1

Towards Multi-Functional 6G Wireless Networks:

Integrating Sensing, Communication and Security
Zhongxiang Wei, Member, IEEE, Fan Liu, Member, IEEE, Christos Masouros, Senior Member, IEEE,
Nanchi Su, Student Member, IEEE, and Athina P. Petropulu, Fellow, IEEE

Abstract—Integrated sensing and communication (ISAC) has framework, ISAC can improve spectral and energy efficiencies,
recently emerged as a candidate 6G technology, aiming to thus addressing the problem of spectrum congestion and at
unify the two key operations of the future network in a the same time reducing hardware and signaling costs, referred
spectrum/energy/cost efficient way. ISAC systems communicate
and sense for targets using a common waveform, a common to as integration gain. Further, by exploiting the possibility
hardware platform and ultimately the same network infrastruc- to the co-design of the two functionalities, ISAC can enable
ture. Nevertheless, the inclusion of information signalling into communication-aided sensing and sensing-aided communica-
the probing waveform for target sensing raises challenges from tion. Hence, it can considerably improve sensing and commu-
the perspective of information security. At the same time, the nication performance, referred to as coordination gain. Benefit-
sensing capability incorporated in the ISAC transmission offers
unique opportunities to design secure ISAC techniques. This ing from the above merits, ISAC can enable emerging appli-
overview paper discusses these unique challenges and opportu- cations, including enhanced localization and tracking, drone
nities for next generation of ISAC networks. We first briefly monitoring/management, human activity recognition, vehicle
discuss the fundamentals of waveform design for sensing and platooning, environmental monitoring, protocols and network-
communication. Then, we detail the challenges and contradictory level sensing, sensing-aided beam training/tracking/prediction,
objectives involved in securing ISAC transmission, along with
state-of-the-art approaches to ensure security. We then identify sensing-aided resource allocation (such as cell handover, band-
the new opportunity of using the sensing capability to obtain width/beamwidth/power allocation).
knowledge target information, as an enabling approach against Nevertheless, ISAC comes with unique security challenges,
the known weaknesses of PHY security. Finally, we illustrate arising due to the shared use of spectrum and the broadcast
some low-cost secure ISAC architectures, followed by a series nature of the wireless transmission. The inclusion of infor-
of open research topics. This family of sensing-aided secure
ISAC techniques brings a new insight on providing information mation messages into the radar probing signal makes the
security, with an eye on robust and hardware-constrained designs communication susceptible to eavesdropping by the target.
tailored for low-cost ISAC devices. Indeed, the target that is being sensed can potentially exploit
the information-bearing signal, and detect the confidential
message intended for the communication destinations [2]. This
I. I NTRODUCTION raises a unique and very interesting tradeoff for the transmitter.
The 6G network, not only an improvement or extension of On one hand, it wishes to illuminate the target by focusing
existing communication technology but also a great paradigm power towards its direction, and on the other hand, it has to
revolution, is envisioned as the new engine of the future limit the useful signal power that reaches the target to prevent
intelligent world. In addition to connecting communication eavesdropping.
nodes, 6G will support ubiquitous sensing, connectivity, and A possible solution to the aforementioned security chal-
intelligence. Among the exciting features of 6G, sensing lenge is to apply cryptographic techniques at high layers of
will rise from an auxiliary functionality to a basic service, the network stack to encrypt the confidential data prior to
providing an extra dimension of capability of the network transmission. However, such solutions have several limitations,
[1]. This has prompted the recent research interest of in- such as a tedious secret key management/maintenance pro-
tegrated sensing and communication (ISAC), a technology cess, unprovable security performance in the presence of a
that enables the integration of sensing and communication computationally strong eavesdropper (Eve), and difficulty in
functionalities with a single transmission, a single device, identifying a compromised secret key. Physical layer (PHY)
and ultimately a single network infrastructure. By exploiting security, an information theory-based methodology could be a
a common spectral, hardware platform and signal processing complementary approach for securing wireless transmission.
By exploiting the channel variability between the Eves and the
Zhongxiang Wei is with the College of Electronic and Information Engi-
neering, at Tongji University, Shanghai, China. Email: z wei@tongji.edu.cn legitimate users (LU)s, the signal quality that the Eves receive
Fan Liu is with the Department of Electronic and Electrical Engineering, can be degraded to the degree that the Eves cannot extract the
at Southern University of Science and Technology, Shenzhen, China. Email: message even when they have full knowledge of the secret
liuf6@sustech.edu.cn
Nanchi Su and Christos Masouros are with the Department of Electronic key [3]. Despite decades of research, the major limitation of
and Electrical Engineering, at University College London, London, UK. a large class of PHY security solutions stems from the either
Email: {nanchi.su.18, c.masouros}@ucl.ac.uk extremely optimistic or overly pessimistic assumptions with
Athina P. Petropulu is with the Department of Electrical and Computer
Engineering, at Rutgers University, NJ, USA. Email: athinap@rutgers.edu regards to what can be known about the Eve. Some methods
Correponding authors: Fan Liu and Zhongxiang Wei. require full knowledge or statistical information of the Eves’
 2

