costs and com-
plexity [84]. The benefits of cylindrical antenna arrays over
traditional planar arrays for 5 G and upcoming 6 G networks
3D multi-user MIMO scenarios. Under realistic 3D chan-
nel models, cylindrical arrays improve sum-rate, spectral
efficiency, and outage performance by reducing sector-edge
issues mitigating beam enlarging and providing uniform cov-
erage [85,86].
4.2 Array layouts
Phased arrays can be categorized into regular and irregular
configurations depending on their layouts, and a sub-array
synthesis strategy is necessary to achieve a desired power
pattern. However, configuring the array for optimal perfor-
mance and operation bandwidth poses a challenge, and the
use of optimization techniques does not always accurately
predict the array’s performance [84].
4.3 Feed architecture
Various factors must be taken into account when choosing
an antenna array feed network, including the number of
beams, peak and average side lobe levels, power, tunable and
instantaneous bandwidth, and more. The feed network can be
divided into three categories: space, constrained, and hybrid
feeds. In the space feed network, a feed antenna is placed
at the focal distance from the aperture of the pickup array,
which illuminates the array elements. However, this approach
requires a large physical volume [84]. On the other hand, the
constrained feed network can be further divided into series
and parallel/corporate feed networks. A series feed network
results in beam squint occurring more frequently due to grad-
ual phase changes in the radiation elements. This leads to a
reduction in antenna gain and an increase in side lobe level
due to the additive insertion loss of serial phase shifters [84].
4.4 Beamforming architecture
When multiple users receive a transmitted signal, they
may experience interference from each other. To overcome
this issue and enhance the received power, we can uti-
lize beamforming techniques. One of the key advantages
of beamforming is its spatial selectivity, which enables us
to generate a focused beam in a specific direction. This is
achieved by aligning the phases of incoming signals from
different locations within an array. Beamforming arrays can
be classified into four categories based on their architecture:
analog, digital, hybrid, and machine learning (ML) aided
[84].
4.4.1 Analog beamforming
Analog beamforming applies amplitude and phase variation
to the transmit signal in analog form and combines the signals
from multiple antennas before analog-to-digital conversion
at the receiver end. [68]. The RF front-end of a beamforming
array typically includes base-band, power amplification, fre-
quency generation, frequency conversion, modulation, and
phase shifter elements. However, the use of a distributed
Local Oscillator (LO) signal can increase the sensitivity to
noise [68,84]. Nantennas at the Base station, Mantennas at
User equipment (UE) RF chain, analog beamformer vector f
and analog combiner w which maximizes the effective SNR
using the conditions in: [4]
wopt ,fopt =argument maximum|w∗Hf|2.(14)
subject to:
wi=√N−1ejφi,∀i.(15)
fl=√M−1ejφl,∀l.(16)
4.4.2 Digital beamforming
Digital beamforming overcomes the constraint of analog
beamforming by connecting each antenna element (AE) to its
separate RF chain. The RF signal is transformed to a digital
signal at the sub-array or element level, and a digital signal
processor performs beamforming. However, digital beam-
forming presents hardware complexities involving ADCs,
DACs, data converters, and mixers. These hardware com-
plexities lead to high energy consumption, processing time,
and costs, which are the limitations of digital beamforming
[68,84]. The precoders, Matched Filter (MF), Zero Forcing
(ZF), and the Wiener Filter (WF), can be ranked in order
of increasing complexity and performance. The ascending
order of complexity and performance for these precoders is
as follows: MF, ZF, and WF. The mathematical expression
123

45 Page 16 of 22 M. Sharma et al.
for the precoder model involves the average received power
denoted by Prand the noise power represented by σ2
n[4]:
DMF =M
tr (FF∗)F,F=H∗.(17)
DZF =M
tr (FF∗)F,F=H∗HH∗−1.(18)
DWF =M
tr (FF∗)F,F=H∗HH∗+σ2
nM
Pr
I−1
.
(19)
4.4.3 Hybrid beamforming
Hybrid beamforming is a technique that combines the ben-
efits of analog and digital beamforming architectures. By
using a reduced number of RF chains, the system’s per-
formance is improved while reducing costs and energy
consumption. Recently, the interest in hybrid, analog and
digital beamforming arrays has grown due to their ability to
lessen the complexity of digital beamforming while enhanc-
ing the effectiveness of analog beamforming [87]. In the
hybrid beamformer, the analog beamformer is connected to
the end of the digital beamforming architecture. However,
this architecture increases the array complexity and system
cost compared to a passive array. The number of RF chains
is denoted by NRF , which is usually much less than Nfor
mmWave arrays. Each RF chain is connected to all antennas
elements (AEs) through Nphase shifters [68,84]. Received
signal vector rkobserved by the kth terminal after precoding
expressed as [4]:
rk=Hk
K

n=1
AnDnsn+nk.(20)
Where Ak,Dkdenotes the Analog precoder and Digital pre-
coder. yksignal becomes after analog combiner wk:[4]
yk=wH
krk=wH
kHk
K

n=1
AnDnsn+wH
knk.(21)
Despite its benefits, the hybrid beamforming architecture has
some limitations that need to be considered. These limita-
tions include lower beamforming gain, increased hardware
complexity, and decreased beamforming performance [88].
4.4.4 Machine learning aided beamforming
ML aided beamforming refers to the application of machine
learning techniques on hybrid beamforming designs.
