A Survey on Security Issues of 5G NR:

A Survey on Security Issues of 5G NR:

Perspective of Artificial Dust and Artificial Rain
Perspective of Artificial Dust and Artificial Rain
a b,*
Misbah Shafi , Member, IEEE, Rakesh Kumar Jha , Senior Member, IEEE, Manish Sabraj c

a,b,c
Department of Electronics and Communication, Shri Mata Vaishno Devi University, J&K India

Abstract—5G NR (New Radio) incorporates concepts of novel envisaged that 5G is likely to be 100x times (10 mW/Mbps/sec
technologies such as spectrum sharing, D2D communication, in IMT-advance) more energy efficient in future IMT than in
UDN, and massive MIMO. However, providing security and IMT-2000(100mW/Mbps/sec) [7]. Radio resource
identifying the security threats to these technologies occupies the management is another challenging field of research associated
prime concern. This paper intends to provide an ample survey of
with the next generation 5G communication network. It
security issues and their countermeasures encompassed in the
technologies of 5G NR. Further, security concerns of each includes spectrum allocation, user association, and
technology are defined mathematically. Thereby defining the management of power [8]. Communication efficiency is
impact on the factors of security. Moreover, a methodology is another aspect that has grabbed serious attention in the field of
developed in which the influence on security due to artificially research. To encounter the communication efficiency in 5G,
generated rain and artificially generated dust on the wireless various techniques such as spectrum using NOMA, network-
communication network is studied. By doing so, an attacking assisted interference suppression and cancellation, coordinated
scenario is identified, where a half-duplex attack in D2D multipoint joint transmission reception, three-dimensional
communication is attained. Half-duplex attack specifies the attack MIMO, full dimensional MIMO are required to be improvised
solely on the downlink to spoof the allocated resources, with
[9]. In addition to it, security is another challenging feature that
reduced miss-rate. Thus, ultra-reliable and adequate advances
are required to be addressed to remove the obstacles that create a occupies immense attention in the field of research.
hindrance in achieving the secured and authenticated With the enhancement of incipient technologies and the
communicating network. massive progression in the figure of devices, the number of
small cell BSs is increased. These are arranged with the
centralized macro BS supporting 1000x capacity, therefore,
Index Terms—5G NR, security, Artificial Rain, Artificial Dust, creating an increased number of handovers. It results in the
D2D, miss-rate, half-duplex attack foundation of feasible sites for the malicious attacker, thus
creating a threat to the field of security [10]-[12]. Moreover,
I. INTRODUCTION next generation wireless communication networks (5G)
With the remarkable upsurge of requirements such as user introduces flat architectures, broad involvement of cloud in
experienced high data rate, low latency, energy efficiency, network communication, and processing, which enhances the
spectral efficiency, UDN, coverage reliability, mobility in the susceptibility of attackers in the network [13]. Also, the
wireless communication network, 5G emerges as a satisfying network of 5G is considered as a heterogeneous network,
approach. In addition to these mobile broadband services, including Wi-Fi hotspots, microcells, small cells, femtocells.
several other services depending upon the usage scenario are Though it provides cost efficiency but increases the probability
predicted to be supported by 5G, including ultra-reliable, of attacks due to the incremented number of breaches in the
gigantic machine type information exchange, boosted mobile architecture of the network [14]. Consequently, advanced
broadband, and truncated latency communications [1]. The solutions are required to be conceived to fulfill the measures of
significant principle of next generation wireless security. The measures of cybersecurity include confidentiality
communication network lies in reconnoitering unused mm- (identity protection, communication confidentiality, and
wave high frequency band, ranging from 3~300GHz [2]. To location privacy), authentication (authorized service access),
justify the ten times more connectivity density of 5G wireless integrity, and availability [15]. Therefore the accuracy of these
networks than that of 4G, the target connectivity density of 5G parameters is required to be fulfilled. For accurate security of
is essentially not to be less than 106 / km2 [4]. Therefore, to the network, the time consumed to detect the existence of
content the stringent challenges of 5G wireless network malicious attackers in the network must be less than the rate of
communication, advanced research with respect to various transmission. This criterion is required to prevent the stark loss
characteristic features are required to be observed. Mobile edge that may be caused in the network due to the entry of malicious
computing is considered as one of the up-to-date paradigms to attackers [16].
solve the issue of burdensome computations and decrease From preceding years, research on the security of the next
latency more significantly [3], [5], [6]. Power optimization is generation 5G wireless communication network has revealed
another step towards the field of research for improving the various security issues. Numerous attacks are observed to
representative factor of energy efficiency, such as the green affect the parameters of security. Various countermeasures,
radio project. For the parameter of energy efficiency, it is including the physical layer security measures [17]-[19], were

1
proposed to thwart these attacks. Conversely, considering network, as shown in Fig. 1. The impact on security, and
security in a systematized manner is still a void to be filled in identifying generic solutions of security enhancement for each
the field of research. The technology of IoT involves the technology. The mathematical aspect of security is expressed
number of sensors to satisfy a variety of applications. It acts as for each technology. The 5G wireless communication network
an evolving technology in 5G architecture. The communication consists of the following technologies:
in IoT takes place via inter mediators (gateway) involving a
A. Massive MIMO
huge number of devices estimated to be 50 billion by 2020.
Consequently, gives rise to the number of attacking breaches Massive MIMO is most widely studied and considered as the
by the large traffic created by numerous IP devices. Thus, fundamental technology which provides the basis for research
secure communication mechanisms are required to be in the 5G generation and the next generation WCN technology
developed with low latency, less complexity, and more energy to satisfy various rising demands of service. Its main advantage
efficiency, improved quality of service, throughput, etc. These involves the potential of high spectral efficiency, energy
parameters achieve proper balance in various security efficiency, and required service demands. Irrespective of its
mechanisms. Considering the research efforts and security numerous supporting applications several limitations bound its
specification of last year, a systematized approach to project number of application such as the interference created due to
security breaches in different technologies is identified in order the numeral of antennas existing in the network, accessibility
to encounter challenges for future particularization. This paper of channel state information, possibility of realistic approach
provides a comprehensive and unified survey on the security of for deployment of the network antennas, pilot contamination,
5G technologies. The scope of the survey in different spoofing attack, MIM (Man in the Middle) attack, and DoS
technologies of research is to occupy the blank void between attack. These challenges give rise to a number of issues;
the researched topics of security and to exclude the legacies of however, security issues occupy the immense interest due to
insecure WCN. the plausibility of an eavesdropper to intercept the information.
Therefore, security management mechanisms are required to
A. Contribution: enhance protection from a malicious attacker.
In summary, the primary contribution of this paper are Various approaches were formulated to advance the security of
mentioned below to provide readers with a better the network in the technology of massive MIMO. It utilizes a
understanding spatial modulation scheme where the physical layer encryption
 We develop a systemized survey of various security technique is followed to increase the security performance of
mechanisms in different technologies of 5G. Starting from the network along with the enhancement of other parameters
the security perspective in massive MIMO, spectrum such as energy efficiency and reliability [22]. However, for
sharing, D2D, beamforming, UDN followed by IoT. A physical layer encryption, the complexity of decoding is
comprehensive picture of security research is summarized examined by mapping massive MIMO to a lattice and therefore
along with their mathematical security models. providing complexity to both quantum and classical computers
 A system model is proposed to thrive the impact of rain on [23]. Secure physical layer multicast transmission where the
the security of WCN, involving the scenario of AR to BS is known of the statistical CSI of the valid and invalid user
with consideration of lower bound of secrecy rate of a multicast
investigate the analysis of the attacking setup.
system [24]. Spoofing
 The impact of dust on the security of WCN is observed by
Direct link
involving the mechanism of AD. Based on these tractable Destination
Source
scenarios, attenuation due to AR and AD are evaluated Control link
Source
 A framework of attack in the form of a half-duplex is Source
Cellular link
proposed. Further, the possibility of such an attack on the Relay

basis of the hit rate in comparison to the conventional Relay D2D Destination Destination

Destination
attack is evaluated. Source

The organization of the paper is mentioned as: section II Beamforming

defines the security background of 5G technologies along with Massive MIMO
their mathematical security models. Section III depicts the
system model involving the artificial rain, artificial dust, and
Half Duplex attack along with mathematical calculations.
5G Smart
Smart
vehicles
Spectrum sharing city
Section IV signifies the challenges in 5G WCN with a security Smart
perspective, which are required to be addressed for secure Primary transmitter Primary receiver
IOT health
communication. Section V characterizes the conclusion Secondary
transmitter
secondary
Smart
home
obtained from the paper, respectively. A list of current security receiver

licensed shared Unlicensed

Smart
projects in WCN and the list of acronyms are mentioned in the appliances
appendix. Spectrum sharing

II. SECURITY BACKGROUND OF 5G TECHNOLOGIES UDN

In this section, we describe the security aspect of crucial
technologies incorporated in a 5G wireless communication Fig. 1. Generalized 5G architecture