channels. Some methods do not require any knowledge on the B. Waveform Design for ISAC
Eves’ channels, such as transmitting artificial noise to jam the
entire space except the legitimate destination. However, such ISAC waveform designs can be categorized into sensing-
methods do not make good use of the available bandwidth centric, communication-centric, and joint designs, as summa-
by transmitting a signal that does not bear communication rized in Table I.
information, as summarized in in Table I. There has also been Sensing-Centric Design: Sensing-centirc design integrates
recent works that monitor the changes caused by Eves’ interac- communication messages into a classical sensing waveform,
tion with the radio frequency electromagnetic wave field at the and hence has high compatibility to the radar architecture.
the PHY to infer Eves’ positions [4]. However, the sensing and Early sensing-centric design works include pulse interval
communication in [4] are performed separately, and hence the modulation, where the interval between radar pulses is utilized
obtained information of the Eves may be outdated, especially for communication. There have also been designs that leverage
in high mobility scenarios. This approach too is not spectrally the concepts of index modulation, or generalized spatial modu-
efficient, as spectral resources are dedicated for sensing only. lation for waveform design [6]. Another sensing-centric design
Interestingly, the joint sensing and communication mecha- approach is to sense the target in the mainlobe of the radar
nism of ISAC ushers in new opportunities for secure design, beampattern, while embedding information in the beampattern
where the additional sensing functionality can serve as a sidelobes [7]. Nevertheless, since the communication symbols
support to facilitate the provision of security. Motivated by are generally embedded into the radar pulses, the sensing-
the aforementioned issue, this article overviews the sensing- centric design results in a low data rate, limited by the pulse
aided secure designs together with the characteristics of ISAC. repetition frequency of the radar, which is well below 5G/6G
Starting from the fundamentals of ISAC systems, we first requirements.
examine a novel secure ISAC design. Then, we discuss a Communication-Centric Design: Communication-centric
practical robust secure ISAC design, where knowledge of designs leverage the standardized communication waveforms,
the target and communication users is imperfectly obtained. protocols and architectures for sensing. For example, pilot
Further, some hardware-efficient secure ISAC architectures signals and frame preambles that have good auto-correlation
are reviewed. Open challenges are then identified, before properties and are typically used for channel estimation or
concluding this article. multi-user access, have been recently employed for sensing
targets [8] [9]. Also, standards-relevant communication wave-
forms, such as the IEEE 802.11p vehicular communication
II. T HE F UNDAMENTALS OF ISAC waveform, have been used for sensing targets in vehicular
applications. These communication-centric ISAC designs can
As an early stage of ISAC, communication and radar realize sensing functionality without sacrificing the communi-
spectrum sharing (CRSS) has been investigated from the cation performance, thereby obtaining high data rate. However,
perspective of spectrum sensing, dynamic spectrum access, the pilot signal, frame preambles and communication wave-
and mutual interference mitigation [5], so that communication forms are not dedicatedly designed for sensing. Accordingly,
and radar systems can share the spectrum without significantly the main drawback of the communication-centric designs lies
interfering with each other. As a further step, ISAC can in the poor, scenario-dependent and difficult-to-tune sensing
realize not only spectral coexistence, but also the shared performance.
use of hardware platform and even network architecture, as Joint Design: In joint-design ISAC approaches, the beam-
shown in Fig. 1. In addition to providing communication pattern is designed to meet an ideal radar beampattern, while
and sensing functionalities, ISAC systems lend themselves to ensuring a high signal-to-interference-plus-noise ratio (SINR)
communication-aided sensing, and sensing-aided communica- at LUs’ for communications [10]. Also, the sum-weighted
tion functionalities. Let us start by discussing the fundamentals sensing and communication quality can also be exploited as
of ISAC, and then elaborate on secure ISAC transmission. an objective function, further leading to a Pareto-optimality of
the multi-objective optimization. Apart from the optimization-
oriented research, the joint-design has also been investigated
from the perspective of information theory, such as the channel
A. Sensing Basics
coding design [11], as well as the theoretic trade-off between
While communication aims to accurately convey the infor- the transmission rate and sensing performance design [12].
mation to a receiver, sensing aims to extract target information Evidently, joint design involves dedicated optimization of both
from the target echoes. Consequently, the useful information functionalities and enables scalable performance trade-offs
for sensing is not in the sensing waveform but in the target between them. It enables flexible use of time, frequency, and
return. Interestingly, since that sensing and communication spatial resources, thereby achieving both high throughput and
performances are evaluated by different key performance in- sensing reliability.
dicators, ISAC waveform design should take different metrics In addition to academic research, there have been ex-
into consideration for implementing the dual functionalities. tensive industrial activities focusing on ISAC, including
This typically incurs conflicting design objective between the 3GPP (such as S1-214036/214056/214100/214101, R1-
sensing and communications, which needs to be carefully 2110894/2104724, and R2-210049), IEEE standards (such
balanced as detailed in the next subsection. as 802.11bf WLAN Sensing, 802.15.22.3-2020, and 802.11-
 3