Recently, a neural network-based hybrid beamforming sys-
tem has been proposed where a feedforward neural network
(FFNN) is utilized for beam user selection. This system
employs Householder Orthogonal analog beamforming and
FFNN to achieve better energy-efficiency for massive MIMO
channels. The input to the neural network is Channel State
Information (CSI), and computational work is replaced by the
supervised machine learning of radio resource management.
Additionally, a deep learning neural network approximates
the hybrid precoder on singular value decomposition (SVD)
[89]. In this system, there are typically Mbeams and Kusers
in the beam domain, with MK. To reduce operational
and capital costs due to power consumption in the reduced RF
chains (as each beam requires one RF chain), Kbeams are
chosen out of M. For wideband systems with Nsubchannels
and NRF RF chains, Kis less than or equal to NRF , which
is less than or equal to M. The input information signal is
denoted as S∈CNs∗N, and the base-station transceiver is
expressed as:
S=⎡
⎢
⎢
⎢
⎢
⎣
s1[1]s2[2]··· s1[N]
s2[1]s2[2]....
.
.
.
.
.......sNs−1[N]
sNs[1]··· sNs[N−1]sNs[N]
⎤
⎥
⎥
⎥
⎥
⎦
(22)
where Ns=Kis the number of information symbols. nth
subchannel symbol vector expressed by:
s[n]=[s1[n],s2[n],...sK[n]]T.(23)
nth subchannel hybrid precoder is expressed by:
FB[n]=FABFDB [n]∈CM∗K.(24)
nth subchannel digital precoder is expressed by:
FDB [n]=[f1[n],f2[n],...fK[n]] ∈CNRF∗K.(25)
After passing through Analog and digital precoder, transmit
signal per subchannel is expressed by:
x[n]=FABFDB [n]s[n]∈CM∗1
=K
i=1FABfDB,i[n]si[n].(26)
Received signal yk[n]at the user kin the subchannel nis
expressed by [89]:
yk[n]=hk[n]HFABFDB [n]s[n]+zk[n].(27)
The proposed scheme utilizing Householder Orthogonal
Analog Beamforming (HH ABF) and Feedforward Neural
Network (FFNN) achieves better energy efficiency perfor-
mance in massive MIMO channels [90].
123

Recent trends and future directions in millimeter wave massive MIMO... Page 17 of 22 45
Table 9 Comparison of Channel Mode LOS parameters for different scenarios
Parameters Channel mode LOS residential Channel mode LOS office
Number of channel realizations 100 100
Distance between Tx and Rx 5[m] 1[m]
LOS component path loss −81.9842[dB] −68.0048[dB]
Number of average arrival clusters (Lmean) 9 6
Cluster power level (0)−88.70000[dB] −89.07000[dB]
Rician factor (k) 4.34 5.04
Cluster angle of-arrival spread (σφ) 49.8000[deg] 102.0000[deg]
Cluster arrival rate (1/Lam) 5.24 24.6
Ray arrival rate (1/lambda) 0.82 1.03
Cluster decay factor (Gam) 4.46 49.8
Standard deviation of log-normal variable for ray fading 13 11.3
Standard deviation of log-normal variable for cluster fading 6.28 6.6
Ray decay factor (γ) 6.25 45.2
Beam-width of measured Tx antenna 360[deg] 30[deg]
Beam-width of measured Rx antenna 15[deg] 30[deg]
Beam-width of simulated Rx antenna 30[deg] 30[deg]
Average RMS delay 4.977[ns] 22.156[ns]
Maximum RMS delay 17.368[ns] 128.333[ns]
Minimum RMS delay 0.608[ns] 4.950[ns]
Average Racian factor 9.287[dB] 14.992[dB]
Maximum Racian factor 33.103[dB] 30.309[dB]
Minimum Racian factor −14.859[dB] −6.990[dB]
5 Performance analysis and applications
This survey examines two scenarios: Residential LOS and
Office LOS. In the Residential scenario, we assumed that
the space was furnished with items such as furniture, TV
sets, and lounges. The size of this scenario is relatively small
compared to an office. The walls and floors were constructed
using either concrete or wood, and then covered with wallpa-
per or carpet. The residential environment includes windows
and wooden doors. On the other hand, an office environ-
ment typically contains numerous desks, chairs, computers,
and workstations. In addition, there are whiteboards, book-
cases, and cupboards in the office environment. The walls
of an office are composed of concrete or metal and covered
in plasterboard or carpet. Typically, an office will have cubi-
cles, a lab, open and closed offices, and long corridors that
connect these offices.
5.1 Compare the residential and office scenarios
Now we compare these environments with the help of Table
9. The LOS model for the residential scenario was deter-
mined through simulation, covering a bandwidth of 3 GHz
with a distance of 5 m between the transmitter and receiver,
and a center frequency of 62.5 GHz. The simulation and data
Fig. 9 Graphical representation of PDP, AoA and ToA for LOS resi-
dential scenario
analysis for this scenario are shown in Fig.9. The model for
the office scenario was also a LOS scenario, with a distance of
1 m between the transmitter and receiver, a frequency band-
width of 3 GHz, and a center frequency of 62.5 GHz. The
simulation and data analysis for this scenario are shown in
Fig. 10.
123

45 Page 18 of 22 M. Sharma et al.
Fig. 10 Graphical representation of PDP, AoA and ToA for LOS office
scenario
5.2 Applications
Wireless connectivity enables a plethora of innovative user
activities. The mmWave Industrial Scientific Medical (ISM)
band offers improved data rates and network capacity, mak-
ing it suitable for various applications. Some of these
applications have specific use cases, such as inter-rack
communication in data centers, while others are related
to network design improvements like single-hop or multi-
hop backhauling, data offloading, and mobile fronthauling.