2
attacks such as pilot spoofing attacks [25], bandwidth spoofing domain transmission [52], hardware quality scaling procedure
attacks [12] are prevented by addressing the techniques of [53], beamforming technique with normalized constrained
PLA. PLA in MIMO is examined on the basis of radio channel policy of power [54], amendment of AN (Artificial Noise) with
information, and interaction between multiple antennas with a low-resolution DACs (Digital to Analog Converters) [55]. AN
receiver is formulated. Consequently, the spoofing attack is facilitated design of jamming in the scenario of Racian fading
detected by using Q-learning [26]. One more approach to detect Channel [56], approach of power scaling [57], AN precoding
active spoofing attacks includes the criterion of MDL. The incorporated with RF chains [58], two AN precoding schemes
scheme of detection involves the allocation of partial power to one for the downlink followed by the second AN scheme for
place over an arbitrary sequence of detection on the sequence the downlink, phase of payload data transmission [59] and
of training of a valid receiver [27]. distributed reconfigurable MIMO forming cooperative MIMO
The BS (Base Station) procures the CSI from the uplink system [60]. Additional techniques that provide physical layer
transmission of pilots for various receivers. It gives rise to the security in massive MIMO include MF (Matched Filter)
susceptibility of contamination of the pilot signal based on the precoding combined with AN [61]. TDD integrated with
estimation of CSI. Therefore, an attack can be initiated, where massive MIMO [62], hybrid massive scheme unified with
an attacker targets at reducing the sum rate of downlink randomization of a signal and the selection of antenna [63].
transmission by incorporating the contamination of uplink Furthermore, other solutions to allay pilot contamination in
pilots [28]. One recent effort towards the solution of a pilot massive MIMO include pilot allocation on the basis of graph
contamination attack is the uncoordinated mechanism of coloring [64], scheme of superimposed pilots [65], game theory
frequency shift [29]. Another resolution to the security of method followed by optimization scheme [66], [67], approach
massive MIMO communication incorporates single-cell of approximate location of mobile devices [68], semi-blind
downlink Time Division Duplexing (TDD) in the presence of scheme for the estimation of the channel in presence of pilot
an eavesdropper. Secure degrees of freedom are obtained even contamination [69], spatial filtering method [70] and
though an eavesdropper is present in the network [30]. For the methodology of channel sparsity [71]. The inclusive survey of
scenario of the wireless network with the consideration of contamination of the pilot signal in massive MIMO is given in
passive eavesdropper only, where secure communication is [72]. The generalized security model for massive MIMO
acquired by employing the massive MIMO relaying technique system can be demonstrated as:
under the non-availability of channel state information of Consider a massive MIMO system entailing of a BS with 𝑖
eavesdropper [31]. In [32]-[34], linear precoding techniques number of antennas and a User Equipment (UE) with 𝑗 number
are utilized for the scenario of a passive eavesdropper for the of antennas where the flat fading channel model from BS to UE
Gaussian MIMO wiretap channel, and these techniques of and from UE to BS can be symbolized as:
linear precoding employed in massive MIMO are mentioned in
𝐻𝑖,𝑗 = 𝑎𝑗 ℎ𝑖,𝑗 𝑏𝑖 (1)
[35]. However, in [36], both active and passive eavesdropper
scenarios were taken into concern. One more approach under or
the consideration of the non-availability of channel state 𝐻𝑗,𝑖 = 𝑎𝑖 ℎ𝑗,𝑖 𝑏𝑗 (2)
information of eavesdropper, security performance is enhanced
where 𝑎𝑗 , 𝑎𝑖 , 𝑏𝑗 and 𝑏𝑖 denotes the response of RF front ends of
by employing the combination of artificial noise and linear
precoding techniques [37], [38]. One more approach to ensure their respective BS and UE during the phase of transmission
PLS in massive MIMO involves the implementation and reception. ℎ𝑖,𝑗 and ℎ𝑗,𝑖 symbolizes propagation channel
beamforming technique under the consideration of channel coefficient from BS to UE and form UE to BS, respectively
estimation errors [39], [40]. Cooperative beamforming [73]. However, from [74] propagation of the channel from BS
provides security by enabling the confidentiality of the to UE can also be defined as:
information [41], [42].
ℎ𝑖,𝑗 = ℎ̅𝑖,𝑗 + ℎ̃𝑖,𝑗 (3)
Moreover, massive MIMO supported C-RAN (Cloud Radio
Access Network), enabling the benefits of secrecy along with Similarly, from UE to BS can also be defined as:
energy efficiency [43]. A mechanism of cooperative relay ℎ𝑗,𝑖 = ℎ̅𝑗,𝑖 + ℎ̃𝑗,𝑖 (4)
network employing in the presence of an eavesdropper whose Equation (4) can also be represented as:
location is determined on the basis of the Poisson point process
incorporating two protocols for relay viz decode-and-forward ℎ𝑖,𝑗 = |ℎ̅𝑖,𝑗 |exp(𝑗2𝜋𝛼𝑖,𝑗 ) + ℎ̃𝑖,𝑗 (5)
and amplify-and-forward, therefore ensuring the security of the ̅
where ℎ𝑖,𝑗 represents the component of the intra array channel
network [44]. Another PLS scheme involves the use of a pilot
model caused by mutual coupling and ℎ̃𝑖,𝑗 models multipath
under the consideration of a multi-antenna jammer. The
detection method for jamming utilizes the likelihood ratio test component of the channel apart from mutual coupling, |ℎ̅𝑖,𝑗 |
for generalized coherence blocks [45, 46]. The mechanism of and 𝛼𝑖,𝑗 designates the magnitude and phase component of ℎ̅𝑖,𝑗
artificial noise is exercised to enhance the security of the correspondingly. The received signal can be obtained as:
network [47]-[50], where an artificial noise is injected in the
𝑦𝑗 = 𝐻𝑗,𝑖 𝑥𝑖 + 𝑛𝑗 (6)
downlink pilot signal in order to prevent the intruder from
getting the correct information about the state of the channel From equation (2), equation (6) can also be re-written as:
[51]. Other different mechanisms that were followed to achieve 𝑦𝑗 = 𝑎𝑖 ℎ𝑗,𝑖 𝑏𝑗 𝑥𝑖 + 𝑛𝑗 (7)
secure communication in massive MIMO include beam Using equation (4), the above equation can be given as:
3
 𝑦𝑗 = 𝑎𝑖 𝑏𝑗 (ℎ̅𝑗,𝑖 + ℎ̃𝑗,𝑖 )𝑥𝑖 + 𝑛𝑗 (8) B. Spectrum Sharing
𝑦𝑗 = 𝑎𝑖 𝑏𝑗 ℎ̅𝑗,𝑖 𝑥𝑖 + 𝑎𝑖 𝑏𝑗 ℎ̃𝑗,𝑖 𝑥𝑖 + 𝑛𝑗 (9) One of the fundamental constraints that are being targeted with
the growing demands of the users is the increased number of
From [75], interference due to other users is taken into spectrum bands. Therefore, the technique of spectrum sharing
consideration for the purpose of analysis. Therefore, equation is introduced. Spectrum sharing technique adapts the primary
(9) can also be deduced as : function to offer efficient service and proficient allocation
𝑗
𝑦𝑗 = 𝑎𝑖 𝑏𝑗 ℎ̅𝑗,𝑖 𝑥𝑖 + 𝑎𝑖 𝑏𝑗 ℎ̃𝑗,𝑖 𝑥𝑖 + 𝑛𝑗 + ∑ 𝐼𝑘 (10) 𝑘=1
among multiple users. Considering the scenario of licensed
spectrum sharing where sharing players are introduced as
where 𝑛𝑗 represents the AWGN with zero mean and 𝜎 2 as the homogenous and heterogeneous sharing players. Based on the
noise power, 𝐼 as the interference due to adjacent users present characteristics of the sharing player, classification of the
in the vicinity. licensed spectrum sharing scheme is introduced. A
Also, the corresponding total power consumption can be homogenous spectrum sharing scheme is the sharing scheme
deduced as: where involved players are of the same nature, while a
𝑂𝑇 = 𝑂𝑡𝑥 + 𝜎 2 + 𝑖𝑝 + 𝑂𝑚𝑐 + 𝑂𝐹 (11) heterogeneous spectrum sharing scheme involves the spectrum
sharing between players of different nature [76]. However,
where 𝑂𝑇 is the total power, 𝑂𝑡𝑥 is the transmitted power by restrictions of the technique affect the security of the network.
the BS, 𝑖𝑝 is the interference power and is given as As LTE was newly projected to operate in an unlicensed band
𝑗
𝑖𝑝 = ∑𝑘=1|𝐼𝑘 |2 , 𝑂𝑚𝑐 power due to mutual coupling of the array, of 5 GHz whose framework coexist with the LSA (Licensed
2 Shared Access) of CBRS for 2.3-2.4 GHz in Europe and 3.5
𝑂𝐹 is the front end power and is given by 𝑂𝐹 = |𝑎𝑖 |2 |𝑏𝑗 | GHz in U.S. Therefore this technique leads to the increased
Now, the SINR between the BS and the authenticated user is interference as interference is one key to the factor affecting the
given by: QoS and thus decreasing the security efficiency of the network
𝑆𝐼𝑁𝑅 =
𝑠𝑖𝑔𝑛𝑎𝑙 𝑝𝑜𝑤𝑒𝑟 [77].
𝑛𝑜𝑖𝑠𝑒 𝑝𝑜𝑤𝑒𝑟+𝑚𝑢𝑡𝑢𝑎𝑙 𝑐𝑜𝑢𝑝𝑙𝑖𝑛𝑔 𝑝𝑜𝑤𝑒𝑟+𝑖𝑛𝑡𝑒𝑟𝑒𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑝𝑜𝑤𝑒𝑟
Congestion is another factor that is required to be examined and
(12) counteracted for preventing security attacks, for example,
2
𝑂𝑡𝑥 |𝐻𝑖,𝑗 | jamming attacks. The taxonomy of threats in spectrum sharing
𝜚𝑗 = (13)
𝜎𝑗2 +𝑖𝑝 +𝑂𝑚𝑐 can be categorized into threats possible in sensing driven
Similarly, SINR for a malicious intruder is modeled as: spectrum sharing and threats possible in database-driven
2 spectrum sharing. Possible menaces to sense driven spectrum
𝑂𝑡𝑥 |𝐻𝑖,𝑖𝑛 |
𝜚𝑖𝑛 = 2 +𝑖 +𝑂 (14) sharing include physical layer threats, MAC layer perils, and
𝜎𝑖𝑛 𝑝 𝑚𝑐
cross-layer perils. Menaces possible in database-driven
The available SR can be expressed as: spectrum sharing include database inference attacks and a
𝑆𝑅 = (𝐶𝑗 − 𝐶𝑖𝑛 )+ (15) menace to database access protocol involving primary and
From the theorem of Shannon’s capacity secondary users [78]. The attacking scenario in the spectrum
sharing can be depicted by considering three possibilities. The
𝐶𝑗 = log 2 (1 + 𝜚𝑗 ) (16) first possibility involves the secondary transmitter acting as an
And 𝐶𝑖𝑛 = log 2 (1 + 𝜚𝑖𝑛 ) (17) illegitimate node in spectrum sharing. As shown in Fig. 2, in
where 𝐶𝑗 defines the capacity of the valid user in the absence Example-1 secondary transmitter, behaves as a node of the
relay, and therefore supports the primary system transmitter
of an intruder and 𝐶𝑖𝑛 defines the capacity of the malicious and primary system receiver with the desired data rate.
intruder. 𝜚𝑗 defines the SINR between the BS and authenticated
Considering the secondary system in which secondary
user. 𝜚𝑖𝑛 defines the SINR between the BS and the malicious transmitter acts as a malevolent node possesses the ability to
2 2
intruder. |𝐻𝑖,𝑗 | , |𝐻𝑖,𝑖𝑛 | models the magnitude of the channel access information of primary system transmitter and primary
gain for the valid user and illegitimate user, respectively. system receiver, therefore, have the capability to create a threat
Using values of equation (13) and (14) in equations (16) and on integrity and confidentiality of the information. In Example-
2 of Fig. 2., considering multiple secondary receivers as
(17) and then substituting those values in equation (15), we get:
2 +
malicious attackers while transmitting secondary data from a
2
𝑂𝑡𝑥 |𝐻𝑖,𝑗 | 𝑂𝑡𝑥 |𝐻𝑖,𝑖𝑛 | Secondary Transmitter (ST) to secondary receiver, there can be
𝑆𝑅 = (log 2 (1 + ) − log 2 (1 + ′ )) > 𝑆𝑅𝑡ℎ
𝜎𝑗2 +𝑖𝑝 +𝑂𝑚𝑐 2
𝜎𝑖𝑛 +𝑖𝑝 +𝑂𝑚𝑐
the possibility of having multiple valid or invalid secondary
(18) receivers. An attack on confidentiality may take place while
where, 𝑆𝑅𝑡ℎ denotes the minimum threshold secrecy rate transmitting information to an invalid secondary receiver.
required to obtain the secured communication. For the case, Example-3 shows the scenario of the Primary Transmitter (PT)
𝑆𝑅 < 𝑆𝑅𝑡ℎ , it can be concluded that the malicious intruder can as an intruder, where the primary system of transmitter and
effectively intrude into the network, 𝑖𝑝 ′ specifies the receiver hold the ability to bug the signal of the secondary
interference power due to the adjacent users in the vicinity of system transmitter [79].
an eavesdropper. Equation (18) specifies SR to characterize the Several solutions to overcome these attacking threats were
security of the massive MIMO system. Moreover, this equation executed to fulfill the requirements of security performance.
can be used to estimate the security performance of the massive PLS is one of the significant technique that makes use of
MIMO system. physical characteristics of the wireless channel to secure
4
wireless communication network against an illegitimate user. For sensing case of 𝐴1 depicts an active state of the primary
A source cooperation mechanism aided with an opportunistic user and is termed as perfect detection while as for sensing case
jamming framework in the presence of an intruder aims to of 𝐴0 , depicting an idle state of the primary user and their
provide confidentiality of transmission for the technology of corresponding probabilities are known as the detection
spectrum sharing [80]. Another technique to improve secrecy probability 𝑃(𝐴1 ) and the false alarm probability 𝑃(𝐴0 ). The
performance in spectrum sharing includes cooperative relay
corresponding probabilities of detection and false alarm are
selection, where three relay selection strategies were followed,
expressed as:
namely PRS, MSRS, ORS. ORS is measured as an
outperformer in comparison to the other two strategies under 𝑡ℎ 𝜇𝑓𝑁
𝑃(𝐴1 ) = 𝑄 [( − Ρ − 1) √ ] (21)
the outlay of having full CSI [81]. The primary aspects of the 𝜎2 2Ρ+1
spectrum sharing system involve spectrum sharing, a decision 𝑡ℎ
for the spectrum, sharing of spectrum and mobility of spectrum. 𝑃(𝐴0 ) = 𝑄 [( − 1) √𝜇𝑓𝑁 ] (22)
𝜎2
However, the two major stages of spectrum sharing are where 𝑡ℎ denotes the threshold for detection, Ρ denotes the
mentioned below: received SNR (Signal to Noise Ratio) at the secondary system
1) Stage of spectrum sensing: transmitter, 𝜎 2 denotes variance of noise n, Q(.) represents the
Spectrum sensing is amenable for determining the state of the
Q-function.
spectrum in the form of idle, busy or available for the
2) Stage of spectrum sharing:
secondary access. It is subjected to decreasing the delay, energy
The second stage involved in spectrum sharing is the stage of
consumption, and interference created to the primary system.
sharing, as in the scenario of sharing secondary users are
Irrespective of its advantages, it is associated with certain
capable of having access to the frequency bands which are
challenging limitations such as hardware requirements in terms
certified to the primary system only. For the case when the
of high-speed processors, high-resolution ADCs (Analog to
primary communicating link between the primary system
Digital Converters) in terms of FPGAs, DSPs. Other
transmitter and receiver is opportunistically weak. The
challenging features include hidden primary user problems,
secondary transmitter possesses the capability to get connected
security issues during sensing. One of the major incursions that
to the network so that the primary receiver can achieve its
can have an effect during sensing is the attack on integrity or
desired data rate. The advantage for the secondary transmitter
intervention where integrity is compromised, is known as PUE
to intervene in the network is that it gets the opportunity for the
(Primary User Emulation), where an eavesdropper
transmission of its particularized data to the secondary system
impersonates the primary system and is capable of deceiving
receiver by utilizing the left out transmit power after
the secondary system. The generalized security modeling of the
connecting with the primary system. Since the data is
spectrum sensing can be expressed as:
transmitted in a scenario of a wireless background. Therefore,
𝐴0 ∶ 𝑥(𝑖) = 𝑛(𝑖) (19) creates more vulnerabilities for insecure communication. To
𝐴1 ∶ 𝑥(𝑖) = ℎ𝑦(𝑖) + 𝑛(𝑗) (20) maintain the security of the spectrum sharing scenario,
Spectrum sharing is used to determine the state of the primary according to Wyner, “the security of the information is
user by using the above mentioned two conditions, where 𝑖= sustained if the authenticated channel is of greater eminence
1,2, …, N, N signifies the total number of samples and is the than the eavesdropping channel”.
function of sensing time 𝜇 and frequency of sampling 𝑓𝑁 . 𝑥(𝑖) Consider the network consisting of a primary system involving
denotes the received signal at secondary user transmitter, 𝑦(𝑖) primary system transmitter, primary system receiver secondary
is the transmitted signal by the primary user, ℎ defines the system transmitter, and secondary system receiver operating in
channel characteristics between the primary system transmitter a full-duplex mode. Considering the attacking scenario in the
and the secondary system receiver, 𝑛(𝑖) depicts the noise. network of spectrum sharing, where the secondary system is
doubtful about the primary system. It involves two phases of
Example-1
Primary Transmitter (PT) transmission included in spectrum sharing. The first
transmission phase encompasses the Request to Cooperate
Primary Receiver (PR) (RTC) sent by the primary transmitter to the primary receiver,
secondary system. The broadcasted signal by the primary
Example-2 Secondary Transmitter (ST) transmitter be 𝑣𝑝𝑡 [79]. The received signal by the primary
Secondary Receiver (SR) system receiver and the secondary system is expressed as:
𝑧 𝑝𝑟 = √𝑜ℎ𝑝𝑟 𝑣𝑝𝑡 + 𝑛 (23)
Malicious ST 𝑧 𝑠𝑡
= √𝑜ℎ𝑠𝑡 𝑣𝑝𝑡 + 𝑛 (24)
Example-3 Malicious SR 𝑠𝑟
𝑧 = √𝑜ℎ𝑠𝑟 𝑣𝑝𝑡 + 𝑛 (25)
𝑝𝑟 𝑠𝑡 𝑠𝑟
Malicious PT where 𝑧 , 𝑧 and 𝑧 models the received signal by the
primary receiver, secondary transmitter, and secondary
receiver correspondingly when the signal is broadcasted by the
Fig. 2. Possible security attack scenario in spectrum sharing primary transmitter, 𝑜 is the total power transmitted, 𝑛 is the
5
AWGN with zero mean presents at the nodes of the primary C. Beamforming
receiver and secondary system. Based on the strategy of It is a propitious technique for next generation WCN. It
allocation of power, the secondary system transmitter involves the maximization of SNR in a definite direction and
combines the signal of the primary system transmitter 𝑣𝑝𝑡 with minimizing it in former directions resulting in the formation of
its own signal 𝑣𝑠𝑡 given by: a beam in a specific direction, which leads to an increase of
𝑣 = √𝑜𝜒𝑣𝑝𝑡 + √𝑜(1 − 𝜒)𝑣𝑠𝑡 (26) energy efficiency as the energy is focused instead of wide-
ranging [82]. Therefore, provides better QoS [83]. It is
where 𝜒 denotes the portion of total transmit power for 𝑣𝑝𝑡
significantly convenient in the field of communication and
signal, 𝑜𝜒 denotes the portion of the transmitted power of a radar. Further, this concept can also be utilized to ameliorate
secondary system for getting the advantage of certified the quality of localization of UE [84]. However, with its
bandwidth from the primary system. 𝑣 denotes the combined numerous advantages, there are certain limitations associated
signal that secondary transmitter broadcasts in the secondary with it. Digital Signal Processing (DSP) chip design is one of
and primary system. Thus, the signal broadcasted by the the main challenges in adapting the beamforming approach.
secondary transmitter, received by the primary and secondary Power requirements in the beamforming technique are
system nodes is given by: relatively more as compared to other promising techniques of
𝑧 𝑝𝑟 ′ = (√𝑜𝜒𝑣𝑝𝑡 + √𝑜(1 − 𝜒)𝑣𝑠𝑡 )ℎ𝑝𝑟 ′ + 𝑛 (27) the future generation. Another aspect that affects the
performance of the network is the parameter of a battery
𝑧 𝑠𝑟 ′ = (√𝑜𝜒𝑣𝑝𝑡 + √𝑜(1 − 𝜒)𝑣𝑠𝑡 )ℎ𝑠𝑟 ′ + 𝑛 (28) lifetime, which is estimated to be comparatively lesser. These
𝑝𝑡 ′ ′
challenging parameters ultimately affect the security of the
𝑧 = (√𝑜𝜒𝑣𝑝𝑡 + √𝑜(1 − 𝜒)𝑣𝑠𝑡 )ℎ𝑝𝑡 + 𝑛 (29)
WCN.
The secondary receiver receives the signals both from the
primary transmitter and secondary transmitter. The received Although directive beamforming is considered as one of the
signal without interference at the secondary system receiver is conceivable tactics to enhance the security performance of the
given as: wireless communication network, due to sharing of
′ information in distributed and collaborative beamforming issue
𝑧 𝑆𝑅 = ( √𝑜(1 − 𝜒)𝑣𝑠𝑡 )ℎ𝑠𝑟 ′ + 𝑛 (30) of security could be a matter of concern. Both the cases of
The SNR received at the secondary receiver for decoding of the beamforming, including distributed and collaborative
signal 𝑣𝑠𝑡 is given by: beamforming. It involves the stage of sharing of information,
(𝑎 −𝜒)𝑜|ℎ𝑠𝑟 ′ |2 from the source node transmission, which will be perceived by
Ρ𝑠𝑟 = = (1 − 𝜒)Ρ𝑔𝑠𝑟 (31) all immediate nodes. The nodes that are not involved in
𝑛
As the primary transmitter is considered to be a malicious cooperating can observe the same information. It , creates a
attacker with an objective of attacking the signal 𝑣𝑠𝑡 of the threat to the integrity of information [85]. The standard of IEEE
secondary transmitter. The secondary transmitter achieves the 802.11ad identifies a selection based protocol for the technique
SR given as: of beamforming. It involves three phases. The first phase is the
𝑆𝑅 = {𝐷𝑠𝑟 − 𝐷𝑝𝑡 }+ (32) phase of SLS involving quasi Omni-pattern to choose the best
transmit and receive antenna sector. The second phase involves
where 𝐷𝑠𝑟 denotes the achieved data rate at the secondary
the phase of beam refinement to select better beamwidth beam
system receiver from the secondary system transmitter, 𝐷𝑝𝑡
pattern pair. This phase is followed by the optional beam
denotes the achieved data rate at the primary transmitter tracking phase, where channel changes are adjusted during data
(malicious attacker) from the secondary transmitter and are transmission [86]. The methods of beamforming encompass
given by: ZF, MRC, and MMSE [87]. Mathematically beamforming can
1
𝐷𝑠𝑟 = log 2 (1 + Ρ𝑠𝑟 ) (33) be represented as:
2
1 Consider a multiple antenna system with j receive antennas
𝐷𝑝𝑡 = log 2 (1 + Ρ𝑝𝑡 ) (34)
2 with 𝑜1 , 𝑜2 , … , 𝑜𝐽 as 𝑗 receive symbols at different receive
The (1/2) factor represents the two phases of transmission. For antennas and single transmitting antenna with 𝑚 as the
two phases of transmission, the signals are required to be transmitted symbol 𝑝1 , 𝑝2 , … , 𝑝𝐽 as the AWGN with zero mean
transmitted in two-time slots. Form the above equations, it is and ℎ1 , ℎ2 , … , ℎ𝐽 represent channel coefficients. The general
observed that security has a direct impact on the data rate
system model can be expressed as:
achieved at the secondary receiver and on the data rate
achieved by the malicious attacker. Form equation (32), the 𝑜̅ = ℎ̅𝑚 + 𝑝̅… (35)
secrecy rate tends to reduce with an upsurge in the rate of data On applying beamforming, represented by the weighted
achieved by the malicious attacker. In other words, the security combination of received symbols as:
of the data is said to be satisfied if the data achieved by the o1
secondary transmitter is higher than the data achieved by the o
𝑠 = [𝑤1 𝑤2 𝑤3 … 𝑤𝑗 ] [ 2⋮ ] (36)
intruder. It indicates the eminence of the wiretap channel, from
o𝑗
secondary system transmitter to primary system transmitter
(intruder) not as much that of the quality of the main channel, It can also be denoted as
i.e., from secondary system transmitter to secondary system 𝑟=𝑤 ̅ ′ 𝑜̅ … (37)
receiver. where 𝑤 ̅ is the vector of weights and is also termed as a vector

