Content uploaded by Shah Zeb
Author content
All content in this area was uploaded by Shah Zeb on May 20, 2022
Content may be subject to copyright.
Interference Mitigation in RIS-assisted 6G Systems
for Indoor Industrial IoT Networks
Naila Rubab∗, Shah Zeb∗, Aamir Mahmood†, Syed Ali Hassan∗, Muhammad Ikram Ashraf§, and Mikael Gidlund†
∗School of Electrical Engineering & Computer Science (SEECS),
National University of Sciences & Technology (NUST), Islamabad 44000, Pakistan.
†Department of Information Systems & Technology, Mid Sweden University, Sweden.
§Nokia Bell Labs, Espoo, Finland.
Email: ∗{nrubab.msee18seecs, szeb.dphd19seecs, ali.hassan}@seecs.edu.pk,†{firstname.lastname}@miun.se,
§ikram.ashraf@nokia-bell-labs.com.
Abstract—Ever necessity of tremendous data traffic and mas-
sive deployment of industrial Internet-of-Things (IIoT) devices
operating in higher bands, i.e., millimeter-wave (mmWave) and
terahertz (THz), have encouraged academia and industry to
transition towards the future sixth-generation (6G) wireless
communication networks. Nevertheless, the recent emergence of
6G enablers, i.e., reflecting intelligent surface (RIS) and massive
multiple-input multiple output (mMIMO), has the ability to po-
tentially change the previous paradigm of indoor mmWave wire-
less communication by modifying the propagation environment.
It can control and establish the favorable and tunable wireless
channel responses by exploiting the multipath and diversity of
the propagation environment. Therefore, this next cutting-edge
technology is capable of massively improving the performance of
mmWave-mMIMO-enabled IIoT data transmissions, making it a
feasible solution for 6G networks. In this paper, we propose a
RIS-assisted mmWave-mMIMO for a multi-cells indoor factory
propagation environment, which provides, 1) aid in mitigating
the impact of radio frequency (RF) interference from interferes
in closed vicinity (i.e., neighbor cells) by employing metasurface
laminated walls, and 2) increased the mmWave-mMIMO sys-
tem performance by controlling and tuning the indoor factory
channel conditions. Our results indicate that the proposed RIS-
assisted mmWave-mMIMO system outperforms the benchmark
link capacity performance in the presence of interference
Index Terms—6G, reconfigurable intelligent surfaces, massive
MIMO, millimeter-Wave, industrial Internet-of-things.
I. INTRODUCTION
The sixth-generation (6G) networks, being extremely ef-
ficient, are envisioned to provide a high data rate commu-
nication of up to Terabit-per-second (Tbps) with minimal
power usage, indiscernible ultra-high reliability and low-
latency for large-scale industrial internet-of-things (IIoT) ap-
plications [1], [2]. Several critical enabling technologies, e.g.,
massive multiple-input multiple-output (mMIMO), millimeter
wave (mmWave), terahertz (THz) communication, as well
as huge antenna array designs and better signal processing
techniques, have been presented as solutions to achieving
challenges of high data rates and energy efficiency in 6G
vision [3], [4].
Nevertheless, while the aforementioned technologies have
demonstrated a significant improvement in spectral efficiency
of wireless networks, though at the cost of increased energy
consumption and hardware cost due to the requirement of
integrating additional active antennas and sophisticated radio
frequency (RF) chains working at higher frequency bands [5].
Thus, making these approaches complex and expensive.
A. Motivation
The key limiting issues in current wireless communication
networks performance include uncontrollable channel environ-
ment and the need to eliminate interference from coexisting
systems, e.g., sufficient isolation between the uncoordinated
and adjacent indoor micro operator deployements [6], [7].
For 5G networks operating at mmWave bands, there are
two primary causes of unreliable link performance: a) strong
directivity, rendering mmWave communications vulnerable
to obstruction, b) increased likelihood of interference from
nearby systems. One of the potential cost-effective solutions
is modifying and tuning the wireless scattering environment
to create programmable channel conditions.