Applications related to consumer electronics can be catego-
rized based on the following use model categories:
5.2.1 Rapid file transfer
The vast bandwidths available in unlicensed band channels
can be effectively utilized by applications requiring ultra-
short range communication. The IEEE 802.11ad standard
specifies a peak data rate of 6756.75 Mbps, which can transfer
591 Mbps of data at a MAC layer efficiency of 70%. This high
data rate makes it possible to download a 1.5 GB standard
definition movie at a kiosk or toll-gate in less than 3s. This
approach has the potential to enable cordless computing in
both domestic and commercial settings [55].
5.2.2 Wireless Display
The dissemination of large amounts of data via point-to-point
(PP) or point-to-multipoint (PMP) network topologies. Such
as video is a crucial application that demands high throughput
and Less performance is required in following case: Uncom-
pressed video only requires a small portion of the 1 Gbps data
rate over a single link, but the compressed file requires the full
amount. The range, on the other hand, covers the entire inhab-
ited area for PMP scenarios like classrooms, aeroplanes, and
exhibitions. [91,92].
5.2.3 UHD video streaming
Much like the data rate, the image resolution that users want
from their video monitors is constantly rising. The following
in line are 8K UHD displays, with a graphic resolution rang-
ing from 7680 to 4320 pixels. Data rates of 28.51 Gbps are
required for 8K UHD video at 60 frames per second with 3B
colour depth [93,94]. Channel bonding, higher order modu-
lation and MIMO are used to accomplish this.
6 Challenges and promising solutions
Massive MIMO communication systems operating in the
mmWave spectrum, particularly at frequencies like 30/60/140
GHz, present a unique set of challenges that need innova-
tive solutions. One significant challenge is short-distance
communication. At these high frequencies, signal propa-
gation is greatly affected by free-space path loss, making
it challenging to maintain reliable connections over short
distances. To address this issue, advanced beamforming tech-
niques and adaptive antenna arrays are being employed
to focus the signal power in the intended direction, effec-
tively mitigating path loss. Another challenge stems from
the dense deployment of devices in the same frequency
bands, leading to interference. Coexistence and interference
management become crucial in such scenarios. Solutions
include interference-aware resource allocation algorithms
and sophisticated interference cancellation techniques.
Additionally, the susceptibility to interference from other
devices operating in the same frequency band poses a signif-
icant hurdle. Cognitive radio and dynamic spectrum sharing
techniques are being explored to intelligently allocate fre-
quency resources and adapt to changing interference patterns.
Moreover, robust modulation and coding schemes are essen-
tial to maintain communication reliability in the presence
of interference. In conclusion, while mMIMO-mmWave
communication systems at 30/60/140 GHz offer promising
opportunities for high-capacity and low-latency communica-
tion, they come with their set of challenges. By employing
adaptive antenna arrays, interference management strategies,
and dynamic spectrum sharing, researchers and engineers are
actively working towards unlocking the full potential of these
advanced communication systems.
7 Summary, open issues and future
directions
This survey provides a comprehensive analysis of the most
significant enabling technologies for future wireless commu-
123

Recent trends and future directions in millimeter wave massive MIMO... Page 19 of 22 45
nication systems, including mmWave and massive MIMO
in various frequency bands (GHz), as well as the latest
channel models and measurements. The survey investigates
current channel measurement and modeling methods for
the most difficult bands, including mmWave communica-
tion, massive MIMO communication, multiple access, and
performance evaluation. The article presents mathematical
formulas, theories, and approaches for mmWave massive
MIMO systems. This contributions thoroughly covers var-
ious crucial channel metrics, including path loss, RMS delay
spread, PDP, CIR, SNR, and more. It also takes into account
several research studies concentrating on mmWave chan-
nel models for carriers within the 30/60/140 GHz range.
These studies have revealed that utilizing spatial multiplexing
techniques for beamforming can significantly enhance the
spectral efficiency, channel capacity, and interference miti-
gation capabilities of large MIMO systems operating in the
mmWave band.
The study also provides information on radio-frequency
spectrum. However, future wireless communication systems
are expected to face new challenges in terms of computa-
tional complexities, signal processing, channel estimation,
and pilot contamination issues, which are considered as open
issues of the wireless communication system.
In our forthcoming research, we aim to delve deeper into
the survey, focusing on augmenting the channel’s capacity,
enhancing the efficiency of applications that require high data
transfer and mmWave and massive MIMO emphasized in
details. Additionally, we plan to conduct supplementary mea-
surements within indoor environments, employing mmWave
communication setups across diverse scenarios, including
libraries, laboratories, kiosks, and more. Finally, the survey
simplifies the various applications of the mmWave wireless
communication band. Therefore, the aim of this survey is to
investigate new developments, problems, and suggested solu-
tions for channel characterization and measurement setups
for mmWave massive MIMO at 30/60/140 GHz bands.
Acknowledgements We extend our gratitude to Science and Engineer-
ing Research Board (SERB) for the financial support and for presenting
us with this exceptional research opportunity.
Author contributions Manish Sharma wrote the manuscript under the
supervision of Dr. Anand Agrawal.He would like to thank his supervisor
for his consistent support and guidance during the writing of this survey
paper. He would also like to thank Dr. Chetna Sharma for her advice
in writing the paper. Furthermore, he would like to thank his friend
Mr. Chandraveer Singh for supporting in drawing some figures and in
references.
Funding The authors disclose that they received funding from Science
and Engineering Research Board (SERB).
Data availibility Not Applicable.
Declarations
Conflict of interest The authors have no Conflict of interest.