6
of beamforming. Therefore, by choosing optimal weights, a Equation (48) is also termed as the maximal ratio combiner as
beam is steered in a specific direction and thus suppressing the it maximizes SNR employing beamforming and the SNR can
beam in other directions. The process of finding the optimal be given by:
value of 𝑤 ̅ is known as the beamforming problem. Beam 𝐸[𝑚2 ]
𝑆𝑁𝑅 = (49)
steering is executed by a process known as electronic steering 𝜃 2 ||𝑤
̅ ||2
offers low complexity, increased energy efficiency, and For any combiner SNR can also be written as:
enhanced precision. For adapting 𝑤 ̅ and obtaining the optimal 𝑆𝑁𝑅 =
𝑎
||ℎ̅||2 … (50)
weight vector from the equation (37) as : 𝜃2
The above equation (50) can also be represented in terms of
𝑤̅ ′ 𝑜̅ = 𝑤 ̅ ′ ℎ̅𝑚+ 𝑤̅ ′ 𝑝̅ … (38)
′̅ ′̅
transmit power is given by:
where 𝑤 ̅ ℎ depicts signal gain and 𝑤 ̅ ℎ𝑝̅ represents the noise
present at the output of beamforming 𝑆𝑁𝑅 = 𝜚||ℎ̅||2 (51)
However, SNR can be maximized by keeping the signal gain where 𝜚 symbolizes the transmit power and is interpreted as the
constant and minimizing the noise power such that: ratio of signal power and noise power. SNR has a definite
impact on the security of the WCN, as the quality of main
1) Case 1:
channel increases with an increase in SNR and decreases the
̅ ′ ℎ̅ =1…
𝑤 (39) interferences present in the channel, which in turn increases the
Or ℎ̅′ 𝑤
̅ = 1… (40) channel capacity as shown in equation (16) and (17), thus
Equation (39) represents the equation of constraint for the affects secrecy rate. From equation (15), the secrecy rate is
optimization problem, where all vector 𝑤 ̅ lie on hyperplane and precisely explained as the difference between the capacity of
indicates a unit gain in signal direction. Formulating SNR the main channel and capacity of the channel of the malicious
maximization forms the objective function for the optimization observer, determining the security of the WCN. From equation
problem. From equation (38), noise power can be specified as: (48) and (50) by obtaining optimal beamformer SNR can be
𝑤̅ ′ 𝑝̅ = ∑𝑖 𝑤𝑖 𝑝𝑖 (41) maximized. It gives rise in capacity of the main channel and
the reduction of the interference resulting in improved
An assumption is made where all noise samples are statistically
performance of the security.
identical with zero mean and identical variance 𝜃 2 signifying
IID (Independent and Identically Distributed) such that D. Device to Device (D2D) communication
𝐸[(∑𝑖 𝑤𝑖 𝑝𝑖 )2 ] = ∑𝑖 𝜃 2 𝑤𝑖 2 (42) The 5G offers device centric architecture of network in
The above can also be represented as: comparison to previous generation network architectures, and
𝐸[(∑𝑖 𝑤𝑖 𝑝𝑖 )2 ] = 𝜃 2 ||𝑤||2 (43) a device is expected to perform dynamically. Therefore,
The optimization problem for beamforming involving defining D2D communication as the shortest communication
minimizing the noise power subject to the constraint 𝑤 ̅ ′ ℎ̅ = between device nodes without the need to pass through the core
1, indicating the constant signal gain network or through an access point [88]. Thus, without the use
2) Case 2: if 𝑤 ̅ ′ ℎ̅ is an affine constraint and the objective is of central authority, including base stations and access points,
convex gain as: the functions of logging, auditing are controlled by resource-
constrained end-user devices. Therefore, D2D communication
𝜃 2 ||𝑤||2 = ||𝑤||2 … (44) executes a crucial role in next generation WCNs. It provides
Equation (44) indicates the convex objective. Both convex the ability to offload heavy data traffic by improving the
objective and convex constraints constitute convex utilization of network resources, provides short-range
optimization problem and to solve a convex optimization communication, thus supports proximity based services. Also,
problem LaGrange’s multiplier can be operated, represented D2D communication depends on the device discovery to
by: perceive peers of communication, which is performed by
𝐿=𝑤 ̅ ′ 𝑤 + 𝜁(1 − 𝑤 ̅ ′ ℎ̅) … (45) transmitting information over wireless channels [89]. In spite
Equation (45) indicates the use of LaGrange’s multiplier of its numerous advantages, several security threats are
denoted by 𝜁and the equation for optimal beamformer can be emerging due to its wireless broadcast communication nature,
given as: which allows malicious attackers to track and locate D2D
𝜁 users, creating a threat to location privacy. The users of D2D
𝑤̅ = ( ) ℎ̅… (46) are spontaneous and self-managed, enforcing and managing
2
Equation (46) indicates the optimal beamformer with security is still a challenge as compared to the centralized
maximizing the SNR ratio. It is also be depicted that 𝑤 ̅ is cellular network. Moreover, next generation WCN architecture
proportional to ℎ̅ which is analogous to the spatially matched and latest application scenarios expose D2D communication to
̅ ′ ℎ̅ = 1 is used.
filter. However, to determine 𝜁, the constraint 𝑤 eavesdropping, jamming, privacy violation, data modification.
Therefore, 𝜁 is calculated as: Mainly D2D communication adopts the connections between
2 devices, which creates a menace to security due to the direct
𝜁= ̅||2
(47) wireless connection, end-user mobility, and privacy issues.
||ℎ
Using the above equation optimal beamformer 𝑤 ∗
̅ is calculated With an increasing number of devices adapting D2D
as: communication, there is an increase in adversaries to target the
ℎ̅ security of the D2D WCN [90].
̅∗ =
𝑤 ̅||2
… (48)
||ℎ

7
 TABLE I
TECHNIQUES TO ELEVATE THE SECURITY OF THE 5G NETWORK
S. no. Technique Objective Description Outcomes/ results
1. Prisoner’s dilemma Analysis of Bandwidth spoofing Approach to evaluate the victory of the intruder Intruder capable of spoofing
game theory [12] attack on various stages of WCN through bandwidth spoofing. bandwidth with a significant
. winning percentage from a
legitimate user.
2. Intrusion detection Detection of an intruder at relay using Intruder detection system ANFIS providing ANFIS provides minimum average
technique using ANFIS minimum average testing error and the malicious testing error which is being chosen
Adaptive Neuro- attackers are identified on the basis of the as a threshold, and therefore the
Fuzzy Inference predetermined threshold value detection of an intruder is done
System (ANFIS) based on the attained value of the
[14] threshold
3. Physical layer Improving the security of the Spatial modeling, mobile association are the PLS Secrecy rate tends to increase
security (PLS) heterogeneous network techniques incorporated to enhance the security profoundly by incorporating
technique [21] beamforming with artificial noise
(AN)
4. Linear precoding Secrecy enhancement using linear Downlink for massive MIMO multi-cell Ergodic secrecy rate is analyzed in
and Artificial Noise precoding of information and by incorporating linear precoding and AN for the presence of contamination of the
(AN) technique [37] utilizing AN enhancement of physical layer security pilot signal by using AN and linear
precoders.
5. Thermal pattern To identify the most probable area of Depending upon the power consumed by user Regions with low security can be
analysis technique illicit access for high-speed handover equipment at any instant and tracing its energy traced out by comparing the real-
[100] setup pattern, detection of the low security region is time energy requirement estimates
observed of devices
6. Secure harvesting Safeguard of extremely penetrating Highly probable regions of attack are being PDF (Probability Density Function)
technique [101] zones, where the possibility of invalid protected by implementing RBA using detection framework is analyzed to
access is maximum using RBA encryption. protect the poorly secured zones
7. Boundary technique To improve the PLS by analyzing the The probability of secure connectivity is The secrecy rate for a high data rate
[112] receiver performance calculated using Wyner coding scheme. The is observed to be increased by
procedure involves the hiding of receiver present hiding the receiver at the boundary
at the corner. or corners.
8. Reinforcement Secure offloading to the edge nodes Learning agent is optimized by using trial and A security solution to provide
learning technique along with a collaborative caching error to obtain the optimal resultant strategy protection against numerous smart
[113] scheme for data privacy attacks in mobile edge caching
systems
9. Multi-antenna To ameliorate the physical layer Use of multiple antennas utilizing a single radio Higher secrecy rates are observed in
transmission security in Millimeter-Wave frequency chain for the transmission of symbols comparison to physical layer
technique [116] Vehicular Communication of information to the destined receiver and noise security techniques using digital or
to non-receiver complex antenna architectures
10. Network function To detect and reduce botnets It involves high-level detection phase by flow Personalized and virtualized
virtualization monitoring function and Deep Packet Inspection honeynet for the proactive detection
technique for Botnet (DPI) for the configuration of the detected bot and mitigation of botnets
Detection and
mitigation [120]
11. Beamforming and To offer PLS by using AN (Artificial An AN is acting as a jamming signal to obstruct the expected SR loss is linearly
jamming technique Noise) aided the beamforming malicious attackers. rising with respect to the power
[124] system dividing factor and to the numeral of
transmit antennas and shows
constant behavior with respect to the
numeral of malicious intruders.
12. Physical Layer To use multiple landmarks with Multiple MIMO landmarks are used to determine Improved accuracy of spoofing
authentication multiple antennas for the the signal RSSI (received signal strength attack detection with reduced
technique [130]. improvement of spoofing detection indicators) with a logistic regression approach overheads
accuracy
13. Robust allocation of To enhance the PLS in the presence Robust covariance matrices of transmitting end Better performance is achieved if
resources technique of an active intruder are obtained for the scenario of Half Duplex the jammer is placed close to the
[145] legitimate receiver friendly jammer (HDJ), Half malicious intruder.
Duplex legitimate receiver (HD), half-duplex
legitimate receiver (Half Duplex legitimate
decode-and-forward relay-HDR), full-duplex
(FD) scenarios
14. Diversity technique , To upsurge the protection mechanism Security performance is evaluated for Security is observed by growing the
[146] of the physical layer cooperative relay present in an environment number of cooperative relays using
pertaining to Rayleigh fading in terms of secrecy the scheme of a better selection of
capacity and the probability of interception relay