TABLE I
A B RIEF S UMMARY OF T HE E XISTING ISAC AND PHY S ECURITY D ESIGNS .

Relevant
Design Principles Pros Cons Remarks
Techniques
Side-lobe based ISAC [7]
Sensing-centric Integrate communication High compatibility
Index or generalized spatial to radar systems Low data rate
design into radar systems
modulation ISAC [6]
Use 802.11p (i) These security-
Leverage the existing communication waveform, agnostic techniques
Communication communication or single carrier PHY frame High compatibility Poor sensing may not preserve the
waveform or protocols of 802.11ad for sensing to communication performance confidentiality of the
-centric design for sensing systems data;
[8] [9] Use OFDM signals
for detection (ii) Even high layer
Optimize sensing subject to encryption/authentica
communication quality tion can be applied,
ISAC Involves optimization of the PHY information
Design one system or the other, Optimize communication, contained in the
subject to certain and meanwhile realize the probing waveform
constraints for sensing functionality can be exploited by
communication or the Eve;
sensing accuracy [10] Optimize sum-weighted
function of communication High performance
of sensing and Need dedicated (iii) Then the Eve is
Joint design and sensing able to decipher the
Channel coding for communication waveform design
data.
improving reception and
channel estimation
Performance analysis by performance at the
information theory destination [12]
Theoretic trade-off analysis
between the communication
and sensing [11]
Exploit the EM wave Encoder design for The secrecy rate
change caused by the providing secrecy even can be close to that Though the sensing
Eve; when the Eve moves to obtained with and communication
Sensing-aided improve its eavesdropping are performed on the
hindsight, had the Only theoretic analysis
secure design then proactively and capability transmitter same channel use, the
is given optimal secure
[4] causally infer the EveĆs Characterize the secrecy obtained the
path loss to assist secure waveform is still
rate for any sequence of Eveÿs condition unknown
design path loss for the Eve non-causally
Exploit the channel Need multiple antennas
disparity between the LU Adjustable secrecy
Secure precoding rate for exploiting spatial (i) Sensing is
and Eve for sending signals disparity disabled;
PHY Secure
Send isotropic AN towards No requirement of Impracticality of
Design network-level (ii) Rely either on
the null-space of LUs Eveÿs CSI interference control
Artificial Noise extremely optimistic
Sensing- Inject spatial AN towards More energy- Need Eveÿs full or or pessimistic
efficient than statistical CSI assumption with
agnostic secure the direction of Eves isotropic AN regards to the Eveÿs
design [3] Difficult for Eve to Need agreement among condition;
PHY Use LUsĆPHY attributes impersonate;
authentication for authentication No requirement of the communication (iii) The transmitter
Eveÿs CSI parties is not capable of
proactively sensing
Constellation Low complexity; the Eves.
Other hardware-efficient rotation and noise No requirement of Reduced throughput
secure design aggregation Eveÿs CSI