Ethical approval Not Applicable.
References
1. Chen, L., Hu, B., Xu, G., & Chen, S. (2021). Energy-efficient power
allocation and splitting for mmWave beamspace MIMO-NOMA
with swipt. IEEE Sensors Journal, 21(14), 16381–16394.
2. Wang, C., Pang, M., Zhong, D., Cui, Y., & Wang, W. (2022). A
mmWave communication testbed based on IEEE 802.11ad with
scalable PtMP configuration. China Communications, 19(4), 44–
56.
3. Lawrence, N. P., Ng, B.W.-H., Hansen, H. J., & Abbott, D. (2017).
5G terrestrial networks: Mobility and coverage-solution in three
dimensions. IEEE Access, 5, 8064–8093.
4. Busari, S. A., Huq, K. M. S., Mumtaz, S., Dai, L., & Rodriguez,
J. (2017). Millimeter-wave massive MIMO communication for
future wireless systems: A survey. IEEE Communications Surveys
& Tutorials, 20(2), 836–869.
5. Uwaechia, A. N., & Mahyuddin, N. M. (2020). A comprehensive
survey on millimeter wave communications for fifth-generation
wireless networks: Feasibility and challenges. IEEE Access, 8,
62367–62414.
6. Al-Saman, A., Cheffena, M., Elijah, O., Al-Gumaei, Y. A., Abdul
Rahim, S. K., & Al-Hadhrami, T. (2021). Survey of millimeter-
wave propagation measurements and models in indoor environ-
ments. Electronics, 10(14), 1653.
7. Singh, C., Sharma, C., Tripathi, S., Sharma, M., & Agrawal, A.
(2024). A comprehensive survey on millimeter wave antennas at
30/60/120 ghz: Design, challenges and applications. Wire les s Per-
sonal Communications, 133, 1547.
8. Zhang, G., Saito, K., Fan, W., Cai, X., Hanpinitsak, P., Takada,
J.-I., & Pedersen, G. F. (2018). Experimental characterization of
millimeter-wave indoor propagation channels at 28 GHz. IEEE
Access, 6, 76516–76526.
9. Ji, Y., Fan, W., & Pedersen, G. F. (2018). Channel character-
ization for wideband large-scale antenna systems based on a
low-complexity maximum likelihood estimator. IEEE Transac-
tions on Wireless Communications, 17(9), 6018–6028.
10. Feng, Y., Wang, R., Zheng, G., Saleem, A., & Xiang, W. (2024).
A 3d non-stationary small-scale fading model for 5g high-speed
train massive MIMO channels. IEEE Transactions on Intelligent
Transportation Systems, 25(11), 16490–16505.
11. Bai, L., Huang, Z., Cui, L., & Cheng, X. (2023). A non-stationary
multi-UAV cooperative channel model for 6g massive MIMO
mmwave communications. IEEE Transactions on Wireless Com-
munications, 22(12), 9233–9247.
12. Samimi, M. K., & Rappaport, T. S. (2016). 3-D millimeter-wave
statistical channel model for 5-G wireless system design. IEEE
Transactions on Microwave Theory and Techniques, 64(7), 2207–
2225.
13. Rappaport, T. S., Sun, S., Mayzus, R., Zhao, H., Azar, Y., Wang,
K., Wong, G. N., Schulz, J. K., Samimi, M., & Gutierrez, F. (2013).
Millimeter wave mobile communications for 5-G cellular: It will
work! IEEE Access, 1, 335–349.
14. Rappaport, T. S., Murdock, J. N., & Gutierrez, F. (2011). State
of the art in 60-GHz integrated circuits and systems for wireless
communications. Proceedings of the IEEE, 99(8), 1390–1436.
15. Wu, X., Wang, C.-X., Sun, J., Huang, J., Feng, R., Yang, Y., & Ge,
X. (2017). 60-GHz millimeter-wave channel measurements and
123

45 Page 20 of 22 M. Sharma et al.
modeling for indoor office environments. IEEE Transactions on
Antennas and Propagation, 65(4), 1912–1924.
16. Wu, X., Wang, C., Sun, J., Huang, J., Feng, R., Yang, Y., & Ge,
X. (2017). 60 GHz millimeter-wave indoor channel measurements
and modeling for 5G systems. IEEE Transactions on Antennas and
Propagation, 65(4), 1912–1924.
17. Jacob, M., Priebe, S., Dickhoff, R., Kleine-Ostmann, T., Schrader,
T.,& Kurner, T.(2012). Diffraction in mm and sub-mm wave indoor
propagation channels. IEEE Transactions on Microwave Theory
and Techniques, 60(3), 833–844.
18. Ariza, A. P. G., Müller, R., Wollenschläger, F., Schulz, A., Elkhouly,
M., Sun, Y., Glisic, S., Trautwein, U., Stephan, R., Müller, J., et al.
(2013). 60-GHz ultrawideband polarimetric MIMO sensing for
wireless multi-gigabit and radar. IEEE Transactions on Antennas
and Propagation, 61(4), 1631–1641.
19. De Beelde, B., Plets, D., Desset, C., Tanghe, E., Bourdoux, A.,
& Joseph, W. (2021). Material characterization and radio channel
modeling at D-band frequencies. IEEE Access, 9, 153528–153539.
20. Pometcu, L., & D’Errico, R. (2020). An indoor channel model for
high data-rate communications in d-band. IEEE Access, 8, 9420–
9433.
21. Chen, Y., Han, C., Yu, Z., & Wang, G. (2021). 140 GHz channel
measurement and characterization in an office room. In ICC 2021-
IEEE international conference on communications, pp. 1–6, IEEE.