8
Table I illustrates the brief summary of various techniques that makes the transmission of the signal received from the first and
have been experienced in improving the security of the 5G its own information to the receiver device. The channel is
communication network. These techniques form the basis and modeled under the Rayleigh fading such that the channel gains
lay the foundation for improving security. on the resource block are given by ℎ1𝑐 , ℎ2𝑐 , ℎ3𝑐 , … , ℎ𝑛𝑐 for the
However, the proper mechanism of security is still required to communication links, ℎ1𝑖 , ℎ2𝑖 , ℎ3𝑖 , … , ℎ𝑛𝑖 as interference links and
be modeled to enhance the capability of secure integration with
ℎ1𝑀 , ℎ2𝑀 , ℎ3𝑀 , … , ℎ𝑛𝑀 as eavesdropper links, such that for the first
future generation architecture and its applications. Through
phase at the BS the received signal, Dr, and eavesdropper M
various research has been studied for full filling the security
respectively is given by:
parameters; however, an effectual mechanism of the security 𝑗 𝑗
solution is still a challenge in the field of research. One of the 𝑅𝑗 = √𝑜𝑐 𝑣𝑐 ℎ𝐶 + √𝑜1 𝑣1 ℎ1 + 𝑛𝑗 (52)
threats to the security in the D2D communication network is where j denotes BS, the relay device 𝐷𝑟 and eavesdropper M,
the proximity based malicious attack where a mobile UE while 𝑜𝑐 and 𝑜1 depicts the power of cellular user in transmit mode
acting as a D2D server can get proximity based mobile viruses and device 𝐷1 , 𝑛𝑗 depicts the AWGN with zero mean and
and are not therefore, able to offer service to other devices. In 𝑗
variance 𝜎 2 , ℎ𝐶 represents the channel coefficient form cellular
addition to it, compromised security of UEs further become a user to corresponding 𝐽 = BS, a relay device,
new source of threat to security while communicating with 𝑗
eavesdropper, ℎ1 represents the channel coefficient form
other devices and thus consequently can spread the attack to
transmitting device 𝐷1 to corresponding BS or relay device or
the entire network. For mitigating such threats, one approach
eavesdropper, 𝑣𝑐 and 𝑣1 represents the transmitted information
involves the utilization of non-cooperative game theory and
from the cellular user and from device 𝐷1 respectively. From
epidemic theory to understand the interdependent security risks
above the instantaneous SINR at the BS, 𝐷𝑟 and at the
and enables security-aware incentive mechanisms [91].
eavesdropper m is given by:
Another approach of enabling a security-enhanced mechanism 𝑗 2
𝑜𝑐 |ℎ𝑐 |
for cellular users in D2D involves the combined optimization Ρ𝑗 = (53)
𝑗 2
of channel assignment and power allocation of communicating 𝑜1 |ℎ1 | +𝜎 2

D2D links where downlink resource sharing is used via multi- For the second phase the received signal at the BS,
channel and single channel communication. On the basis of the eavesdropper M and at the receiver device 𝐷2 is given by:
optimized D2D data rate in view of the power budget, further ′
𝑅𝐵𝑆 = √𝑜𝑐 𝑣𝑐 ℎ𝐶𝐵𝑆 + ℎ2𝐵𝑆 (√𝑙𝑜1 𝑅𝐷𝑟 + √(1 − 𝑙)𝑜𝐷𝑟 𝑣𝐷𝑟 ) + 𝑛𝐵𝑆
improvements in security are made accordingly [92]. (54)
Additional efforts that were made to secure D2D
𝑅2′ = √𝑜𝑐 𝑣𝑐 ℎ𝐶2 + ℎ22 (√𝑙𝑜𝐷𝑟 𝑅𝐷𝑟 + √(1 − 𝑙)𝑜𝐷𝑟 𝑣𝐷𝑟 ) + 𝑛2
communication include pairwise establishment of key to
introduce short authentication based string agreement protocol (55)
′
[93]. The technique of link adaptation to create a balance 𝑅𝑀 = √𝑜𝑐 𝑣𝑐 ℎ𝐶𝑀 + ℎ2𝑀 (√𝑙𝑜𝐷𝑟 𝑅𝐷𝑟 + √(1 − 𝑙)𝑜𝐷𝑟 𝑣𝐷𝑟 ) + 𝑛𝑀
between secrecy spectral efficiency and secrecy energy (56)
efficiency [94]. The PLS technique were adapted to analyze the where 𝑙 depicts the cooperation level, 𝑣𝑐 and 𝑣𝐷𝑟 represents the
rate of generation of secure key on the basis of cooperative transmitted information from cellular users and from the relay
relays (trusted and non-trusted) [95]. The power transfer model device 𝐷𝑟 , 𝑛𝐵𝑆 , 𝑛2 and 𝑛𝑀 depicts the circularly symmetric
is evaluated for the transmission of secured information by AWGN with zero mean and variance 𝜎 2 at the BS, device 𝐷2
making use of three defined policies: best power beacon, and at the eavesdropper. Therefore, the instantaneous SINR
cooperative power beacon, and nearest power beacon [96], obtained at the BS, device 𝐷2 and an eavesdropper M is given
combined scheme of public-key cryptography and mutual by:
2
authentication [97]. The generalized security model of D2D 𝐵𝑆
𝑜𝑐 |ℎ𝐶 |
communication: Ν𝐵𝑆 = 2 2 2
𝐵𝑆 | +𝜎 2 )+(1−𝑙))+𝜎 2
(57)
𝑜𝐷𝑟 |ℎ2𝐵𝑆 | +(𝑙(𝑜1 |ℎ1𝐵𝑆 | +𝑜𝑐 |ℎ𝐶
Consider the scenario of a D2D communication network, 2 2 𝐷 2
𝑜𝐷𝑟 |ℎ𝐶 | (𝑙𝑜1 |ℎ1 𝑟 | +(1−𝑙))
where a device 𝐷1 is far from receiving device denoted by 𝐷2 Ν2 = (58)
2 2
2 | +𝑙𝑜 |ℎ2 | (𝜎 2 +𝑜 |ℎ2 | )+𝜎 2 2
and it is not possible for the device 𝐷1 to directly communicate 𝑜𝑐 |ℎ𝐶 𝐷𝑟 1 𝑐 𝐶
2 𝐷 2
with the receiving device denoted by 𝐷2 . Consider another 𝑀
𝑜𝐷𝑟 |ℎ𝐶 | (𝑙𝑜1 |ℎ1 𝑟 | +(1−𝑙))
device 𝐷𝑟 present in the network in such a manner such that it Ν𝑀 = 2 2
𝑀 | +𝑙𝑜 |ℎ𝑀 | (𝜎 2 +𝑜 |ℎ𝑀 | )+𝜎 2 2 (59)
𝑜𝑐 |ℎ𝐶 𝐷𝑟 1 𝑐 𝐶
is present in connectable premises with both devices and
Therefore the data rate for the scenario where the
therefore, can act as a relay device. Consider M as an
communication takes place through the relay device such that,
eavesdropper under the assumption that the CSI of the uplink
if c number of channels are reused such that 𝑐 < 𝑛, for the
is known and the eavesdropper attempts to obtain the
D2D link, the data rate on channel c is given by:
information that the devices 𝐷1 and 𝐷𝑟 communicates. 𝑜𝑐𝑥 𝑔𝑥
𝑐
Communication follows two phases. In the first phase, BS 𝐷𝑐 ≜ log 2 (1 + ′ ) (60)
1+𝑜𝑐 𝑔𝑐𝑥
makes the decision of allocating the RB (Resource Block) to where 𝐷𝑛 denotes data rate on channel n of D2D link 𝑥 𝑛, 𝑜𝑐𝑥 is
the D2D communication, and in the second phase relay device

9
the power transmitted with D2D link 𝑥 on channel c, 𝑔𝑥𝑐 and increase the security of UDN, as with the growth in the number
𝑔𝑐𝑥 ′ are the normalized gains and are expressed as: of BS. Consequently, gain tends to increase due to which path
′ 2 loss gets decreased. However, this phenomenon is also
|ℎ𝑐𝑥 |
𝑔𝑐𝑥 ′ ≜ 2 (61) associated with frequent handovers and increased inter-cell
𝜎𝑥,𝑐
𝑐 2
interference from neighboring active BSs. The large antenna
𝑐 ′ |ℎ𝑀 |
𝑔𝑀 = 2 (62) array is another tactic to improve physical layer security. BSs
𝜎𝑀,𝑥
employed with a large number of arrays of an antenna can
2
′ |ℎ𝑐𝑀 | significantly provide large gains to the valid users and therefore
𝑔𝑐𝑀 = 2 (63)
𝜎𝑀,𝑥 improves the performance of secrecy. In case of heterogeneous
𝑐
|ℎ𝑥 |2
𝑔𝑥𝑐 ≜ (64) UDN, mm-Wave and sub 6 GHz BSs are capable of providing
2
𝜎𝑐,𝑥
varying levels of array gains depending on the precoding and
Where ℎ𝑐𝑥 is the channel coefficient on channel c from D2D beamforming designs, therefore degrades channels of
link x transmitter to its respective receiver, ℎ𝑐𝑥 ′ is the cross eavesdropping
channel coefficient from the BS to D2D link 𝑥 receiver. PLS has referred to the core idea to improve security
Therefore the secrecy rate can be achieved as: performance for the densified networks. Resource management
′ +
𝑐
𝑜𝑐 𝑔𝑥 𝑐
𝑜𝑐 𝑔𝑀 is one of the approaches to enhance security in a UDN. Another
𝐷𝑐,𝑀 ≜ [log 2 (1 + 𝑥′
) − log 2 (1 + 𝑥′
)] (65) approach towards an increase in security is the blockage of an
1+𝑜𝑐 𝑔𝑐 1+𝑜𝑐 𝑔𝑀

From the above equation achieved, it is clarified that the eavesdropper [105]. Physical layer security can also be
achieved rate of secrecy is dependent on the normalized gains upgraded by considering the UDN, where users are examined
and their respective powers [151]. Where normalized gains are under the close proximity to the cells. The effects of user
in turn, the function of channel gains depicting an improved density, cell density, and eavesdropper are observed on the
gain of the main channel without intruder improves the average SR. Considering the stochastic geometry under the
performance of the security. Rician fading channel, an expression for average SR is
determined for the wiretap channel and the main channel
E. UDN (Ultra Dense Network) without leakage [107]. Additional access to improve the
UDN is investigated as the technology capable of justifying the security of the network is user-centric UDN by utilizing the
growing demands of data rate, low latency, seamless methodology of the network providing service to user and de-
connectivity, ubiquitous coverage. For the fulfillment of cellular. The secure service of the network is provided by the
extensive applications of existing generation and next- dynamic AP (Access Point) grouping enabling mobility
generation WCN, UDN supports a large increase in a number management, interference management, resource management,
of small cells (macro, pico, femto) with increased density of and most importantly security issues [108]. Another user-
small cell access (SCA) points. There are certain challenges centric approach involving clustering is followed for UDN
associated with it, such as small cell discovery, user occupying the secrecy and energy efficiency perspective. A
association, energy efficiency, backhauling, propagation dedicated jamming strategy, along with an embedded jamming
modeling, interference management [98]. However, it exposes strategy, is followed to satisfy the security requirements of
more vulnerable breaches for an attacker to attack and decrease UDN. This design is formulated for both unknown and known
the security of the network. Various possibilities of invalid channel state information. A set of empirical greedy protected
access in UDN and other security challenges associated with it user-centric clustering algorithms is operated for different
are mentioned in [99]. Handover management attacks are scenarios of the UDN [109].
expected to be considered with more frequent possibility, as the In the case of the machine to machine communication in UDN,
figure of handovers is instinctively increased due to the the vulnerabilities considerably exist more due to the cellular
improved number of small cell densities. Various approaches network of the UDN. Also, the high density of BSs creates the
have been investigated for the management of handover attacks issue of frequent handovers and hence, more vulnerable sites
such as RBA, TPA, but an efficient security mechanism is quite for an eavesdropper [110]. However, various categories of
devoid of being contented [100],[101]. attacks and their source causes are stated in [111]. A technique
Several approaches were followed to improve the performance of secrecy improvement is given in [112], where the average
of the security in UDN. PLS approach is used in UDN while capacity of secrecy is evaluated on the basis of Poisson
considering the density of users and the density of interferers. Another technique is contrary to a jamming attack
eavesdropper. Their impact on the average secrecy rate includes the adaptation of reinforcement learning to provide
following the method of stochastic geometry to obtain the secured offloading [113]. For high-speed users in UDN under
secrecy rate [102]. A mechanism of blockchains is employed the scenario of dense picocells, where users are considered to
to provide the authentication of the network. [103] A be randomly distributed, the issue of security for the framework
comprehensive overview of efficient solutions is given in of vehicular users in UDN is explained by the technique of BM
[104], [106] to improve physical layer security. An approach of and BB. This technique improves the secrecy of
BS densification is considered as one of the approaches to communication by improving the reliability of the information