Reconfigurable Intelligent Surfaces. In this regard, reconfig-
urable intelligent surfaces (RIS) for 6G networks is considered
to be a critical enabling technology to address interference
mitigation,blockage issues and achieve uninterrupted wireless
connectivity for mmWave and THz networks by attaining in-
telligent radio environments [8]. By deploying the frequency-
selective surfaces (FSS)-based RIS wallpaper (a planar struc-
ture with PIN diodes implanted with an external bias) to block
unwanted co-channel interference while enabling desired radio
communication to travel across the walls is an intelligent
strategy for enhancing the signal-to-interference-plus-noise
ratio (SINR) at receivers [9]–[11]. Hence, FSS-based RIS
can be used to achieve radio isolation, minimize interference
between nearby co-channel wireless systems, and strengthen
wireless security by limiting radio wave spill over outside of
defined regions in an indoor factory environment (InF) [12].
RIS 2D Structure and Working. RIS two-dimensional (2D)
metasurface consists of a massive number of low-cost passive
reflecting electronic elements, each of which may indepen-
dently modify the phase, amplitude, frequency, or even polar-
ization of incident wavefront. The electromagnetic (EM) wave
characteristics such as scattering, reflection, and refraction
directed at these enormous number of adjustable elements are
intelligently controlled individually in real-time through ap-
Cell-3 Cell-4
Cell-2
Cell-1
gNB-3 gNB-4
gNB-2gNB-1 (Serving BS)
IT-3
IT-4
IT-2
IT-1
Conventional InF Wall
Signal
Interference
Link
Useful
Signal Link
Fig. 1. Illustration of an RIS-assisted mmWave-mMIMO system deployment
in a multi-cell InF environment.
plying desired control signal, resulting in highly defined three-
dimensional passive beamforming with achieving exceptional
massive MIMO gains [10], [13]. Moreover, it functions as
a signal scatterer without demanding any energy source for
signal processing, unlike a traditional multi-antenna relay.
Hence, RIS-2D is regarded as an extension of the mMIMO
system, substantially reducing the wireless network energy in
contrast to the existing wireless link adaptation techniques.
However, the reflection coefficient must be designed very
carefully so that the required signal strength can be increased
by coherently adding the signal reflections by other paths or
destructively to mitigate co-channel interference [8].
B. Contributions
This study presents two scenarios of mmWave-mMIMO-
enabled system deployment in an InF setting, one with conven-
tional walls and the other with RIS wallpaper-covered walls.
With this, it evaluates the increase in attainable mmWave-
mMIMO link capacity in the presence of signal interference.
The main contributions of this paper are as follows
•We used the 3D geometry-based statistical channel model
(SCM) technique in our work by following the framework
of [14], [15] to realistically recreate the InF environmen-
tal conditions for mmWave-mMIMO links in both con-
sidered scenarios (conventional walls and walls covered
with the RIS wallpapers).
•A system model, which incorporates the co-system in-
terference in InF multi-cell setting, is implemented and
used to compute the mmWave-mMIMO link capacity in
both scenarios.
The rest of the paper is organized as follows. The developed
system model for analysis is described in Section II. Simula-
tion results are presented in Section III. Finally, Section IV
concludes the paper.
II. RIS-AS SI ST ED M MWAVE-MMIMO SYSTEM MODEL
Consider a closely placed multiple InFs, i.e., multi-cells (as
shown in Fig. 1) where four gNodeB (gNBs) by different op-
erators are deployed to provide high data rate communication
links to four industrial devices/terminals (ITs) on manufactur-
ing floor. Let B={1,2, ...., B}and I={1,2, ...., U }be
defined as the set of deployed gNBs and ITs, respectively,
while, Band Uare the maximum number of gNBs and
ITs. We assumed a downlink RIS-assisted mmWave-mMIMO
communication scenario for this analysis. The mMIMO com-
munication module at both gNBs and ITs has a uniform planar
array (UPA) system that operate at mmWave frequency fc.