22. Ge, X., Zi, R., Xiong, X., Li, Q., & Wang, L. (2017). Millimeter
wave communications with OAM-SM scheme for future mobile
networks. IEEE Journal on Selected Areas in Communications,
35(9), 2163–2177.
23. Heath, R. W., Gonzalez-Prelcic, N., Rangan, S., Roh, W., & Sayeed,
A. M. (2016). An overview of signal processing techniques for
millimeter wave MIMO systems. IEEE Journal of Selected Topics
in Signal Processing, 10(3), 436–453.
24. Alkhateeb, A., El Ayach, O., Leus, G., & Heath, R. W. (2013).
Hybrid precoding for millimeter wave cellular systems with partial
channel knowledge. In 2013 information theory and applications
workshop (ITA), pp. 1–5, IEEE.
25. Alkhateeb, A., Mo, J., Gonzalez-Prelcic, N., & Heath, R. W. (2014).
MIMO precoding and combining solutions for millimeter-wave
systems. IEEE Communications Magazine, 52(12), 122–131.
26. Sharma, I. B., Joshi, P., Shrimal, S., Kalra, B., & Sharma, M. M.
(2023). The design of multi band antenna with improved higher
order mode radiation using CMA for l5-band, l1-band, and s-
band application. Scientia Iranica.https://doi.org/10.24200/sci.
2023.62142.7669
27. Shalavi, N., Atashbar, M., & Mohassel Feghhi, M. (2020). Down-
link channel estimation of FDD based massive MIMO using spatial
partial-common sparsity modeling. Physical Communication, 42,
101138.
28. Yu, H., Tuan, H., Nasir, A., Debbah, M., & Fang, Y. (2022). New
generalized zero forcing beamforming for serving more users in
energy-harvesting enabled networks. Physical Communication, 50,
101500.
29. Basar, E., Yildirim, I., & Kilinc, F. (2021). Indoor and outdoor
physical channel modeling and efficient positioning for reconfig-
urable intelligent surfaces in mmwave bands. IEEE Transactions
on Communications, 69(12), 8600–8611.
30. Zamanian, S. F., & Razavizadeh, S. M. (2023). 3D beamforming
in intelligent reflecting surface (IRS)-assisted multi-user cognitive
radio networks. Physical Communication, 56, 101951.
31. Wang, X., Lin, Z., Lin, F., & Hanzo, L. (2022). Joint hybrid 3D
beamforming relying on sensor-based training for reconfigurable
intelligent surface aided terahertz-based multiuser massive MIMO
systems. IEEE Sensors Journal, 22(14), 14540–14552.
32. Qin, H., Liu, Z., & Yang, C. (2022). Indoor mm-wave cover-
age enhancement: Reconfigurable intelligent surface deployment
strategy based on human mobility model. IEEE Communications
Letters, 26(10), 2475–2479.
33. Martalò, M., & Raheli, R. (2022). Outage capacity analysis of the
massive MIMO diversity channel. Physical Communication, 53,
101683.
34. Firouzjaei, H. M., Moghaddam, J. Z., & Ardebilipour, M. (2023). A
virtual MIMO communication for a UAV-enabled cognitive relay
network. IEEE Sensors Journal, 23(17), 20267–20274.
35. Waqar, S., Muaaz, M., & Pätzold, M. (2023). Direction-
independent human activity recognition using a distributed MIMO
radar system and deep learning. IEEE Sensors Journal, 23(20),
24916–24929.
36. Guan, K., Peng, B., He, D., Eckhardt, J. M., Rey, S., Ai, B., Zhong,
Z., & Kürner, T. (2019). Channel characterization for intra-wagon
communication at 60 and 300 GHz bands. IEEE Transactions on
Vehicular Technology, 68(6), 5193–5207.
37. Naviasky, E., Iotti, L., LaCaille, G., Nikoli´c, B., Alon, E., &
Niknejad, A. M. (2021). A 71-to-86-GHz 16-element by 16-beam
multi-user beamforming integrated receiver sub-array for massive
MIMO. IEEE Journal of Solid-State Circuits, 56(12), 3811–3826.
38. Ghasempour, Y., Da Silva, C. R., Cordeiro, C., & Knightly, E.
W. (2017). IEEE802.11ay: Next-generation 60-GHz communica-
tion for 100 Gb/s Wi-Fi. IEEE Communications Magazine, 55(12),
186–192.
39. Zaaimia, M., Touhami, R., Talbi, L., Nedil, M., & Yagoub, M.
(2015). 60-GHz statistical channel characterization for wireless
data centers. IEEE Antennas and Wireless Propagation Letters,
15, 976–979.
40. Daniels, R. C., Murdock, J. N., Rappaport, T. S., & Heath, R. W.
(2010). 60-GHz wireless: Up close and personal. IEEE Microwave
magazine, 11(7), 44–50.
41. Chi, T., Park, J. S., Li, S., & Wang, H. (2018). A millimeter-wave
polarization-division-duplex transceiver front-end with an on-chip
multifeed self-interference-canceling antenna and an all-passive
reconfigurable canceller. IEEE Journal of Solid-State Circuits,
53(12), 3628–3639.
42. Wang, C., Pang, M., Zhong, D., Cui, Y., & Wang, W. (2022). A
mm-Wave communication testbed based on IEEE802.11ad with
scalable PtMP configuration. China Communications, 19(4), 44–
56.