10
exchange between source and destination. The parameter of ∞ 2 𝑃ℎ𝑒𝑣 𝑏 −𝛾
S̅𝑒𝑣 = 2𝜋 Ψ𝑒𝑣 ∫0 𝑏𝑒 −𝜋Ψ𝑒𝑣 𝑏 𝐸 [𝑙𝑛 (1 + )] 𝑑𝑏 (68)
𝜎 2 +𝐹𝑒𝑣 (𝑏)
FSA (Fairness Security Assessment) is analyzed to determine
the potential capability of BM and BB and hence improved the where E[.] denotes the operator of expectation, ℎ𝑒𝑣 denotes the
value of secrecy rate [114]. A scheme of mobile association channel characteristics of the leakage channel of eavesdropper
based on the threshold of the received signal power is modeled and b is the distance between BS and nearest eavesdropper and
in [115] to improve the security of the network. The use of 𝐹𝑒𝑣 (𝑏) is the interference present in the leakage link between
millimeter waves in UDN is considered to be more secure and BS and eavesdropper and is given by:
more suitable for the transfer of power wirelessly, as the path 𝐹𝑒𝑣 (𝑏) = ∑𝑐∈Γ𝑑 /c𝑜 (b) 𝑃ℎ𝑒𝑣 𝑏 −𝛾 (69)
loss incorporated in short distance is lesser as compared to the where Γ𝑑 are the thinned small cells point process Γ𝑑 is the
path loss for long-distance communication[116]. The effective subset of PPP of BS.
secrecy throughput in the presence of multiple malicious ∞ 2 𝑃ℎ𝑠 𝑎 −𝛾
observers is analyzed, and protected communication for S̅𝑠 = 2𝜋 Ψ𝑠 ∫0 𝑎𝑒 −𝜋Ψ𝑠 𝑎 𝐸 [𝑙𝑛 (1 + )] 𝑑𝑎 (70)
𝜎 2 +𝐹𝑠 (𝑎)
SWIPT is observed [117], [118]. Also, in [119], Matern Hard where ℎ𝑠 is the gain of the main channel, a is the distance
Core Point Processes is used to characterize the haphazard between the nearest BS and the specific user, 𝐹𝑠 (𝑎) is the
position of the BS and users and thus determine the secrecy aggregate interference present in the main link due to other
outage, respectively. The generalized security model of UDN active base stations and is given by:
communication is expressed as: −𝛾
𝐹𝑠 (𝑎) = ∑ Γ𝑑 𝑃ℎ𝑠 𝑎𝑐 (71)
Consider the scenario of UDN consisting of small cell base 𝑐∈
c𝑜 (a)
station transmitting a power ‘Pbs’ where the position of the base From the above equations, it can be concluded that nearest
station is spatially distributed according to homogenous intruders decrease the secrecy rate more effectively as
Poisons Point Process (PPP) Γ𝑏𝑠 with a density of Ψ𝑏𝑠 and compared to intruders that are far from the respective Base
another homogenous PPP of mobile stations Γ𝑚𝑠 with a density Station, thereby creates a primary impact on the security of the
of Ψ𝑚𝑠 held by valid users independent from BSs PP. Consider communicating network.
a malicious attacker distributed according to homogenous PPP
F. IoT (Internet of Things)
Γ𝑒𝑣 having a density of Ψ𝑚𝑎 and is independent of PPPs of BS
and valid users, present along with the valid users, and attempts IoT describes the future generation epoch of wireless
to intercept the communicating data of valid users. communication providing immense convenience and value. It
is capable of connecting the number of entities, including
Let the density of valid users is less than the density of
sensors, mobile phones, automation, smart vehicles, smart
malicious attackers such that Ψ𝑒𝑣 < Ψ𝑚𝑠 . Also, channel
home appliances, on-body sensors in the scientific and
characteristics of small scale fading are considered adapting the
industrial community. In general, both the survival of humans,
fundamental property of UDN, where LOS (Line of Sight) is
as well as industries, are drastically affected by the massive
the most possible propagation. The main channel is
deployment of application critical IoT devices. It has become
characterized as the channel between the valid users and base
the collective term for upcoming networks [121]. From
stations denoted by h𝑚 while as the eavesdroppers channel or
preceding years, the foundation of undefended project masses
the leakage channel is characterized as the channel between the
created vulnerable sites causing numerous “proof of concept”
malicious attacker and transmitting BS denoted by h𝑒𝑣 . To
attacks to exploit those exposures [122]. Therefore, with the
evaluate the secure communication in UDN, the parameter of
rapid deployment of such a promising era, it carries certain
secrecy rate is examined denoted by S𝑟 is expressed as:
challenges with it in terms of security management, energy
S𝑟 = [S𝑚 − S𝑒𝑣 ]+ (66) efficiency, autonomy, scalability, and other performance
where, S𝑚 is the secrecy rate of the main channel and S𝑒𝑣 is tradeoffs. The cause of these challenges is easy to follow, as
the secrecy rate of the channel of an eavesdropper, where the information exchange is now much more life-critical and
[𝑎]+ = max[a,0] and the average secrecy rate can be estimated sensitive.
as: Personal health monitoring, biometric data, information of
S̅𝑟 ≈ S̅𝑚 − S̅𝑒𝑣 (67) location are present all over the internet, and the possibility of
̅ ̅
where S𝑟 is the average secrecy rate, S𝑚 is the average main breaches could create the susceptibility of organized e-crimes
channel secrecy rate and S̅𝑒𝑣 is the average secrecy rate of the resulting in enormous financial the damages, the depiction of
channel of the eavesdropper reputation of the company to stakeholders and customers, and
For the Eavesdropper channel or the link with leakage, where more importantly, creates a threat to the lives of human [123].
the nearest malicious attacker is considered as the most Therefore, among these challenges, security occupies the
damaging malicious attacker [107] and the average channel critical priority for the secure communication irrespective of its
secrecy rate with leakage in the presence of the nearest distributed and broadcast nature of communication. As the
eavesdropper where other eavesdroppers are independent of internet of things is centered by the number of seamless
connections and generally IoT architectures encompass four
each other is given by:
layers viz network layer, perceptual layer, application layer,

11
and the support layer. The perceptual layer is the primary layer principles for the implementation of PETs at various layers of
responsible for collecting information of all types from the IoTs were demonstrated. Privacy legislation maps validated
physical world by making use of devices, including radio with respect to the principles of privacy are employed for the
frequency identification tags or accelerometers. design of PETs in the architecture of IoT stack.
On the basis of simplicity and resource-oriented nature of the A deep neural network that permits authentication of wireless
IoT entities present at the perceptual layer, the security of IoT nodes in real-time on the basis of variation in properties of RF
communicating signals at this layer is a challenging task. of the transmitter detected by in-situ ML present at the
Various security solutions were proposed for the authentication destination end [129]. This scheme makes use of 65-nm node
of IoT signals such as traditional encryption techniques, PLS standard, including factors such as local oscillator offset, I-Q
techniques to secure communication. A lightweight biometric imbalance where detection using neural network consists of the
system specifically considered for limited resources of IoT hidden layer having 50 neurons and is capable of providing
entities is proposed in [121] for the authentication of the user. accuracy of 99.9% with an ability to distinguish 4800
On the basis of the MCC code, which is basically a state-of- transmitters. It is responsible for providing authentication
the-art matching fingerprint algorithm, it raises the security of under varying channel conditions. A mechanism of spoofing
the network by maintaining the accuracy of recognition. The attack detection by operating the system of PL authentication
biometric system makes use of block logic operation, which by exploiting the CSI with more accuracy using multiple
results in a decrease of biometric feature size, therefore reduces landmark schemes [130]. The risks that lie in the privacy and
the number of computations and memory. security of IoT are delved in [131].
In [123], a PUF based cryptosystem design was proposed for In [132], the IoT attack model was identified based on the
the lightweight security of IoTs. This cryptosystem is based on learning approach to improve security in terms of secure
the principle that two chips that are identical from the same line offloading, access control, authentication, malware detection.
of production cannot share the alike and exact physical For malware detection, an approach of Q-learning is followed
characteristics. These characteristics may include frequency of to obtain optimal offloading rate with unknown bandwidth
oscillating components, speed of racing signals, the model and trace generation of the neighboring devices of IoT.
randomness of memory components in their initial states, For secure offloading, the technique of reinforcement learning
speckle outline and arrangement of optically delicate is approached in the dynamic radio environment. For access
substantial. On the basis of such characteristics, devices for control, the machine learning approach, including techniques
cryptographic functions can be mechanized in the form of such as K-NNs (K-Nearest Neighbor), SVMs, and NNs, are
PUFs. A mechanism of PLS enhancement in IoTs is proposed capable of detection of an intruder in IoTs. Moreover, another
in [126], where the performance of security is evaluated for a field where the security of IoTs occupies fundamental
multiuser system that makes use of antenna selection scheme importance in the field of healthcare. In the case of the smart
at the transmitter end TAS and incorporates diversity healthcare system, various security issues exist in the form of
opportunistic scheduling TSD on the basis of threshold over password guessing attacks, which can lead to DoS attacks,
legitimate nodes. The valid communication between the active and passive attacks, respectively, creating a menace to
legitimate source and authentic receiver experiences the the privacy of health data. Keeping in view the concept of a
existence of colluding and non-colluding passive malicious learning approach, IoT security challenges are required to be
attacker(s). Closed system expression for PDFs and CDFs addressed proactively where the devices can adapt dynamically
(Cumulative Density Functions) of SINR (signal-to- security threats and possess reconfigurability [133].
interference-plus-noise ratio) and other evaluating parameters Various approaches were followed to increase the strength of
of security such as Secrecy Outage Probability (SOP) of the the password and therefore reduce the security vulnerabilities.
system, asymptotic outage secrecy probability for the scenario PSMs approach is proposed in [134] to progress the security of
of colluding and non-colluding eavesdroppers were observed. healthcare system in IoT on the basis of personal identification
Another approach to secure communication in the internet of information, and thereafter label processing to detect the
things involves a watermarking algorithm [127]. In such an personal information present in the password and other hidden
approach, dynamic authentication is performed for the variations. The GSC and RSC are the two antenna reception
detection of cyber-attacks. This approach is based on a deep techniques that are used to improve the security on the basis of
learning mechanism to provide short term memory structure in the zero-forcing beamforming algorithm [135]. One more
order to abstract a determined set of probabilistic components approach to enhance the security of the IoTs is to secure the
of the produced signal and therefore enabling the entities of IoT downlink of the transmission by using cooperative jammer
to dynamically watermark the features of the set into the signal. against non-colluding and passive eavesdroppers, where
The authentication process is effectively performed by analysis of security parameters such as SR and SOP are
enabling the IoT gateway to collect signals from the devices of analyzed [138]. The emerging challenges in the security of
IoT to verify the reliability of the signal. In [128], PETs were IoTs whose countermeasures are yet to be solved are an
studied where state-of-the-art principles for the laws of privacy, exponential rise in the numeral of poor links and unpredicted
considered for the architecture, were analyzed. The valid use of information [139]. The significant areas that are present
12
in the current IoT landscape include security in connected channel coefficient present between the 𝑖𝑒 -th interferer and the
vehicles (VANETs-Vehicular Ad-hoc Networks), security in intruder, 𝑑𝑖𝑒 shows symbols transmitted from the co-channel
MEC (Mobile Edge Computing), blockchains, artificial interferer with unit variance and zero mean, 𝐼𝑒 is the total of all
intelligence as a solution to enhance the security in the IoTs interferers present at the intruder and 𝑛𝑒 is the AWGN with unit
[140], [142]. variance and zero mean at the intruder. The received SINR
A polar low complexity coding method is utilized to provide (𝜚𝑚,𝑒 , 𝜚𝑚,𝑝 ) at the eavesdropper and valid node is given by:
confidentiality to the information under improved secrecy 2
𝛼𝑚 |ℎ𝑚,𝑑 |
[141]. Blockchain, as a possible solution to the security of IoT, 𝜚𝑚,𝑑 = 𝐼𝑑 2 (74)
∑𝑖 =1 𝛼𝑖 |ℎ𝑖 ,𝑑 | +1
𝑑 𝑑 𝑑
acts as a decentralized database on the basis of cryptographic
where
techniques to improve the security of the IoTs and, therefore, 𝐶𝑚 𝑐𝑖
provides reliability, desirable scalability, and authentication 𝑑 𝜖{𝑝, 𝑒}, 𝛼𝑚 = , 𝛼𝑖𝑑 = 𝑑
and 𝑐𝑖𝑑 = 𝜂 𝑐𝑚 , 0≤ 𝜂 ≤ 1
𝑁𝑜 𝑁𝑜
[143]. However, conventional mechanisms to improve the The secrecy capacity denoted by 𝐶𝑠 can be given as:
security of IoT requires an optimization framework for the +
𝐶𝑠 = [𝐶𝑝 − 𝐶𝐸 ] = max(𝐶𝑝 − 𝐶𝐸 , 0) (75)
involved nodes in the network to provide valid access to nodes
[149]. Also, a countermeasure to the Sybil attack involves K- where 𝐶𝑝 = log(1 + 𝜚𝑝 ) and 𝐶𝑒 = log(1 + 𝜚𝑒 ) , assume CSI
means clustering, followed by the detection scheme via age of an intruder is not accessible. The BS considers for secure
replacement policy is identified [150]. Yet, It involves the communication, the rate of the channel which under the
examination of only one intruder node. Considering all these influence of eavesdropper is given as (𝐶𝐸 )′ = 𝐶𝑝 − 𝐶𝑟 .where
above mentioned security enhancement schemes, a novel and 𝐶𝑟 is the set SR by the BS. The BS attempts to construct the
sophisticated procedure of filtering, firewalls, potent IDS are wiretap codes by using (𝐶𝐸 )′and 𝐶𝑝 . To ensure perfect secrecy
required to enable secure health care, business practices, and rate where secrecy is not compromised 𝐶𝑟 ≤ 𝐶𝑠 . An outage of
other essential applications. The mathematical security model secrecy is said to occur if 𝐶𝑠 < 𝐶𝑟 . This performance of analysis
for IoTs can be evaluated as: is said to be SOP analysis [126]. The likelihood of SOP under
Consider a system with n number of users where n≥1 such that the fading distribution is given by:
a scenario of the multiuser downlink is obtained. Assume co- 𝑃𝑜 (𝐶𝑟 ) = Pr[𝐶𝑠 < 𝐶𝑟 ] (76)
channel interference is present, and each user (valid and or
eavesdropper) suffers identical co-channel interference. 𝑃𝑜 (𝐶𝑟 ) = Pr(𝐶𝑠 < 𝐶𝑟 |𝜚𝑝 > 𝜚𝑒 ) Pr(𝜚𝑝 > 𝜚𝑒 ) + Pr(𝜚𝑝 < 𝜚𝑒 ) (77)
Consider A number of antennas at the transmitter end and B or
∞
number of users, which are involved in communication from a 𝑃𝑜 (𝐶𝑟 ) = ∫0 𝜚𝑝′ (2𝐶𝑟 (1 + 𝑅) − 1)𝑓𝜚𝑒 (𝑅)𝑑𝑅 (78)
set of n number of users. Consider the presence of an the relationship is said to exist between the CDF and its
eavesdropper either in an active or passive attacking mode, predefined threshold value (𝜚𝑡ℎ ) as 2𝐶𝑟 (1 + 𝑅) − 1 ≥ 𝜚𝑡ℎ or
possessing the capability to intercept the communicating 2𝐶𝑟 (1 + 𝑅) − 1 < 𝜚𝑡ℎ . The piecewise secrecy outage
information. In addition to it, assume the communication link probability with respect to any bound point 𝑇(𝜚𝑡ℎ ) =
is I.I.D. The expected signal at the 𝑚𝑡ℎ user at the 𝑝𝑡ℎ antenna 2−𝐶𝑟 (1 + 𝑅) − 1 is given by:
BS is derived as: 𝑇(𝜚𝑡ℎ )
𝐼 ∫0 𝜚𝑝′ (𝜛𝑅 )𝑓𝜚𝑝 (𝑅)𝑑𝑅
𝑅𝑚,𝑝 = √𝑐𝑚 ℎ𝑚,𝑝 𝑑 + ∑𝑖𝑝 =1 √𝑐𝑖𝑝 ℎ𝑖𝑝 ,𝑝 𝑑𝑖𝑝 + 𝑛𝑚 (72) ∞
𝑝 𝑃𝑜 (𝐶𝑟 ) = ∫𝑇(𝜚 𝜚𝑝′ (𝜛𝑅 )𝑓𝜚𝑝 (𝑅)𝑑𝑅 𝑇(𝜚𝑡ℎ ) ≥ 0
𝑡ℎ )
where 𝑐𝑚 is the power transmitted to the valid user, ℎ𝑚,𝑝 ∞ ′
𝜚 (𝜛 )𝑓 (𝑅)𝑑𝑅 𝑇(𝜚𝑡ℎ ) < 0
characterizes the channel coefficient between the 𝑚𝑡ℎ antenna {∫0 𝑝 𝑅 𝜚𝑝
(79)
and 𝑝𝑡ℎ valid-user where m ranges from 1 to A and p ranges 𝐶𝑟 (1
from 1 to B, d is the symbol transmitted to the valid user having where 𝜛𝑅 = 2 + 𝑅) − 1. Therefore on the basis of the
above-attained closed-form equations, SOP can be calculated,
zero mean and unit variance, 𝑐𝑖𝑝 is the interference power at the
and the factors affecting the SOP include the set secrecy
𝑝𝑡ℎ valid user, ℎ𝑖𝑝 ,𝑝 is the channel coefficient between threshold, average SNR of the valid and eavesdropper node,
𝑖𝑝 interferer and 𝑝𝑡ℎ valid node, where 𝑖𝑝 ranges from 1 to 𝐼𝑝 , and the number of eavesdropping nodes. From the above
𝑑𝑖𝑝 are the symbol transmitted from co-channel interference mentioned closed-form equations, it can be deduced that
interference and the number of interfering devices lead to a
sources with unit variance and zero mean, 𝐼𝑝 is the number of decrease in the performance of the security.
co-channel interferers, 𝑛𝑚 is the AWGN with unit variance and Fig. 3. condenses the security attacks in different technologies
zero mean. The obtained signal at each malicious attacker from of 5G along with the perspective of their countermeasures. For
the 𝑚𝑡ℎ antenna is expressed as: D2D communication, the security attacks and preventive
𝐼
𝑅𝑚,𝑒 = √𝑐𝑚 ℎ𝑚,𝑒 𝑑 + ∑𝑖𝑒𝑒 =1 √𝑐𝑖𝑒 ℎ𝑖𝑒 ,𝑒 𝑑𝑖𝑒 + 𝑛𝑒 (73) aspects of security in inband and outband configurations are
where ℎ𝑚,𝑒 signifies channel coefficient between the 𝑚𝑡ℎ briefly mentioned. Further, attacks and the related blocking and
precautionary security aspects in the technologies of spectrum
antenna and 𝑒 𝑡ℎ malicious observer, 𝑐𝑖𝑒 is the intervention
sharing, UDN, massive MIMO and IoT are summarized
power from the 𝑖𝑒 -th interferer at the intruder, ℎ𝑖𝑒 ,𝑒 denotes