2020), and ITU recommendations (such as ITU-T Y.4809 and as the target might be an Eve, the angle of the sensing beam
ITU-T X.1080.2). that enters the SCNR objective is the same as the angle of
the Eve, as shown in Fig. 2. This implies that the target has
III. F ROM D UAL -F UNCTIONAL TO M ULTI -F UNCTIONAL : high reception SINR on the embedded communication signal,
I NTEGRATING S ECURITY INTO ISAC which significantly increases the susceptibility of information
to eavesdropping by the target. Therefore, one should carefully
In this section, we discuss security issues around ISAC and strike a trade-off between sending sufficient power towards the
how the sensing functionality can be judiciously utilized for target’s direction for sensing, while limiting the useful signal
benefiting the provision of information security. power at the target to prevent eavesdropping. In the following,
we examine recent research results of sensing-aided secure
A. The Unique Security Challenges and Opportunities of ISAC ISAC techniques.
The ISAC transmitter needs to focus its power towards The ISAC transmitter is able to sense the angle of arrival
directions that contain targets and ensure that the target echo at (AoA) of the echo signal reflected from the target, and infer
the receiver has good enough signal-to-clutter-plus-noise ratio the target’s position based on the received signal strength.
(SCNR) for achieving certain sensing performance. However, Leveraging this round-trip channel, the wiretap channel from
 4

phase shifters sensing & com

Communication module
vehicles

RF chain

...
Digital

. . .
001110

. . .
bits Precoding
sequence modulation

...
RF chain LUs

Sensing-aided Proactive Design

clutter
RF chain

...
Radar Matched target

. . .
signal

. . .
filter echo (Eve)
processing signal
RF chain

...
Target detection and
estimation module

Fig. 1. A generic joint communication and sensing design for ISAC. At transmitter side, the modulated signal is manipulated by digital precoding at baseband,
then passes through radio frequency (RF) chains, and finally is dissipated by antennas. On the other hand, while the reflected echo is analyzed for target
detection, the sensing results also assist secure waveform design in a proactive and causal manner.