43. Martinez-Ingles, M.-T., Sanchis-Borras, C., Molina-Garcia-Pardo,
J.-M., Juan-Llácer, L., et al. (2013). Experimental evaluation
of an indoor MIMO-OFDM system at 60-GHz based on the
IEEE802.15.3c standard. IEEE Antennas and Wireless Propaga-
tion Letters, 12, 1562–1565.
44. Cid, E. L., Taboas, M. P., Sanchez, M. G., & Alejos, A. V. (2016).
Microcellular radio channel characterization at 60-GHz for 5-G
communications. IEEE Antennas and Wireless Propagation Let-
ters, 16, 1476–1479.
45. Lota, J., Ju, S., Kanhere, O., Rappaport, T. S., & Demosthenous, A.
(2022). mm-Wave V2V localization in MU-MIMO hybrid beam-
forming. IEEE Open Journal of Vehicular Technology, 3, 210–220.
46. Gustafson, C., Haneda, K., Wyne, S., & Tufvesson, F. (2013). On
mm-wave multipath clustering and channel modeling. IEEE Trans-
actions on Antennas and Propagation, 62(3), 1445–1455.
47. Mezzavilla, M., Zhang, M., Polese, M., Ford, R., Dutta, S., Ran-
gan, S., & Zorzi, M. (2018). End-to-End simulation of 5G mmwave
networks. IEEE Communications Surveys & Tutorials, 20(3),
2237–2263.
48. Saleh, A. A., & Valenzuela, R. (1987). A Statistical Model for
Indoor Multipath Propagation. IEEE Journal of Selected Area in
Communications, 5(2), 128–137.
49. Maltsev, A., et al. (2010). IEEE802.11 wireless LANs channelmod-
els for 60-GHz WLAN systems. Tech. Rep.: IEEE.
50. Yong, S.-K. (2006). TG3C channel modeling sub-committee final
report. IEEE802. 15-06-0195-07-003c.
123

Recent trends and future directions in millimeter wave massive MIMO... Page 21 of 22 45
51. Shoji, Y., Sawada, H., Choi, C.-S., & Ogawa, H. (2009). A
modified SV-model suitable for line-of-sight desktop usage of
millimeter-wave WPAN systems. IEEE Transactions on Antennas
and Propagation, 57(10), 2940–2948.
52. Charfi, E., Chaari, L., & Kamoun, L. (2013). PHY/MAC enhance-
ments and QoS mechanisms for very high throughput WLANs: A
survey. IEEE Communications Surveys & Tutorials, 15(4), 1714–
1735.
53. Rappaport, T. S., Heath, R. W., Jr., Daniels, R. C., & Murdock,
J. N. (2015). Millimeter wave wireless communications.Prentice
Hall.
54. Seyedi, A. (2007). TG3C system requirements. IEEEP802.15-07-
0583-01-003c.
55. Grodzinsky, M. (2009). Amendment to WFA usage models. IEEE
802.11–09/0583r0.
56. Ghasempour, Y., da Silva, C. R. C. M., Cordeiro, C., & Knightly, E.
W. (2017). IEEE 802.11 ay: Next-generation 60 GHz communica-
tion for 100 Gb/s Wi-Fi. IEEE Communications Magazine, 55(12),
186–192.
57. Zhou, P., Cheng, K., Han, X., Fang, X., Fang, Y., He, R., Long,
Y., & Liu, Y. (2018). IEEE 802.11ay-based mm-Wave WLANs:
Design challenges and solutions. IEEE Communications Surveys
& Tutorials, 20(3), 1654–1681.
58. Ait Taleb, H., Ghanem, K., Zaaimia, M. Z., Ben Mabrouk, I., &
Nedil, M. (2020). On 60-GHz MIMO diversity in an underground
mine propagation channel. IEEE Antennas and Wireless Propaga-
tion Letters, 19(10), 1769–1773.
59. Bandyopadhyay, L., Chaulya, S., & Mishra, P. (2010). Wireless
communication in underground mines. RFID-Based Sens. Netw,
vol. 22.
60. El Azhari, M., Nedil, M., Alj, Y. S., Mabrouk, I. B., Ghanem, K.,
& Talbi, L. (2015). Off-body LOS and NLOS channel characteri-
zation in a mine environment. In: 2015 International Conference
on Electrical and Information Technologies (ICEIT), pp. 114–118,
IEEE.
61. Mohamed, E. M., Abdelghany, M. A., & Zareei, M. (2019). An effi-
cient paradigm for multiband WiGig D2D networks. IEEE Access,
7, 70032–70045.
62. Deng, N., & Haenggi, M. (2017). A fine-grained analysis of
millimeter-wave device-to-device networks. IEEE Transactions on
Communications, 65(11), 4940–4954.
63. Siligaris, A., Richard, O., Martineau, B., Mounet, C., Chaix, F.,
Ferragut, R., Dehos, C., Lanteri, J., Dussopt, L., Yamamoto, S. D.,
et al. (2011). A 65-nm CMOS fully integrated transceiver module
for 60-GHz wireless HD applications. IEEE Journal of Solid-State
Circuits, 46(12), 3005–3017.
64. Romero-Jerez, J. M., Lopez-Martinez, F. J., Paris, J. F., & Gold-
smith, A. J. (2017). The fluctuating two-ray fading model: Statisti-
cal characterization and performance analysis. IEEE Transactions
on Wireless Communications, 16(7), 4420–4432.
65. Fuschini, F., Häfner, S., Zoli, M., Müller, R., Vitucci, E., Dupleich,
D., Barbiroli, M., Luo, J., Schulz, E., Degli-Esposti, V., et al. (2017).