13
 Technology Attack Type Possible Solution
Underlay Malware attack Access control
Inference attack confidentiality
Inband eavesdropping Integrity
Overlay Session hijacking obfuscation
D2D Man-in-the-middle Anomymity
Impersonation Cryptography
Controlled
Masquerading Key management
Outband DDos Authentication
IP spoofing Authorization
Autonomous Bandwidth spoofing Availability and dependability
Location poofing Secure routing transmission
Jamming attack IDS

Signal superposition
RF fingerprinting
EM signature identification
Watermarking
Physical layer Spectrum sensing data falsification
IDS
Primary user emulation Manipulation in SDR and CR
MAC layer or Rogue transmitter (DoS) Policy enabled CR
Spectrum data-link layer Beacon falsification Rule based policy reasoner
Sharing Tamper resistance
Loin attack
Primary system Fine imposed on adversary
Confidentiality attacks
Spectrum inaccessible to
DIA (Database Inference Attack) rogue transmitter
Secondary system Location privacy attack Checking RSS for false
transmitter
5G
Encryption schemes (WEP,WPA2)
Picocells (D)DoS, botnet attack, jamming attack
MAC address filtering, SSID removal
UDN User privacy, Traffic monitoring
Pseudonyms and signatures
Authentication based attack (MITM, Impersonation)
Detection of apps using code sequence
Femtocells Service theft and fraud, Rogue access point
SAFH scheme
False location attack, location tracing
PDR, carrier sensing time analysis
Relay based attack,Bandwidth spoofing Random routing approach
Relays
IDS

TAS-ML scheme
Pilot contamination attack
Beamforming
Information leakage
Game theoretic study
Eavesdropping
Privacy Het C-RAN
Spoofing attack
Authentication AF-Relaying
Massive Jamming attack, (D)DoS
MIMO Confidentiality Beamforming & AN
Impersonation
Integrity Linear precoding & AN
Active pilot spoofing attack
Reliability PL-Cryptography
Location privacy
MRC-MIMO Relay system
Computational attack, PT
attack, Ciphertext attack Sub-space estimation approach,
Phishing attack pilot based estimation approach
Packet injection and modification attack
Malware attack

Eavesdropping Isolation

Tracking Kill/sleep command

RFID tags
Tag cloning Anonymous tag

inventorying Blocking
Counterfeiting Personal firewall
Physical attacks Information flooding
Corrupted node Distance estimation
Edge Nodes
Computing nodes Node Replication attack Secure firmware update
IOT Hardware trojans Trojan activation methods
Camouflage Side-channel signal analysis
Denial of service IDS
Side channel attack Design/circuit modification
Edge Computing Integrity attacks Outlier Detection
Inessential/insufficient logging De-patterning & decentralization
Malicious injection Pretesting
Communication links Causative attack
Cryptographic schemes
impersonation
Role-based authorization
Unauthorized links
Reliable routing
Routing attack
Fig. 3. Visualization of various attacking categories in 5G

14
 b) Dynamic seeding of a cloud
III. THE PROPOSED SCENARIO OF ARTIFICIAL DUST
AND ARTIFICIAL RAIN WITH HALF DUPLEX It includes the precipitation using erect air currents to rise more
ATTACKING MODEL water droplets to pass through clouds, therefore converting into
more rainfall. This method of seeding is considered to be more
The proposed scheme is divided into two main phases. The first complex as it involves a sequence of proceedings. It includes
phase further comprises of two cases. The first case 100 times more ice crystals than the static method of cloud
demonstrates the scenario of AR, and the second case seeding.
represents the scenario of AD. The second phase involves the
modeling D2D half-duplex attack. c) Hygroscopic seeding of the cloud
A. Phase First This technique defines the seeding process by scattering salts
This phase reveals the impact of attenuation due to AR and AD in the lower part of clouds. The size of salt molecules increases
on WCN and thereby creating a vulnerable environment for the by adjacent combing of water molecules. Due to the increased
eavesdropper by introducing the attenuation, consequently size and weight, these molecules are expected to precipitate as
degrading the communicating channel. rainfall. An intruder is examining the communication taking
place between the various nodes of defense wireless
1) Case 1: Scenario of AR (Artificial Rain) communication networks and has complete information about
The utmost phenomenon of weather is the rainfall. In the the traffic of the communication network. therefore, on the
present era of WCN, the attenuation of propagating signals due basis of traffic pattern analysis, at any time when intruder finds
to rainfall is not equitably approximated. The estimates are the essential commands or information that is required to be
reasonably considered for the WCN operating below 6GHz. transmitted to the destined node of the defense communication
However, the diameter of the raindrop lies in the range from network, intruder starts the instant procedures of artificial dust,
0.1mm to 10mm, which is less than the wavelength of the followed by the procedure of artificial rain. Due to the presence
operating signal in WCN. Therefore, creates attenuation in the of the artificial dust particles or rain particles present in the
signal, decreases its strength, and increases path loss, atmosphere, the interference tends to increase, thereby
consequently decreases the capacity of the system, which leads increases the path loss and thus degrades the communication
to the ultimate decrease in secrecy rate. Thus creates a environment.
favorable scenario for the malicious intruder to attack the
communicating network. Artificial rain is the phenomenon of 2) Case 2: Scenario of AD (Artificial Dust)
causing deliberate rainfall by using the process of precipitation AD shows an important aspect that can affect the
in the form of rain or drizzle or sleet or snow or graupel or hail. communicating link. The impact of AD is illustrated by
This process follows the method of cloud seeding defined as an considering the cross-polarization and attenuation on the
artificial approach of injecting condensation nuclei typically millimeter and microwave WCNs. The wave attenuation due to
0.2 µm or 1/100th the dimensions of a cloud droplet on which dust to AD can be approximated by Rayleigh scattering theory
water vapor condenses. This mechanism makes use of cloud or Mie scattering approximation. AD facilitates an inventive
seeding chemicals such as potassium iodide, silver iodide, method of communicating with the environment. CARE
liquid propane ammonium nitrate, solid carbon dioxide (dry (Charged Aerosol Release Experiment) represents the AD
ice), chloride calcium carbonate, the compound of urea calcium project handled by NASA to observe the specific concern of
oxide. These chemicals are considered as water-absorbent artificially generated dust cloud named Noctilucent Clouds.
substances to enhance the formation of the cloud, which Nevertheless, the potential aspect of AD decreases the secrecy
increases or induces rainfall. The injection of these chemicals of the network. Therefore, the use of AD by an intruder requires
is being dispersed in the clouds through aircraft, dispersion primary attention. The AD is introduced by an
devices positioned on the ground or through remote-controlled
rockets, or through generators. There are three methods
through which artificial rain can be achieved. Water droplets, dust
Destination Node
Source Node

a) Static cloud seeding: Signal

This method involves the dispersing of chemical substances

such as silver iodide (AgI) into the cloud. This chemical Water droplets, dust, Artificial dust
substance exhibits crystalline structure, and due to the
difference in vapor pressure, the tiny water droplets or the
moisture accumulate on the molecules of the chemical
Artificial Rain (AR)
substance, thereby increasing the size of the molecules. Due to Fig. 4. The impact of artificial dust and artificial rain on the strength of the
sufficient size, the particles enough weighty to fall from clouds signal

and produce the rain.

15
intruder in the communicating environment of WCN aerially In the dBm scale
provides the capability to the intruder to examine the 𝑃𝑟 (𝑑𝑏𝑚) = 𝑃𝑡 (𝑑𝑏𝑚) + 10 log10 (𝐴𝑙 ) + 20 log10 (𝜆) −
communication, sense the information, and thus increases the 20 log10 (4𝜋) − 20 log10 (𝑟) (85)
possibility of eavesdropping in the network. As a result, creates Therefore, the free space loss can be expressed as:
an ability to the intruder to monitor the rate of communication 𝑃𝑡
𝑃𝑙 (𝑑𝑏) = 10 log10 (86)
and thus possesses the capability to create hindrances or delay 𝑃𝑟
the communication. The impact of AD and AR on the signal Using equation (84), we get
strength is shown in Fig. 4. From the source node to the 𝐴𝑙 𝜆2
𝑃𝑙 (𝑑𝑏) = −10 log10 (87)
destination node in the absence of AR and AD, the signal (4𝜋𝑟)2
strength of the signal occupies minimum losses due to the Also, the free-space path gain is given by:
absence or minimum of obstacles present in the communication 𝑃𝑔 (𝑑𝑏) = −𝑃𝑙 (𝑑𝑏) (88)
channel. Due to the addition of AD, the signal is transmitted 𝐴𝑙 𝜆2
through the communication channel by passing through the 𝑃𝑔 (𝑑𝑏) = 10 log 10 (4𝜋𝑟)2 (89)
obstacles creating the decrease of the signal strength, as shown Approximated PL as a function of distance r is expressed as:
in Fig. 4. However, for AR, the particle size is considerably 𝜓
𝑟
more than the dust particles, thereby creates a more impact on 𝑃𝑟 = 𝑃𝑡 𝑞 [ 𝑜 ] (90)
𝑟
the signal strength.
In dB, the above equation (90) can be rewritten as:
From Fig 5 at stage-I, wireless communication environment is 𝑟
devoid of artificial dust or artificial rain, seamless 𝑃𝑟 (𝑑𝑏𝑚) = 𝑃𝑡 (𝑑𝑏𝑚) + 𝑞(𝑑𝑏) − 10 ψlog10 ( 𝑜 ) (91)
𝑟
communication is expected to occur. At stage-II, when an where q is a dimensionless constant depending on the
intruder introduces the AD particles in the atmosphere, more characteristics of average attenuation of channel and antenna,
interference, more path loss is expected to occur. At stage-III, 𝑟𝑜 is the distance of reference, 𝜓 is the PL exponent
the presence of AR particles creates more interference and where q can be expressed as:
attenuation due to the large size molecules of the rain droplets, 𝜆
𝑞(𝑑𝑏) = 20 log10 ( ) (92)
thereby creating multipath propagation attenuation and thus 𝑟𝑜
decreasing the signal strength. The mathematical modeling to Using a model of log-normal shadowing, suppose the ratio of
determine the impact of AR and AD on WCN can be deduced transmit power to receive power is represented by‘𝜙’ be
as: random, given by log-normal distribution as:
Δ (10 log 10 𝜙 − 𝜇𝜙 (𝑑𝑏) )2
a) Stage I: Free space PL (Path Loss) 𝑝(𝜙) = 𝑒𝑥𝑝 {− }, 𝜙>0
√2𝜋 𝜎𝜙 (𝑑𝑏) 𝜙 2 𝜎 2 𝜙 (𝑑𝑏)
Consider a signal transmitted to a receiver through free space (93)
at a distance of ‘r’. Suppose there are no hindrances in
where Δ is given as :
reference to the receiver and transmitter, and thus the signal is
10
propagating in a straightforward path. The path loss of free Δ= (94)
ln 10
space presents a complex factor of scaling in the receiving where 𝜇𝜙 (𝑑𝑏) denotes the mean of 𝜙(𝑑𝑏) , 𝜎 𝜙 (𝑑𝑏) represents 2
signal such that the received signal can be calculated as:
the variance of 𝜙(𝑑𝑏) and 𝜎𝜙 (𝑑𝑏) represents the standard
𝑗2𝜋𝑟
−
𝜆√𝐴𝑙 𝑒 𝜆 deviation of 𝜙(𝑑𝑏) . The expectation of 𝜙 or the linear average
𝑟(𝑡) = 𝑅𝑒 { 𝑥 ′ (𝑡) 𝑒 −2𝜋𝑓𝑐𝑡 }, (80)
4𝜋𝑟 path gain or the mean of 𝜙 can be given as:
where 𝐴𝑙 is the product of field radiation patterns of transmit 𝜇𝜙 (𝑑𝑏) 𝜎 2 𝜙 (𝑑𝑏)
𝐸[𝜙] = 𝜇𝜙 = 𝑒𝑥𝑝 [ + ] (95)
𝑗2𝜋𝑟 Δ 2Δ2
antenna and receive antenna, 𝑒 − 𝜆 denotes the phase shift that
In the dB scale:
arises due to the traveling distance of wave from transmit
𝜎 2 𝜙 (𝑑𝑏)
antenna to receive antenna ‘r’, 𝑥 ′ (𝑡) denotes the complex 10 log10 𝜇𝜙 = 𝜇𝜙 (𝑑𝑏) +
2Δ2
(96)
baseband signal and is the combination of an in-phase Equation (93) can also be written as:
component and quadrature component is given by: 2
1 (𝜙(𝑑𝑏) − 𝜇𝜙 (𝑑𝑏) )
𝑥 ′ (𝑡) = 𝑢𝐼 (𝑡) + 𝑗𝑢𝑄 (𝑡) (81) 𝑝(𝜙(𝑑𝑏) ) = exp {−
2 𝜎2
} (97)
√2𝜋 𝜎𝜙 (𝑑𝑏) 𝜙 (𝑑𝑏)
where 𝑢𝐼 (𝑡) denotes in-phase component and 𝑢𝑄 (𝑡) denotes
Therefore the combined path loss and shadowing can be
quadrature component and are represented as: evaluated as:
𝑢𝐼 (𝑡) = 𝑅𝑒 {𝑥 ′ (𝑡)} (82) 𝑃 𝑟
′ ( 𝑟) = 10 log10 𝑞 + 10𝜓 log10 − 𝜙(𝑑𝑏) (98)
𝑢𝑄 (𝑡) = 𝐼𝑚 {𝑥 (𝑡)} (83) 𝑃𝑡 (𝑑𝑏) 𝑟𝑜