The NTx and NRx are total number of antenna elements at
both transmitter (Tx) and receiver (Rx), respectively, while the
horizontal and vertical inter-element spacing is kept at λc/2.
A. Scenario-1: Conventional Walls
In this scenario, when no walls are laminated with RIS coat-
ings, the useful mMIMO link experiences signal interference
originating from other neighboring gNBs.
Geometry-based 3D SCM. For the conventional wall sce-
nario, we used the geometry-based 3D SCM framework
of [15] to realistically model the channel gain matrix for each
mMIMO link (useful or interference) and is represented by
HC∈CNRx×NTx , where each entry of HCis,
hpq =
NTC
X
f=1
NSP
X
g=1
αfgej2π τfg fcGRx
ϕRx
fg, θRx
fgGTx
ϕTx
fg, θTx
fg,
(1)
In (1), NTC,NSP are the total number of InF scatter clusters
and associated cluster subpaths (SP), respectively, αf g and
τfg are the associated SP amplitude and time delay; GRx (·)
is the 3D spatial gain signature along angle-of-arrival spread
of azimuth and elevation angular components while GTx (·)is
along angle-of-departure angular spread.
mMIMO System Capacity with Signal Interference. To
compute the resultant capacity for mMIMO link analysis in
the presence of external interferers, the achievable capacity
for j-th IT (Rx) and i-th gNB (Tx) is given as
Cij = log2
1+ΩTxl−1
ij
NRx
X
m=1
NTx
X
n=1
hpq
2
÷
Xy̸=jΩTxl−1
yj
NRx
X
m=1
NTx
X
n=1
hpq
2
+N0
,(2)
where lyj represents the interfering link pathloss between j-th
Rx and y-th interfering gNB, ΩTx is the transmitted power,
N0represents the noise power spectral density, assumed to
be −174 dBm/Hz. We selected InF pathloss model (in dB)
of [16] for both the line-of-sight (LOS) and no-LOS (NLOS)
conditions.
B. Scenario-2: RIS-laminated Walls
In the absence of RIS-assisted communication, because
of the blockage probability, associated high pathloss and
signal interference from other cells, the signal received by
IT devices situated at the cell’s edge would be considerably
Cell-1
gNB-3 gNB-4
gNB-2
gNB-1 (Serving BS)
IT-1
RIS-lamented InF Walls
Blockage of External
Interference Links
(Radio Isolation)
RIS-assisted
Reflecting Link
G
D
H
RIS
Controller
Fig. 2. A detailed description of RIS-assisted mmWave-mMIMO communi-
cation of cell-1 in InF environment.
reduced. Consequently, the SINR in (2) can fall precipitously,
impacting the link capacity. In this scenario, the FSS-based
RIS planes are dispersed on InF walls (e.g., cell-1 of Fig. 2
to aid mmWave-mMIMO communication by shielding signal
interference. RIS-laminated planes consists of numerous RIS
passive reflecting elements connected with the central RIS
controller, managing the RIS’s reflecting elements.
RIS Metasurfaces Modes & Interference Mitigation. In
RIS, a control signal from RIS microcontroller, attached to
each RIS lamenated wall (c.f. Fig. 2), manages to control the
RIS metasurface in a transparent mode and/or a reflective
mode dynamically to impact the propagation environment.
Here, for the cell-1 signal transmitter, the RIS is configured
in reflective mode or as a passive repeater; therefore, the
transmitted signals are only limited to inside cell-1. Conse-
quently, the signal power received by IT-1 as a result of a
reflected contribution from a RIS-laminated wall is always
greater than that received from any other object/material that
does not reflect strongly. While, for outside interferers, it
works in transparent mode, such that it is configured to provide
isolation, which mitigates the interference between co-channel
wireless systems present in adjacent cells; hence, it acts as an
isolator at operating frequency by rejecting a frequency range
selectively. While rest of the signals for communication on
other bands are permitted to enter and leave cell-1, and as a
result the system performance improves.