the ISAC transmitter to the target can be estimated. Hence, the For example, given N antennas arranged in uniform linear
proactively and casually obtained wiretap channel, or the tar- structure with half-wavelength spacing, the angular resolution
get’s AoA as a minimum, can be exploited to design a number is approximately calculated as N2 (in rad) [10], which means
of secure approaches such as secure beamforming, artificial the targets within that angular interval can not be detected
noise, and cooperative jamming, amongst many others. individually. When the target’s position can only be roughly
The unique ISAC transmission however requires the afore- sensed within an angular region, a wider beam needs to be
mentioned approaches to be redesigned for achieving the “in- formulated towards that region to avoid missing the target.
band” dual functionality. As an example, in designing a secure However, focusing the beam to a region of space inevitably
dual functional transmission, one can optimize the sensing leads to an increased possibility of the information leakage,
performance by maximizing the echo signal’s SCNR at the giving rise to a need for robust secure waveform design.
ISAC’s receiver, while limiting the eavesdropping SINR at the When the target is only known to be located within a certain
target and at the same time guaranteeing the signal’s SINR angular region of space, the robust secure waveform can be
at LUs above a certain threshold. This equivalently improves obtained by minimizing the sum of the target’s reception
the value of the secrecy rate, calculated as the achievable SINR at the possible locations in this angular interval. In this
rate difference between the LUs and target. Alternatively, way, the achievable rate of the target can be upper-bounded,
one can maximize the secrecy rate and meanwhile ensure thus guaranteeing information security. On the other hand,
the echo’s SCNR at the ISAC receiver for guaranteeing the when the LUs’ channels are also not perfectly known by the
radar’s functionality. Though designing of a sensing-aided ISAC transmitter, the channel estimation error can be generally
secure waveform is not convex in nature, due to the fractional- formulated using bounded or un-bounded error models [13].
structured SINR and SCNR constraints of the sensing and With the bounded or un-bounded error model, ensuring the
communication functionalities, there has been extensive opti- LUs’ SINR can be further transformed into deterministic or
mization tools for handling these typical fractional-structured probabilistic constraints, which can be readily handled by a
optimizations in ISAC systems. Detailed discussions can be series of established stochastic optimization tools [13].
found in [2] [5] [10]. Note that in the rare case that the target Let us consider the scenario shown in Fig. 2, where the
and LU are in the same direction and both have strong LOS possible angular interval of the target is [−5◦ , 5◦ ], while
channels, their channels are strongly correlated. In this context, the channel estimation error of the LUs follows Gaussian
ensuring security at the PHY layer is extremely challenging, distribution with variance 0.05. There are 4 LUs and their
where authentication and encryption secure techniques are still SINR threshold is 40 dB. The power budget is 20 dBm.
needed at the higher layers. The objective of the secure waveform optimization is to
suppress the targets’ reception SINR, subject to per-LU’s
B. Robust Secure ISAC Waveform Design SINR requirement, while ensuring the resulting waveform
In practice, a target’s position may not be always perfectly approximates the desired sensing beampattern. As observed
obtained, due to sensing error and finite detection resolution. in Fig. 2(a), a narrow beampattern is obtained when the
 5

Beam gain (dBm)

eavesdropping
SINR
Uncertainty
interval [-ƃ, ƃ]

same direction
imperfect CSI and 10o target location uncertainty
--- perfect CSI and precise target location
communication echo with
SINR SCNR (a) Angle (degree)
Wider beampattern --- perfect CSI and precise target location
for sensing imperfect CSI and 10o target location uncertainty
Robust statistical CSI and 10o target location uncertainty

Secrecy rate (bit/s/Hz)

waveform

ISAC

(b)
Communication SINR threshold

Fig. 2. A practical scenario, where the target’s position is roughly sensed within an uncertainty interval, and the LUs’ channels are also imperfectly known.
(a) The width of the beampattern is adaptively manipulated in different scenarios for sensing the target. (b) By proactively sensing the target, a high level of
secrecy rate is achieved.

target location is accurately sensed. Leveraging the proactively logue beamforming. However, in both fully-digital or hybrid
obtained location, the transmitter is able to manipulate the ISAC, the required number of RF chains is no smaller than the
dissipated waveform to suppress the eavesdropping SINR of total number of data streams for multi-user communications.
the target, thereby improving the secrecy rate in Fig. 2(b).
When the target’s location can only be imperfectly sensed, a To remove the expensive and power-consuming digital-to-
wider beampattern is formed, directing the same power over analogue converters (DAC)s, a more hardware-efficient secure
the possible region, with reduced power gain of mainbeam. ISAC technique, built on the concept of directional modulation
Nevertheless, by suppressing the sum of the target’s SINR at (DM) is emerging, where parasitic antennas are used as main
the possible locations in the angular interval, a high level of components in the transmitter. Aided by the LU’s channel state
secrecy rate is achieved, even if the ISAC transmitter only information (CSI), symbol modulation happens at the antenna
knows the statistics of the LUs’ channels. level instead of the baseband level, and the received beam
pattern at the LUs is treated as a spatial complex constellation
point. In particular, the constructed signal of the LUs does not
C. Hardware Efficient Secure ISAC Design necessarily align with the desired symbols, but can be pushed
At millimeter wave band, a candidate frequency for 5G/6G away from the detection thresholds of de-modulation, built
systems, low-cost and -power consuming hardware is pre- on the concept of constructive interference (CI) regions [13].
ferred. However, the hardware limitations may jeopardize An example is illustrated by Fig. 3 for quadrature phase shift
the sensing and communication performance, and impor- keying (QPSK). Since the decision thresholds for QPSK are
tantly the security of the transmission. A recent abundance the real and imaginary axes, the constructed symbols (denoted
of hardware efficient techniques that have been developed by blue dots) at the LUs can be judiciously pushed away
for communication-only systems can be leveraged to design from both the real and imaginary axes, where the resultant
hardware-informed secure ISAC transmission [14]. On the increased distance with respect to the detection threshold
feasibility of secure waveform with high hardware efficiency, benefits the LUs’ communication quality. In a similar vein,
one approach is to reduce the RF chains through analog the symbols can be constructed for the LUs with higher-
architectures that involve phase shifters (PS)s and/or switchers, order modulations. On the other hand, with the proactively
as illustrated in Fig. 1. This hybrid ISAC involves low- obtained Eve’s information, one can intentionally locate the
dimensional digital beamforming and high-dimensional ana- Eve’s received symbols (denoted by red stars) into destructive
 6