Analysis of in-room mm-wave propagation: Directional channel
measurements and ray tracing simulations. Journal of Infrared,
Millimeter, and Terahertz Waves, 38(6), 727–744.
66. Zhang, L., Zhao, H., Hou, S., Zhao, Z., Xu, H., Wu, X., Wu,
Q., & Zhang, R. (2019). A survey on 5-G millimeter wave com-
munications for UAV-assisted wireless networks. IEEE Access, 7,
117460–117504.
67. Serghiou, D., Khalily, M., Brown, T. W. C., & Tafazolli, R. (2022).
Terahertz channel propagation phenomena, measurement tech-
niques and modeling for 6G wireless communication applications:
A survey, open challenges and future research directions. IEEE
Communications Surveys & Tutorials, 24(4), 1957–1996.
68. Xiao, Z., Zhu, L., Liu, Y., Yi, P., Zhang, R., Xia, X.-G., & Schober,
R. (2021). A survey on millimeter-wave beamforming enabled
UAV communications and networking. IEEE Communications
Surveys & Tutorials, 24(1), 557–610.
69. Federal Communications Commission (2016). Spectrum fron-
tiers report and order and further notice of proposed rulemaking:
FCC16-89.
70. Al-Samman, A., Rahman, T., Azmi, M., Hindia, M., Khan, I., &
Hanafi, E. (2016). Statistical modelling and characterization of
experimental mm-Wave indoor channels for future 5-G wireless
communication networks. PloS one, 11(9), e0163034.
71. Khademi, S., Chepuri, S. P., Irahhauten, Z., Janssen, G. J., & van
der Veen, A.-J. (2015). Channel measurements and modeling for
a 60-GHz wireless link within a metal cabinet. IEEE Transactions
on Wireless Communications, 14(9), 5098–5110.
72. Agrawal, A.BER Analysis of UWB and mm-Wave Communication
Systems Adhering to IEEE 802.15 WPAN Standards. PhD thesis,
2018.
73. Sudhamani, C., Roslee, M., Chuan, L. L., Waseem, A., Osman, A.
F., & Jusoh, M. H. (2024). Enhanced indoor path loss and RSRP of
5G mmWave communication system with multi-objective genetic
algorithm. Wireless Personal Communications, 138(1), 603–621.
74. Obeidat, H. (2024). Investigations on millimeter-waveindoor chan-
nel simulations for 5G networks. Applied Sciences (Basel), 14,
8972.
75. Oestges, C., Vanhoenacker-Janvier, D., & Clerckx, B. (2006).
Channel characterization of indoor wireless personal area net-
works. IEEE Transactions on Antennas and Propagation, 54(11),
3143–3150.
76. Yin, X. (2005). Cluster angular spreads in a MIMO indoor propa-
gation environment. In 2005 IEEE 16th international symposium
on personal, indoor and mobile radio communications,vol.1,
pp. 664–668.
77. Pirkl, R. J., & Durgin, G. D. (2008). Optimal sliding correlator
channel sounder design. IEEE Transactions on Wireless Commu-
nications, 7(9), 3488–3497.
78. Wu, T., Rappaport, T. S., Knox, M. E., & Shahrjerdi, D. (2019). A
wideband sliding correlator-based channel sounder with synchro-
nizationin65nmCMOS.In2019 IEEE International Symposium
on Circuits and Systems (ISCAS), pp. 1–5, IEEE.
79. Nguyen, S. L., Jarvelainen, J., Karttunen, A., Haneda, K., & Putko-
nen, J. (2018). Comparing radio propagation channels between 28
and 140 GHz bands in a shopping mall.
80. Pometcu, L., & D’Errico, R. (2019). Channel model characteristics
in D-band for NLOS indoor scenarios. In 2019 13th European
Conference on Antennas and Propagation (EuCAP), pp. 1–4, IEEE
81. Pometcu, L., & D’Errico, R. (2019). An indoor channel model for
high data-rate communications in D-band. IEEE Access, 8, 9420–
9433.
82. Shakya, D., Wu, T., Knox, M. E., & Rappaport, T. S. (2021).
A wideband sliding correlation channel sounder in 65 nm cmos:
Evaluation board performance. IEEE Transactions on Circuits and
Systems II: Express Briefs, 68(9), 3043–3047.
83. Lecci, M., Testolina, P., Polese, M., Giordani, M., & Zorzi, M.
(2021). Accuracy versus complexity for mmWave ray-tracing: A
full stack perspective. IEEE Transactions on Wireless Communi-
cations, 20(12), 7826–7841.
84. Ghosh, S., & Sen, D. (2019). An inclusive survey on array antenna
design for millimeter-wave communications. IEEE Access, 7,
83137–83161.
85. Lin, Z., Lv, T., Ni, W., Zhang, J. A., & Liu, R. P. (2021). Nested
hybrid cylindrical array design and DoA estimation for massive
IoT networks. IEEE Journal on Selected Areasin Communications,
39(4), 919–933.
86. Riviello, D. G., Tuninato, R., & Garello, R. (2024). Multi-layer
multi-user MIMO with cylindrical arrays under 3g pp 3d channel
model for b5g/6g networks. IEEE Access, 12, 145753–145767.
123

45 Page 22 of 22 M. Sharma et al.
87. Han, K., Qian, H., Hu, S., & Kang, K. (2022). Performance analysis
of hybrid beamforming systems with analog mismatches. Physical
Communication, 53, 101770.
88. Ahmed, I., Khammari, H., Shahid, A., Musa, A., Kim, K. S., De
Poorter, E., & Moerman, I. (2018). A survey on hybrid beamform-
ing techniques in 5G: Architecture and system model perspectives.