The ratio of the received power to transmitter power is given In dbm scale,
as: 𝑟
𝑃𝑟 (𝑑𝑏𝑚) = 𝑃𝑡 (𝑑𝑏𝑚) + 𝑞(𝑑𝑏) − 10 ψlog10 ( 𝑜) − 𝜙(𝑑𝑏𝑚) (99)
2 𝑟
𝑃𝑟 √𝐴𝑙 𝜆
= [ ] (84)
𝑃𝑡 4𝜋𝑟

16
 Stage - I Stage - II

Intruder Artificial Rain

Relay
Defense
Spectrum
Sharing communication
Beamforming

SCA SCA

Stage - III

SCA Intruder Artificial Dust

Base (to disrupt communication)
Station
UDN SCA

Device to Device
Communication

Fig. 5. Impact of artificial rain and artificial dust in the 5G communicating network

b) Stage-II: Impact of attenuation on security due to AR or

𝑃 𝑟
The attenuation occurring in a signal while traveling through ( 𝑟) = 10 log10 𝑞 + 10𝜓 log10 − 𝜙(𝑑𝑏) − 𝑁𝑅 𝑑𝑆𝐶 −
𝑃𝑡 (𝑑𝑏) 𝑟𝑜
obstacles with depth 𝜔 is approximately given as:
𝑁𝑅 𝑑𝐴𝐵 − 𝑁𝑅 𝑑𝑅𝐸𝐹 − 𝑁𝑅 𝑑𝑃𝑂𝐿 (108)
𝑗(𝜔) = 𝑒 −𝑁𝜔 (100) where, 𝑑𝑆𝐶 is the scattering depth, 𝑑𝐴𝐵 is the absorption depth,
where N denotes the constant of attenuation and depends on the 𝑑𝑅𝐸𝐹 is the refraction depth, 𝑑𝑃𝑂𝐿 is the polarization depth, 𝑁𝑅
material of the object and its dielectric properties. Equation specific attenuation that depends on the dielectric properties of
(100) can also be written as the rain droplet and its composition. The specific attenuation in
presence of an artificial rainfall can be determined by using
𝑗(𝜔𝑡 ) = 𝑒 −𝑁 ∑𝑖 𝜔 (101) power-law relationship given as:
𝑗(𝜔𝑡 ) = 𝑒 −𝑁𝜔𝑡 (102) 𝑁𝑅 = 𝜃𝑅𝜀 (109)
where 𝜃 and 𝜀 are the coefficients and depends on the
where 𝜔𝑡 defines the sum of random object depths from which
frequency f (in GHz) and lies in ranges from 1to 6 GHz, 28 –
signals travel. Using the central limit theorem, the variable 𝜔𝑡
32GHz and up to 64GHz. Using power-law coefficients (curve-
can be simplified by approximating the Gaussian random
fitting), 𝜃 can also be expressed as:
variable. Consider a distance 𝑟𝑥,𝑦 such that NR denotes the
log10 𝑓−𝜁𝑖 2
specific attenuation constant of a signal, due to artificial log10 𝜃 = ∑4𝑖=1 (𝛿𝑖 exp [− ( ) ]) + 𝑎𝑘 log10 𝑓 + 𝑏𝑘
𝜗𝑖
rainfall. (110)
𝑑𝑅 = 𝑑𝑆𝐶 + 𝑑𝐴𝐵 + 𝑑𝑅𝐸𝐹 + 𝑑𝑃𝑂𝐿 (103) where f is the frequency, a, b, 𝛿, 𝜁, and 𝜗 are coefficients of 𝜃
Such that attenuation due to rainfall can be given as: 𝜀 can be determined as :
𝑥 ′ (𝑑𝑅 ) = 𝑒 𝑁𝑅 𝑑𝑅 (104) 4
log10 𝑓 − 𝜁𝑖 2
𝑥 ′ (𝑑𝑅 ) = 𝑒 𝑁𝑅 𝑑𝑆𝐶 + 𝑒 𝑁𝑅 𝑑𝐴𝐵 + 𝑒 𝑁𝑅 𝑑𝑅𝐸𝐹 + 𝑒 𝑁𝑅 𝑑𝑃𝑂𝐿 (105) 𝜀 = ∑ (𝛿𝑖 exp [− ( ) ]) + 𝑎𝑘 log10 𝑓 + 𝑏𝑘
𝜗𝑖
𝑖=1
In the dB scale:
(111)
(𝑥 ′ (𝑑𝑅 ))𝑑𝑏 = 𝑁𝑅 𝑑𝑆𝐶 + 𝑁𝑅 𝑑𝐴𝐵 + 𝑁𝑅 𝑑𝑅𝐸𝐹 + 𝑁𝑅 𝑑𝑃𝑂𝐿 (106) The coefficient 𝜃 can be determined either for horizontal
polarization 𝜃𝐻 or for vertical polarization (𝜃𝑉 ). Similarly, 𝜀
where 𝑥 ′ (𝑑𝑅 ) denotes the attenuation of a signal due to
can also be determined for horizontal polarization (𝜀𝐻 ) or
artificial rain caused by an intruder. The equation (100) for path
vertical polarization (𝜀𝑉 ). For circular, linear and other path
loss, including the effect of artificial attenuation, can be given
geometries the coefficients 𝜃 and 𝜀 can be calculated as:
as:
(𝜃𝐻 + 𝜃𝑉 +(𝜃𝐻 −𝜃𝑉 ) cos2 𝛼 cos2𝛽)
𝑃
( 𝑟) = 10 log10 𝑞 + 10𝜓 log10
𝑟
− 𝜙(𝑑𝑏) − (𝑥 ′ (𝑑𝑅 ))𝑑𝑏 𝜃= (112)
2
𝑃𝑡 (𝑑𝑏) 𝑟𝑜
(𝜃𝐻 𝜀𝐻 + 𝜃𝑉 𝜀𝑉 +(𝜃𝐻 𝜀𝐻 −𝜃𝑉 𝜀𝑉 ) cos2 𝛼 cos2𝛽)
(107) 𝜀= (113)
2𝜃

17
 TABLE II D2D communication involves the relay device for
RANGE OF PATH LOSS COMPONENT FOR DIFFERENT SCENARIOS
communication, consequently gives upsurge to the growth of
Scenario Path loss component (𝝍) security issues, as relays are considered to be least secure
1. Urban macrocells 3.7 - 6.5
communication identities in comparison to the BS and SCA.
2. Building (same floor) 1.6 – 3.5
3. Urban microcells 2.7 - 3.5 Therefore makes the D2D communication more prone to
4. Building (multiple floors) 2-6 attacks. In Fig. 6. considering the case of D2D communication
5. Home 3 with device relaying assisted by the controlled link provided by
6. Store 1.8 – 2.2 the BS. A device present at the edge of the cell where the
7. Factory 1.6 - 3.3 coverage of the signal is weak. Following the D2D
where 𝛼 is the angle of path elevation and 𝛽is the tilt angle of communication where another device present in suitable
polarization relative to the horizontal coverage zone of the BS communicates with a device presen at
an inadequate coverage, thereby acting as a relay. The BS is
c) Stage-III: Impact of attenuation on security due AD
responsible for allocating resources to the devices.
From Fig. 5, consider the same distance 𝑟𝑥,𝑦 from transmitter
The scenario of a half-duplex attack is being modeled by
antenna to receiver antenna such that V denotes the constant of involving three stages. The first stage involves the intrusion of
attenuation occurring due to artificial dust x’(D) denotes the artificial noise to the device present at a poor coverage area by
attenuation of signal due to artificial dust due to destructive an intruder who continuously examines the communication
interferences are resulting in a decrease in signal strength. environment. The intrusion of artificially generated noise
Table II shows the range of path loss components defined for creates the effect of jamming while performing D2D
various scenarios, such as macrocells, microcells. From communication. The second stage involves the continuous ping
equation (100) the attenuation in a signal while crossing a at the relay device. This stage involves the mechanism
distance through artificial dust with depth D, as the associated with the uplink communication from a relay device
combination of depths occurring due to scattering, absorption to the BS. Authentication of the relay device is being requested
and cross-polarization is evaluated as: during the uplink phase for the allocation of resources. The
𝑥 ′ (𝐷) = 𝑒 𝑉𝐷 (114) third stage involves the response received by the intruder from
The above equation (109) can also be evaluated as: the relay device, followed by the occupancy of the allocated
𝑥 ′ (𝐷) = 𝑒 𝑉𝐷𝑆𝐶 + 𝑒 𝑉𝐷𝐴𝐵 + 𝑒 𝑉𝐷𝑐𝑝 (115) resources. During the downlink communication from the BS to
In the dB scale relay device, the intruder pretends as a relay device and grabs
the information present in the downlink and takes hold of
(𝑥 ′ (𝐷))𝑑𝑏 = 𝑉𝐷𝑆𝐶 + 𝑉𝐷𝐴𝐵 + 𝑉𝐷𝑐𝑝 (116)
allocated resources. The mathematical analysis for the HD
Therefore, the attenuation occurring in a signal due to artificial attack is modeled as:
dust can be evaluated for path-loss by using equation (98) as: The equations of state transition can be evaluated by
𝑃 𝑟
( 𝑟) = 10 log10 𝑞 + 10𝜓 log10 − 𝜙(𝑑𝑏) − (𝑥 ′ (𝐷))𝑑𝑏 (117) considering the scenario of transmission time intervals for RRC
𝑃𝑡 (𝑑𝑏) 𝑟𝑜
configuration. At time 𝑡1 , idle RRC_ state of authenticated user
In the dBm scale equipment makes a request to gNB for RRC setup, as shown in
𝑟
𝑃𝑟(𝑑𝑏𝑚) = 𝑞𝑑𝑏 + 10𝜓 log10 − 𝜙(𝑑𝑏) − (𝑥 ′ (𝐷))𝑑𝑏𝑚 (118) Fig. 7. The setup request consists of an authentication stamp
𝑟𝑜
Such that, (𝑘𝑡1 ), identification id 𝑒𝑡1 , indication bits 𝐶𝑡1 specifying
(𝐶𝑠 )𝑖 = (𝐶𝑢 − (𝐶𝑒𝑣 )𝑖 ) < 𝐶𝑇 proximity indication or Feature Group Indication (FGI).
(118𝑎)
Therefore, for the transmission time interval 𝑡1 in the presence
where 𝑖 =AR, AD, 𝐶𝑢 and (𝐶𝑒𝑣 )𝑖 is the capacity of the user and
of an intruder, the received signal at gNB send by the UE can
intruder (AR, AD), (𝐶𝑠 )𝑖 is the secrecy capacity achieved after
be defined as:
AR and AD, less than the threshold capacity 𝐶𝑇 . Therefore,
creating a favorable attacking scenario for the intruder (𝐶𝑠 )𝑖 . 𝑦𝑡1 = (𝑘𝑡1 , 𝑒𝑡1 , 𝐶𝑡1 , 𝑥𝑡1 , ℎ𝑡1 )𝑜𝑡1 + 𝑛𝑡1 (119)
An assumption is made where an intruder is completely known
B. Phase Second: Execution of Half-duplex attack in the D2D
WCN: of channel state information and possesses access to the
identity id of the valid user equipment. During the Transition
D2D is one of the emerging technology of wireless Time Interval (TTI) 𝑡2 , it is assumed that the intruder is capable
communication to satisfy demands as reduced delay, efficient of concealing the identity of authenticated user equipment and
throughput, reduced power consumption. However, D2D receives the RRC setup in the downlink phase. The received
wireless communication technology does not involve the direct signal by the intruder instead of valid UE can be expressed as:
monitoring of the central network, thus creates a question over
𝑦𝑡2𝑒𝑣 = (𝑘𝑡2 , 𝑒𝑡2 , 𝐶𝑡2 , 𝑥𝑡2 , 𝑎𝑡2 , ℎ𝑡2𝑒𝑣 )𝑜𝑡2 + 𝑛𝑡2𝑒𝑣 (120)
the security of D2D communication. Also, D2D technology
entails two main stages. The initial stage involves At the same transition time interval, intruder transmits the RRC
communication through the central network such as BS, and setup signal along with a random sequence to the authenticated
the second stage involves the communication between devices. UE. The received signal by the valid UE sent by an intruder is
given by:
18
 Device to Device Uplink
Communication
Smart homes Smart grid
Intrusion Continuous
Downlink Health care
of Ping attack IOT
Artificial Healthcare
Noise Half duplex
attack Industrial
automation

Botnet
UDN
Base station
Valid user Bandwidth
Spoofing
Intruder
SCA(Small cell
access) point
Relay Spectrum
Sharing
Mobile User
Pico cell Base
Station