..........CI-aided
Power
LU .... QPSK
PA parasitic

...
Divider antennas
LO Eve
DM signal generation
2 . . . .. .
Modu-
QPSK2
. .
LU
16QAM
.. .. .. ..
lator
. .. .

...
Com-
Modu- biner PA
. .

...
lator 1 Eve
LO
. .
QPSK1
CDA signal generation

Fig. 3. The DM uses power amplifier (PA)s and parasitic antennas as main components, while CDA uses modulators, linear combiners and PAs as main
components.

regions of the signal demodulation, which further impedes the Secure ISAC Design for 5G/6G KPIs: In recent years,
Eve’s intercepting behavior at a symbol level. ultra-reliable and low-latency, and massive-device communi-
Another hardware-efficient architecture, namely constella- cations have received much attention in 5G/6G applications.
tion decomposition array (CDA), also has a high potential Those applications involve new metrics and protocols, such
for securing ISAC. A simplified block of the CDA is shown as latency, reliability, grant-free massive access, short packets,
in Fig. 3, including the local oscillators, modulators, linear and so on. Rethinking secure ISAC techniques to align with
combiners and PAs, but the costly DACs are completely these stringent requirements, and also to maintain a low level
avoided [14]. Evidently, a high-order quadrature amplitude of complexity and overhead is a fertile area of research.
modulation (QAM) can be treated as a vectorial combination On Compatibility of Secure ISAC and 5G NR: 5G new
of several low-order QAM/PSK signals. For example, a 16- radio (NR) has standardized a series of waveforms, including
QAM signal can be seen as a combination of a QPSK1 and a but not limited to filter-OFDM, DFTS-OFDM, and FBMC-
QPSK2 signals, where the superscript denotes the normalized QAM. Also, 5G NR has also proposed adaptive wireless
Euclidean distance between two adjacent symbols. By properly interface configuration, such as changeable frame structure and
controlling the array with the LU’s CSI, a LU can see a adaptive 15-120 KHz carrier spacing. With different commu-
correct combination of the intended signal, while any Eve nication environments and specific performance requirement,
(including the sensing target) located at a different angle will how to leverage the flexible waveform specification and wire-
obtain a distorted signal in demodulation. Also, since the CDA less interface configuration? Essential work is needed to bridge
transmits low-order modulation signals with a low level of the gap between theory and implementations.
peak-to-power-ratio, the stringent linearity requirement of the
Network Level ISAC Design and Secure Performance
PAs is properly relaxed.
Analysis: Networking design has been investigated for cellu-
IV. O PEN C HALLENGES AND F UTURE W ORKS lar communication systems, where coverage probability and
ISAC-relevant design is still broadly open, and the remain- ergodic capacity are analyzed in a systematic manner. This
ing challenges can benefit from the communications literature. network level investigation advises networking planing and
Radar Location and Identity Privacy-Preserving Design: engineering design with an eye to the interests of the whole
On the evolution road of ISAC design, the CRSS system still system. While the existing ISAC-related research is investi-
has its market. Designed to control mutual-interference, there gated in simple scenarios, considering the heterogeneity and
are parameters transformed to one system that contain implicit high nodes density in future communication systems, the
information about the other. This raises privacy concerns for systematic ISAC designs need fundamental research.
the two systems, and especially for the military radar. Recent ISAC for New Security Metrics: Apart from data confi-
research has unveiled some machine-learning based schemes, dentiality, the concept of security has been greatly generalized
which exploit the information contained in the precoder to in 5G/6G communications, such as covertness and privacy.
infer the radar’s location [15]. As a result, how to exchange In some scenarios, users want to communicate with others
parameters between radar and communication units without covertly, referred to as low-probability of detection communi-
loss of each other’s privacy, while maintaining a minimum cation. Coordinated with sensing, it becomes easier to detect
level of mutual interference remains an open challenge. an intruding adversary’s information, which is then exploited
 7