3D GB-SCM for RIS-assisted InF communication. In
RIS-assisted communication for cell-1 (c.f. Fig. 2), the two
propagation paths exist between IT (Rx) and serving gNB
(Tx), namely the direct-link and the RIS-assisted reflecting-
link. The direct-link is assumed to be the LOS path for which
signal attenuation is proportional to the 2D Euclidean distance
between i-th gNB and j-th IT, giving the mMIMO channel
gain matrix D∈CNRx×NTx . The probability of an IT device
being in LOS conditions , i.e., PLOS, inside InF environment
is given by
PLOS =e
ln(1−rd)(hc−hIT)dij
(hgNB−hIT )dc,(3)
where dcand rcare the InF clutter size and density, respec-
tively, hcis the effective clutter height while hIT and hgNB
are the height of gNB and IT, respectively. Similarly, the
RIS-assisted reflecting-link is comprised of three phases in
reflecting propagation mode: gNB-to-RIS path, RIS-reflection,
and lastly RIS-to-IT path; all three phases contribute in chan-
nel gain as H∈CN×NTx ,Φ∈CN×Nand G∈CNRx×N,
respectively, where Nis the total reflective elements in RIS-
laminated wall. We produce G,Φ,H, and Dby following
the 3D GB-SCM framework of [14] to realistically integrate
the InF environment conditions in our RIS-assisted mmWave-
mMIMO deployed system. It mitigates the negative effects
of multipath fading by coherently integrating radio waves
being reflected, refracted, and dispersed from these path. The
overall composite end-to-end channel gain matrix HRIS in RIS-
assisted link between gNB-to-IT in InF settings is defined
as [14] HRIS =GΦH +D, where HRIS ∈CNRx×NTx . After
computing the composite gain matrix HRIS, (2) is used to
evaluate the final achievable capacity.
III. SYS TE M SIMULATION AND RES ULT S
This section initially discusses the simulation environment
setup and the impact of increasing mMIMO antenna elements
on the achievable mmWave-mMIMO link capacity in the
presence of interference from neighboring cells. Afterward,
we analyze the achievable mmWave-mMIMO link capacity
performance for both scenario-1 and scenario-2 of the Sec. II.
A. Simulation Environment Setup and Parameters
We assume a single user (U= 1) and four gNBs (B= 4)
scenario where the closest gNB (i= 1) is selected for
communication at fc= 28 GHz while the remaining three
gNBs (i= 2,3,4) act as interferers. The distances are kept as,
d11 = 10 m, d21 = 30 m, d31 = 50 m, and d41 = 100 m. We
selected 16 antenna element (4×4)-based UPA configuration in
both IT and gNBs. We set the total number of RIS elements,
i.e., N= 2048. Note that, in both performance analyses of
Fig. 3 and Fig. 4, we used the 3D geometry-based SCM
technique to create the associated channel gains of mmWave-
mMIMO links between the IT and each 5G gNB (serving or
interferer) using Monte-Carlo simulations, which incorporates
the large-scale parameters (LSPs) and small-scale parameters
(SSPs) of indoor environmental conditions realistically. The
necessary steps to create 3D SCM with InF LSPs (i.e, environ-
mental settings)-SSPs (temporal and spatial channel statistics)
are taken from [16, Table I].