IEEE Communications Surveys & Tutorials, 20(4), 3060–3097.
89. Ahmed, I., Shahid, M. K., Khammari, H., & Masud, M. (2021).
Machine learning based beam selection with low complexity
hybrid beamforming design for 5-G massive MIMO systems. IEEE
Transactions on Green Communications and Networking, 5(4),
2160–2173.
90. Szott, S., Kosek-Szott, K., Gawłowicz, P., Gómez, J. T., Bellalta,
B., Zubow, A., & Dressler, F. (2022). Wi-Fi meets ML: A survey
on improving IEEE 802.11 performance with machine learning.
IEEE Communications Surveys & Tutorials, 24(3), 1843–1893.
91. Linfoot, S. L., & Sherratt, R. S. (2000). Analysis of a DVB-T
compliant receiver simulation under various multipath conditions.
IEEE Transactions on Consumer Electronics, 46 (1), 201–206.
92. Choi, J. H., Kwon, H., Kim, J., Lee, W.-Y., & You, Y. (2013).
Implementation of a seamless uncompressed video transmission
system in 60-GHz bands. In 2013 IEEE international conference
on consumer electronics (ICCE), pp. 422–423, IEEE.
93. Park, Y.-H., Kim, J., Kim, M., Lee, W., & Lee, S. (2015). Pro-
grammable multimedia platform based on reconfigurable processor
for 8k UHD TV. IEEE Transactions on Consumer Electronics,
61(4), 516–523.
94. Xie, F., Pourazad, M. T., Nasiopoulos, P., & Slevinsky, J. (2016).
Determining bitrate requirement for UHD video content delivery.
In 2016 IEEE international conference on consumer electronics
(ICCE), pp. 241–242, IEEE.
Publisher’s Note Springer Nature remains neutral with regard to juris-
dictional claims in published maps and institutional affiliations.
Springer Nature or its licensor (e.g. a society or other partner) holds
exclusive rights to this article under a publishing agreement with the
author(s) or other rightsholder(s); author self-archiving of the accepted
manuscript version of this article is solely governed by the terms of such
publishing agreement and applicable law.
Mr. Manish Sharma He has
received his B.Tech degree from
Univerity College of Engineering
Kota, Rajasthan, India, in 2010,
M.Tech degree from Jagannath
University Jaipur, Rajasthan, India,
in 2014. He is currently pursu-
ing the Ph.D. degree in the depart-
ment of Electronics and commu-
nication Engineering at Indian
Institute of Information Technol-
ogy Kota, Rajasthan, India. He
has over 14 years of Teaching,
Research and Industrial experience.
He is working as a Junior Research
Fellow at department of Electronics and Communication Engineer-
ing in Indian Institute of Information Technology Kota, Rajasthan,
India. His current research interests include mm-Wave, 60GHz, m-
MIMO, Network Slicing, Beamforming, OFDM, Channel Characteri-
zation and 5G/6G wireless communication.
Dr. Anand Agrawal He has
received his B.Tech degree from
Samrat Ashok Technological Insti-
tute, Vidisha, India, in 2008, M.
Tech degree from National Insti-
tute of Technology Bhopal, India,
in 2010 and Ph.D from Indian
Institute of Technology Guwahati,
India in 2018. Since June 2019, he
is working as an Assistant Profes-
sor in the department of Electron-
ics and Communication Engineer-
ing at Indian Institute of Infor-
mation Technology Kota, India.
From Jan. 2018 to May 2019, he
worked as an Assistant Professor in the department of Electronics
and Communication Engineering at Jaypee Institute of Information
Technology Noida, India. During PhD, he was a visiting research
scholar at National Tsing Hua University (NTHU), Taiwan from
Sept 2015 to Jan 2016. He has been involved in organizing sev-
eral national/international conferences in various capacities. He is
also a reviewer of various reputed national and international journals
including, Physical Communication, Wireless Personal Communica-
tion, Transaction of emerging Telecommunication technologies, IEEE
communication letter and so on. His current area of research inter-
est are UWB Communication, mm-Wave Communication at 60 GHz,
Cooperative Communication, IEEE802.15 Standards, 5G Technology,
Massive MIMO and D2D Communication.
Mr. Chandraveer Singh He has
over 14 years of teaching, research
and industrial experience. Previ-
ously he was holding the posi-
tion of Deputy Manager (Opera-
tion and Maintenance) instrumen-
tation and Control in ESSAR Oil
Ltd, India. Also, he has worked
with Govt. Engineering College,
Ajmer as Guest Lecturer in the
Department of EI&CE. Presently,
he is working as Assistant Pro-
fessor in School of Automation,
Banasthali Vidyapith (Deemed to
be University), Rajasthan, India.
He is also taking care of training and placement in his department.
His area of interest includes Microstrip Antenna, Ultra- Wideband
antenna, mmWave antenna, flexible antenna, antenna for medical
applications.
Dr. Chetna Sharma She has
received her BE from Govern-
ment Engineering College,
Bilaspur and Dual Degree (M.
Tech. & PhD) from PDPM Indian
Institute of Information Technol-
ogy Design and Manufacturing,
Jabalpur, India in 20011 and 2018,
respectively. She has been work-
ing as an Assistant Professor with
the ECE Department, Indian Insti-
tute of Information Technology
Kota, India, since 2019. She is
actively involved in research-
oriented antenna systems for
defense and surveillance systems, RF energy harvesting systems and
biomedical applications. She has also been mentoring one research
project for developing UWB antenna system for surveillance RADAR.
123