to design covert waveform to hide an ongoing communication. [11] M. Kobayashi, G. Caire, and G. Kramer, “Joint state sensing and
On the other hand, sensing may be leveraged by an adversary communication: optimal tradeoff for a memory-less case,” in IEEE Proc.
ISIT’18, Vail, USA, pp. 111-115.
to violate users’ privacy, such as sensing pedestrians’ non- [12] W. Zhang, S. Vedantam, and U. Mitra, “Joint transmission and state
shared positions and trajectories, imaging users’ indoor activ- estimation: a constrained channel coding approach,” IEEE Trans. Inf.
ities. Hence, it is demanding to rethink the role of ISAC from Theory, vol. 57, no. 10, pp. 7084–7095, Oct. 2011.
[13] Z. Wei et al., “Multi-cell interference exploitation: enhancing the power
the perspective of new security metrics. efficiency in cell coordination” IEEE Trans. Wireless Commun., vol. 19,
no. 1, pp. 547-562, Jan. 2020.
[14] N. S. Mannem et al., “A mm-wave transmitter MIMO with constella-
V. CONCLUSIONS tion decomposition array for key-less physical secured high-throughput
This article has discussed the exciting intersection of ISAC links,” in Proc. IEEE RFIC’21, Denver, US, pp. 199-202.
[15] A. Dimas et al., “On radar privacy in shared spectrum scenarios,” in
and security. Starting from the fundamentals of the ISAC, Proc. IEEE ICASSP’19, Brighton, UK.
we first have introduced the methodology of the waveform
design for joint sensing and communication. Then, we have
examined the sensing-aided secure ISAC techniques to prevent
the confidential signal embedded in the probing waveform Zhongxiang Wei is an associate professor of Electronic and Information
from being eavesdropped upon by the sensing target. Finally, Engineering at Tongji University. He received the Ph.D. from the University
the recent interests in robust and hardware-efficient secure of Liverpool (2017). He was a postdoc researcher at the University College
London (2018-2021), and was a Research Assistant at the A∗STAR Singapore
ISAC has been reviewed. This family of sensing-aided secure (2016-2017). He has served as a TPC chair/member of various international
ISAC design offers a broad field of preserving information flagship conferences. He was a recipient of an Exemplary Reviewer of the
security in a proactive manner, which holds the promise of IEEE TWC, the Outstanding Self-Financed Students Abroad in 2018, and
the A∗STAR Research Attachment Programme in 2016. His interests include
exciting research in the years to come. MIMO systems, PHY security, and anonymous communication designs.