B. Impact of Increasing Tx-Rx Antenna Elements
First, we assess the impact of interference on the achievable
link capacities between the gNB-1 (Tx) and IT-1 (Rx) from
neighboring cells under the assumptions of scenario-1. We
created the channel gains for links between IT-1 and each
gNBs, and evaluate the SINR12 using eq. (2) to obtain the
achievable capacities under no interference and l= 1,2,and 3
total number of interferers. Fig. 2 reveals that with increasing
interference from neighbouring cells, the impact on reduction
Fig. 3. Impact of increasing antenna elements on achievable mmWave-
mMIMO link capacities in the presence of neighbouring gNBs interference
from adjacent InF cells. Note that for the benchmark performance, we assumed
no RIS-laminated indoor walls and no neighbouring cell interference.
of achievable link capacity is high. As a solution to reduce
signal interference impact, we increased the mMIMO antenna
elements at both the Tx and Rx module, which can increase
the useful signal power. By increasing antenna elements at
both Tx and Rx UPA, we observe that the impact of inter-
ference on achievable capacities reduces. However, this way
of improving received power results in increased cost and
energy consumption due to the addition of active RF chains.
Furthermore, it does not allow any modification and control to
the wireless scattering environment. As a result, it is ineffective
for attaining radio isolation and decreasing neighbouring co-
channel interference.
C. Impact of RIS-laminated InF Walls
We use the RIS-laminated InF walls to reduce the interfer-
ence and increase the channel sparsity. Considering scenario-
2, when RIS metasurface is deployed on only one wall of
the factory hall, a significant increment in achievable capacity
can be observed in Fig. 4 as compared to the benchmark
performance. This observation indicates that, compared to the
unmodified environment, the RIS-laminated InF wall prototype
could successfully restrict all the signals at 28 GHz and
enhances the received SINR at the IT device by offering extra
isolation to the required signal coming outside the laminated
wall, i.e., providing shielding from outside interference coming
from three nearby co-system interferences.
Now, considering the case when shielding from two inter-
ferers is achieved by laminating two InF walls. It can be
observed from the figure that the achievable link capacity
increases substantially, as compared to the case when only
one wall is laminated. Similarly, at the end, when all three
walls of the hall are shielded from the outside interferers
present in close proximity of the cell-1, a drastic change in
achievable MIMO capacity has been noted. This analysis states
that when RIS-assisted mmWave-mMIMO system is deployed
in a InF conditions, it increases the existence of strongly
reflected multipath components received by IT device. The
system performance is no longer solely determined by MIMO
Fig. 4. RIS-aided achievable mmWave-mMIMO link capacities in the
presence of neighbouring BSs interference from adjacent InF cells.
gains. Instead, it majorly depends upon the RIS-assisted gain,
which includes two system gains, one is direct gain, and the
other is RIS-assisted reflection gain in contrast to the system
with an unmodified environment. Therefore, two significant
improvements are experienced by the RIS-assisted mmWave-
mMIMO system, which further improves the radio channel
characteristics and link capacity.
IV. CONCLUSION AND FUTURE WOR K
The RIS and metasurfaces deployments for indoor factory
scenarios are poised to form a critical part of the 6G vision.
The explored research area is a promising research direction,
which will significantly increase the NLOS communication
coverage behind obstructed areas, shielding interference, and
increase the mMIMO channel’s sparsity. In this respect, we
proposed a RIS-assisted mmWave-mMIMO system in the InF
conditions and assessed the system performance in terms of
achievable link capacities under interference from gNBs of
neighboring manufacturing cells. To realistically incorporate
the InF channel conditions, we used the 3D geometry-based
SCM technique to create the RIS-assisted and without RIS
wireless links between the IT and each deployed gNBs,
i.e., serving and interferers. The obtained results show that
integrating RIS technology with the mmWave-mMIMO system
is poised to perform better in an interference-rich InF environ-
ment because it effectively shields the manufacturing cell from
incoming interference at the receiver, allowing for channel gain
optimization. We are currently working towards incorporating
other essential features in our proposed RIS-assisted mmWave-
mMIMO system and analyzing its performance in realistic InF
conditions. The upcoming work will consist of detailed math-
ematical system model with incorporated updated features.