ACKNOWLEDGEMENT
Z. Wei would like to acknowledge the financial support of
Fan Liu is an Assistant Professor of the Department of Electronic and Electri-
the NSFC under Grant 62101384, as well as of the Chongqing cal Engineering, Southern University of Science and Technology. He received
Key Laboratory of Mobile Communication Technology un- the Ph.D. and the BEng. degrees from Beijing Institute of Technology, China,
der Grant cqupt-mct-202101. C. Masouros would like to in 2018 and 2013, respectively. He was a Marie Curie Research Fellow at
University College London, UK, from 2018 to 2020. He is an Associate
acknowledge the financial support of the EPSRC under Grant Editor of the IEEE OJSP and IEEE COMML, and a Guest Editor of the
EP/R007934/1. A. P. Petropulu would like to acknowledge the IEEE JSAC and IEEE WCM. He is also the Founding Academic Chair of the
financial support of the ARO under Grant W911NF2110071. IEEE ComSoc ISAC Emerging Technology Initiative. He was the recipient of
the 2021 IEEE SPS Young Author Best Paper Award, and the 2019 Chinese
Institute of Electronics Best PhD Thesis Award. His research interests include
R EFERENCES ISAC, vehicular networks, and mmWave communications.
[1] Y. Cui et al., “Integrating radio sensing and communications for ubiq-
uitous IoT: applications, trends and challenges,” IEEE Network, vol. 35,
issue. 5, pp. 158-167, Oct. 2021.
[2] N. Su, F. Liu, and C. Masouros, “Secure radar-communication systems Christos Masouros is the Professor of Electrical and Electronic Engineering
with malicious targets: integrating radar, communications and jamming at UCL. He received his PhD from the University of Manchester, UK (2009).
functionalities,” IEEE Trans. Wireless Commun., vol. 20, no. 1, pp. 83- His interests include wireless communications and signal processing with
95, Jan. 2021. specialty on Large Scale Antenna Systems and Interference Exploitation. He
[3] M. Bloch et al., “An overview of information-theoretic security and has held a Royal Academy of Engineering Research Fellowship (2011-2016).
privacy: metrics, limits and applications,” IEEE J. Sel. Inf. Theory, vol. He is co-author of the 2021 IEEE SPS Young Author Best Paper Award (F.
2, no. 1, pp. 5-22, Mar. 2021. Liu). He is an Editor and Guest Editor for IEEE TWC/TCOM/JSTSP/JSAC
[4] M. Tahmasbi, M. Bloch, and A. Yener, “Learning an adversary’s actions and Vice-Chair of the IEEE ComSoc ISAC Emerging Technology Initiative.
for secret communication,” IEEE Trans. Info. Theory, vol. 66, no. 3, pp.
1607-1624, Mar. 2020.
[5] B. Li, A. P. Petropulu, W. Trappe, “Optimum co-design for spectrum
sharing between matrix completion based MIMO radars and a MIMO
communication system,” IEEE Trans. Sig. Process., vol. 64, no. 7, pp. Nanchi Su (S’18) received the B.E. and M.E. degrees from Harbin Institute of
4562-4575, Jul. 2016. Technology, China, in 2015 and 2018, respectively. She is currently pursuing
[6] T. Huang et al., “MAJoRCom: a dual-function radar communication her Ph.D. degree at UCL, U.K. Her research interests include CI design, PHY
system using index modulation,” IEEE Trans. Signal Process., vol. 68, security, and ISAC signal processing.
no. 5, pp. 3423-3438, May 2020.
[7] A. Hassanien et al., “Dual-function radar communications: Information
embedding using sidelobe control and waveform diversity,” IEEE Trans.
Signal Process., vol. 64, no. 8, pp. 2168–2181, Apr. 2016.
[8] P. Kumari, N. Myers, and R. W. Heath, “Adaptive and fast com-
Athina P. Petropulu is a Distinguished Professor of Electrical and Com-
bined waveform beamforming design for mmWave automotive joint
puter Engineering at Rutgers University. Her interests include radar signal
communication-radar,” IEEE J. Sel. Topics Sig. Process., vol. 15, no.
processing and PHY security. She received the Presidential Faculty Fellow
4, pp. 996-1012, Jun. 2021.
Award (1995) from NSF and the US White House, and the 2012 IEEE Signal
[9] P. Kumari et al., “IEEE 802.11ad-based radar: an approach to joint
Processing Society (SPS) Meritorious Service Award. She is IEEE and AAAS
vehicular communication-radar system, in IEEE Trans. Veh. Tech., vol.
Fellow. She is co-author of the 2005 IEEE Signal Processing Magazine Best
67, no. 4, pp. 3012-3027, Apr. 2018.
Paper Award, the 2020 IEEE SPS Young Author Best Paper Award (B. Li),
[10] F. Liu et al., “Toward dual-functional radar-communication systems:
the 2021 IEEE SPS Young Author Best Paper Award (F. Liu), and the 2021
optimal waveform design,” IEEE Trans. Signal Process., vol. 66, no.
Aerospace and Electronic Systems Society Barry Carlton Best Paper Award.
16, pp. 4264–4279, Aug. 2018.
She is currently President-Elect of the IEEE SPS.