These essential features are, 1) integrating AI-driven tech-
niques in RIS-aided 6G system for designing, optimizing, and
reconfiguration of large metasurfaces according to the realistic
environmental conditions, 2) performance of RIS-aided 6G
system in the presence of practical sectored antennas or custom
designed antennas, and their impact on interference reduction
in indoor IIoT networks, and 3) incorporating human blockage
conditions in RIS-aided 6G system for InF environment.
REFERENCES
[1] S. Zeb et al., “Industrial digital twins at the nexus of nextG wireless
networks and computational intelligence: A survey,” Journal of Network
and Computer Applications, p. 103309, 2022.
[2] S. Zeb, A. Mahmood et al., “Industry 5.0 is coming: A survey on
intelligent nextG wireless networks as technological enablers,” arXiv
e-prints: 2205.09084, 2022.
[3] J. Zhang, E. Bj¨
ornson, and others., “Prospective multiple antenna
technologies for beyond 5G,” IEEE Journal on Selected Areas in
Communications, vol. 38, no. 8, pp. 1637–1660, 2020.
[4] Y. Zeng and X. Xu, “Toward environment-aware 6G communications
via channel knowledge map,” IEEE Wireless Communications, vol. 28,
no. 3, pp. 84–91, 2021.
[5] S. A. Busari et al., “Millimeter-Wave massive mimo communication for
future wireless systems: A survey,” IEEE Commun. Surv. Tutor, vol. 20,
no. 2, pp. 836–869, 2018.
[6] A. Mahmood et al., “Industrial iot in 5G-and-beyond networks: Vision,
architecture, and design trends,” IEEE Transactions on Industrial Infor-
matics, vol. 18, no. 6, pp. 4122–4137, 2022.
[7] K. Hiltunen, M. Matinmikko-Blue, and M. Latva-Aho, “Interference
between 5G indoor micro operators utilizing beamforming and dynamic
tdd in 26 ghz band,” in 2019 IEEE Wireless Communications and
Networking Conference (WCNC). IEEE, 2019, pp. 1–6.
[8] Q. Wu et al., “Intelligent reflecting surface enhanced wireless network
via joint active and passive beamforming,” IEEE Transactions on
Wireless Communications, vol. 18, no. 11, pp. 5394–5409, 2019.
[9] E. Basar et al., “Wireless communications through reconfigurable intel-
ligent surfaces,” IEEE Access, vol. 7, pp. 116 753–116 773, 2019.
[10] G. C. Alexandropoulos et al., “Reconfigurable intelligent surfaces and
metamaterials: The potential of wave propagation control for 6g wireless
communications,” arXiv preprint arXiv:2006.11136, 2020.
[11] A. Kapoor, R. Mishra, and P. Kumar, “Frequency selective surfaces
as spatial filters: Fundamentals, analysis and applications,” Alexandria
Engineering Journal, vol. 61, no. 6, pp. 4263–4293, 2022.
[12] D. Sarkar et al., “An electromagnetic framework for the deployment
of reconfigurable intelligent surfaces to control massive mimo channel
characteristics,” in 2020 14th European Conference on Antennas and
Propagation, 2020, pp. 1–4.
[13] S. Zeng et al., “Reconfigurable intelligent surfaces in 6g: Reflective,
transmissive, or both?” IEEE Commun. Lett, vol. 25, no. 6, pp. 2063–
2067, 2021.
[14] E. Basar et al., “Indoor and outdoor physical channel modeling and
efficient positioning for reconfigurable intelligent surfaces in mmWave
bands,” IEEE Trans Commun, vol. 69, no. 12, pp. 8600–8611, 2021.
[15] S. Ju et al., “Millimeter wave and sub-terahertz spatial statistical channel
model for an indoor office building,” IEEE Journal on Selected Areas
in Communications, vol. 39, no. 6, pp. 1561–1575, 2021.
[16] S. Zeb, A. Mahmood et al., “Analysis of beyond 5G integrated com-
munication and ranging services under indoor 3D mmWave stochastic
channels,” IEEE Transactions on Industrial Informatics, pp. 1–1, 2022.