Fig. 6. A scenario of Half-Duplex (HD) attack

𝑦𝑡2 = (𝑘𝑡2 , 𝑒𝑡2 , 𝐶𝑡2 , 𝑥𝑡2 , 𝑎𝑡2 , ℎ𝑡2 , 𝑟𝑡2 )𝑜𝑡2 + 𝑛𝑡2 (121) 𝑆(𝑦𝑡𝑛 [𝑡𝑛 ] = 𝑢) = 𝑦𝑡1 [𝑡1 ] + 𝑦𝑡2 [𝑡2 ] + 𝑦𝑡3 [𝑡3 ] + ⋯ + 𝑦𝑡𝑁 [𝑡𝑁 ]
For the case of a connection failure mechanism, such that (125)
intruder again occupies the idle state and repeats the process. Therefore, the probability of successful reception at the valid
The attacking mechanism of Half- Duplex attack is defined as UE for TTI 𝑡𝑛 , can be expressed as:
the involvement of eavesdropping of sole downlink process for
the TTI 𝑡2 instead of attacking both the phases, i.e., UL and DL. 𝑃𝐷𝐿 (𝑆(𝑦𝑡𝑛 [𝑡𝑛 ] = 𝑢)) = (𝑛𝑢)𝑝𝐷𝐿
𝑢 𝑛−𝑢
𝑞𝐷𝐿 (126)
The probability of occurrence of Half-Duplex attack can be Similarly, for TTI 𝑡𝑛 the probability of successful reception at
evaluated as: an intruder or the probability of HD attack can be expressed as:
Considering two possible scenarios for downlink reception: 𝑃𝐷𝐿𝑒𝑣 (𝑆(𝑦𝑡𝑛 [𝑡𝑛 ] = 𝑢)) = (𝑛𝑢)𝑞𝐷𝐿
𝑢 𝑛−𝑢
𝑝𝐷𝐿 (127)
1. Successful reception of DL RRC setup by the valid UE For the scenario, when the intruder attacks both UL and DL
denoted 𝑝𝐷𝐿 (success) during RRC state transition from idle to connected. For UL
2. Successful reception of DL RRC setup by an intruder RRC request from UE to gNB at TTI 𝑡1 , from equation (119)
denoted by 1 − 𝑝𝐷𝐿 (failure) the received signal by an intruder can be expressed as:
𝑦𝑡𝑛 [𝑡𝑛 ] = 𝑦𝑡1 [𝑡1 ], 𝑦𝑡2 [𝑡2 ], 𝑦𝑡3 [𝑡3 ], … , 𝑦𝑡𝑁 [𝑡𝑁 ] (122) 𝑦𝑡1𝑒𝑣 = (𝑘𝑡1 , 𝑒𝑡1 , 𝐶𝑡1 , 𝑥𝑡1 , ℎ𝑡𝑒𝑣 )𝑜𝑡1 + 𝑛𝑡𝑒𝑣 (128)
where n=1,2,3, …, N. N denotes Nth natural number. 𝑦𝑡𝑛 [𝑡𝑛 ] From equation (125) and (126), equation (128) can also be
is considered to be a discrete-time process whose density represented for the condition of successful transmission to gNB
function is given by: (𝑝𝑈𝐿 ) as:
𝑓𝑦𝑡𝑛 [𝑡𝑛] (𝑦𝑡𝑛 [𝑡𝑛 ]) = 𝑞𝐷𝐿 𝛿(𝑦𝑡𝑛 [𝑡𝑛 ]) + 𝑝𝐷𝐿 𝛿(𝑦𝑡𝑛 [𝑡𝑛 ] − 1) 𝑃𝑈𝐿 (𝑆(𝑦𝑡𝑛 [𝑡𝑛 ] = 𝑢)) = (𝑛𝑢)𝑝𝑈𝐿
𝑢 𝑛−𝑢
𝑞𝑈𝐿 (129)
(123) where 𝑞𝑈𝐿 = 1 − 𝑝𝑈𝐿 , for the scenario where the successful
where 𝛿(. ) depicts unit impulse function, 𝑞𝐷𝐿 = 1 − transmission to intruder takes place, is expressed as:
𝑝𝐷𝐿 represents the condition of failure. The second order 𝑃𝑈𝐿𝑒𝑣 (𝑆(𝑦𝑡𝑛 [𝑡𝑛 ] = 𝑢)) = (𝑛𝑢)𝑞𝑈𝐿
𝑢 𝑛−𝑢
𝑝𝑈𝐿 (130)
density function for the received signal 𝑦𝑡𝑛 [𝑡𝑛 ] can be 𝑃𝑡𝑜𝑡𝑎𝑙 = 𝑃𝑈𝐿 (𝑆(𝑦𝑡𝑛 [𝑡𝑛 ])) + 𝑃𝐷𝐿 (𝑆(𝑦𝑡𝑛 [𝑡𝑛 ])) −
expressed as: 𝑃𝑈𝐿 (𝑆(𝑦𝑡𝑛 [𝑡𝑛 ])) ∩ 𝑃𝐷𝐿 (𝑆(𝑦𝑡𝑛 [𝑡𝑛 ])) (131)
𝑓𝑦𝑡1 [𝑡1] 𝑦𝑡2 [𝑡2] (𝑦𝑡1 [𝑡1 ] 𝑦𝑡2 [𝑡2 ]) = 𝑞𝐷𝐿 2 𝛿(𝑦𝑡1 [𝑡1 ])𝛿(𝑦𝑡2 [𝑡2 ]) + Since, 𝑦𝑡𝑛 [𝑡𝑛 ] is IID., therefore,
𝑝𝐷𝐿 𝑞𝐷𝐿 𝛿(𝑦𝑡1 [𝑡1 ] − 1)𝛿(𝑦𝑡2 [𝑡2 ] − 1) (124) 𝑃𝑈𝐿 (𝑆(𝑦𝑡𝑛 [𝑡𝑛 ])) ∩ 𝑃𝐷𝐿 (𝑆(𝑦𝑡𝑛 [𝑡𝑛 ])) = 0 (132)
At the authenticated UE, for 𝑡𝑛 time intervals, the sum of the Thus, the above equation can be expressed as:
received signal is given by:

19
 TABLE III
COMPARISON OF THE PROPOSED SCHEME WITH EXISTING SCHEMES
Ref. Objective Associate parameters Attacking scenario Dust Rain

Analysis Security Half General Security General Security

impact Duplex Analysis impact Analysis impact
DL UL attack
[92] Single and multi-channel D2D sum rate,        
downlink resource management security probability
for the improvement of physical
layer security
[151] Enhancement of the security by D2D sum rate,        
utilizing the interference caused probability of CU
by D2D to cellular users security QoS is
satisfied
[152] Security enhancement by Secrecy probability of        
analyzing the inference created cellular link, SINR,
by D2D communication using connection
accurate closed-form bounds probability of cellular
link
[153] Impact of rain on the Rain attenuation,        
propagation of millimeter specific attenuation
waves at 26 GHz
[154] Effect of atmospheric gases, Specific attenuation,        
rain, foliage, and diffraction at fog attenuation,
28GHz, 30GHz and 60GHz foliage attenuation,
rain attenuation
[155] Statistics of rain attenuation Rain attenuation,        
over terrestrial paths for the cell excedance
size of 200m probability, predicted
FMT gain
[156] Impact of rain on achievable Rain attenuation,        
rates and capacities of capacity, bit rate
millimeter wave MIMO system
[157] Study of phase delay and Attenuation, visibility        
attenuation caused by dust and
sand storms using Ghobrial et
al. ’s formula
[158] Effect of dust and sand storms Specific attenuation,        
on millimeter wave signal propagation factor
attenuation
[159] Calculation of wave attenuation Averaged extinction        
in the storms of rain and dust at cross-section (ECS),
a frequency of 10-100GHz attenuation
[160] Effectt of dust and sand storms Attenuation, cross        
on propagation of microwaves polar discrimination
[161] Effect of dust on FSO system of Attenuation, specific        
communication in semi-arid attenuation, visibility
and arid atmosphere
[162] An equation for radar based on Power ratios        
parabolic wave equation is
computed in the presence of
dust and sand storm for the
prediction of backscattering
millimeter wave
[163] To observe the effect of dust Cross polarization        
and sand-dust storms in earth discrimination,
satellite links on microwave attenuation,
propagation attenuation
coefficient
proposed To observe the impact of rain, Secrecy rate, secrecy        
dust and half duplex attack at a non-outage
frequency 10-100 GHz probability, capacity,
distance, attenuation,
specific attenuation

20
 1. RRC Setup request (t1) of an outage when the SC is smaller than a certain defined
2. RRC Setup (t2)
value. SOP tends to progress with the rise in the number of
3. RRC Setup malicious intruders.
4. RRC Setup
with random
complete (t4)
UE sequence (t3) INTRUDER gNB B. Resource spoofing
RRC_IDLE RRC_IDLE

6. Initial UE message
5a. Connection failure

7. DL NAS transport
5. RRC connected
Due to increased vulnerable breaches in 5G communication,
network attacks are prominently increasing day by day. These
breaches led to catching sight for the intruder to attack. The
INTRUDER
interest of an attacker can be either of active type or of the
RRC_CON AMF passive type. The resource spoofing is the fundamental
NECTED attention grabbing the interest of an attacker to attack the user
Fig. 7. Half-Duplex attack using RRC setup and spoof the resources allocated to the legitimate user. It
𝑃𝑡𝑜𝑡𝑎𝑙 = 𝑃𝑈𝐿 (𝑆(𝑦𝑡𝑛 [𝑡𝑛 ])) + 𝑃𝐷𝐿 (𝑆(𝑦𝑡𝑛 [𝑡𝑛 ])) (133) targets control channels to spoof allocated resources. It
primarily affects UDN, D2D technologies of 5G. Therefore,
Now, the probability of miss-rate can be obtained as: security performance is required to be enhanced by proper
(1−𝑃𝑈𝐿𝑒𝑣 )+(1−𝑃𝐷𝐿𝑒𝑣 ) mechanisms to prevent the inefficient, illegal, and
𝑃(𝑚𝑖𝑠𝑠 − 𝑟𝑎𝑡𝑒𝐹𝐷 ) = (134)
(𝑃𝑇𝑜𝑡𝑎𝑙 ) unproductive usage of resources by an attacker.
where FD denotes the Full-Duplex mode, i.e., intruder trying
C. Complexity
to eavesdrop both during uplink and downlink.
Similarly, As the demands of the users increase nowadays, which
correspondingly increases the deployment of the increased
(1−𝑃𝐷𝐿𝑒𝑣 )
𝑃(𝑚𝑖𝑠 − 𝑟𝑎𝑡𝑒𝐻𝐷 ) = (135) number of access points, the number of cells. This leads to the
(𝑃𝑇𝑜𝑡𝑎𝑙 )
development of UDNs, Massive MIMO technologies of 5G
Comparing equation (134)and (135), we get:
WCN. These deployed structures require proper management
Such that 𝑃(𝑚𝑖𝑠 − 𝑟𝑎𝑡𝑒𝐹𝐷 ) > 𝑃(𝑚𝑖𝑠 − 𝑟𝑎𝑡𝑒𝐻𝐷 ) and control for their appropriate functioning to fulfill the
From equation (135), it can be concluded that the miss-rate of seamless connectivity of links. This results in an upsurge in a
the Full Duplex attack is greater than the miss-rate of HD number of divisions present in the WCN, creating suitable
attack. Thereby increasing the chance of the intruder to attack breaches for a malicious user to attack. Therefore, creates a
the network by the execution of HD attack as compared to the security issue in existing technologies of 5G WCN.
full-duplex attacking mechanism.
D. Identity spoofing
Table III provides a brief comparison of related study and the
proposed methodology in view of security. It defines the One of the chief challenges in the current security of the 5G
objective of previously attempted work along with the WCN is the attack on the identity of the authentic users by
considered associated parameters of security. malicious attackers. In the case of UDN, D2D identity spoofing
is one of the chief challenges which is required to be
IV. CHALLENGES IN CONTEXT OF SECURITY encountered. As an identity parameter is included in the
Regardless of the existing work on the security of 5G WCN, primary aspect of the security, such as authentication,
various significant research challenges are required to be authorization of authorized users. Therefore, the security of
addressed to prevent compromised features of the security. identity is required to enable valid users to authorized access to
This section describes some present and future research the authenticated resources at the right time while maintaining
directions in view of security of WCN integrity.

A. Secrecy capacity (SC) and SOP E. CSI availability

SC and SOP are the two essential performance parameters of The availability of CSI is another challenge that is required to
security. Based on the security impact of the network, these two determine the security performance of the network. In case
parameters are analyzed. SC is precisely defined as the UDN, Massive MIMO, D2D, availability of CSI of users
difference between the capacity of the main channel and necessitates the secure transmission of the information.
capacity of the channel malicious intruder. It defines the Security tends to increase with an increase in the accuracy of
maximum rate at which data can be communicated from the CSI.
sender to the receiver under the appropriate performance of F. Handover management attacks
secrecy. Security capacity is major affected by the presence of
In 5G WCN, UDN consists of the bulky number of microcells,
the eavesdropper. As the number of eavesdroppers increase,
picocells, femtocells to fulfill the growing demands of the
secrecy capacity inclines to decrease. In the case of 5G WCN,
subscribers. Consequently, with an increased number of cells,
SC and SOP are examining parameters that primarily affect the
handover tends to increase frequently. As handover site is the
performance of the network, including a D2D communication,
most appropriate spot where an attack is most probable to
UDN. However, these two factors also affect the performance
happen due to a decrease in secrecy rate. Therefore, in
of other technologies of 5G WCN. SOP defines the probability
technologies of 5G such as UDN, Massive MIMO suitable
21
handover management is required to prevent the sensitive site of security are required to be acquired for the development of
of handover from being attacked and thus increasing the security services in 5G wireless communication networks
security of the network.
APPENDIX
G. Latency TABLE IV
LIST OF ACRONYMS
One of the chief issues that deteriorate the functionality of the [Link]. Acronym Full form
5G security network is latency. It creates a key impact on the 1. AD Artificial Dust
service of security to the fundamental parameters of the 2. ANFIS Adaptive Neuro-Fuzzy Inference System
3. AR Artificial Rain
security, including integrity, confidentiality, and
4. AP Access Point
authentication. The users existing in the 5G network 5. AN Artificial Noise
necessitate high pace connectivity with uninterrupted 6. AWGN Additive White Gaussian Noise
communication links. However, due to latency, the delay is 7. BB Beam Broadening
8. BM Beam Merging
created in the communication network followed by its 9. CSI Channel State Information
consequential effect with an adequate increase of chance for 10. CBRS Citizens Broadband Radio Service
the intruders to examine or eavesdrop the communication 11. D2D Device to Device
between the transmitter and receiver. By reason of its effect, 12. DoS Denial of Service
13. ECS Extinction Cross section
the threat is introduced to the security environment of the 14. FMT Fade Mitigation Techniques
communication network. 15. FGI Feature Group Indication
16. GSC General Selection Combining
H. Jamming attack 17. IDS Intrusion Detection System
An important parameter that affects the performance of the 18. IoT Internet of Things
19. MCC Minutia Cylinder Code
security of the 5G network. Jamming in UDN, D2D gives an 20. MDL Minimum Description Length
optimal decrease in the security of these technologies of 5G. In 21. MIMO Multiple Input Multiple Output
a jamming attack, an attacker sends more powered signals 22. ML Machine Learning
23. MMSE Minimal Mean Square Error
using an equivalent frequency range, thereby disrupting
24. MRC Maximal Ratio Combining
legitimate communication between transmitter and receiver. 25. MSRS Minimal Self-interference Relay Selection
These challenges are the chief factors that affect the security of 26. NN Neural Networks
the 5G network. Therefore, the solution to these problems 27. NOMA Non Orthogonal Multiple Access
provides a challenging aspect in the field of research and 28. ORS Optimal Relay Selection
29. OFDMA Orthogonal Frequency Division Multiple
academia to provide secure communication. Several solutions Access
were approached to countermeasure the challenges mentioned 30. PDF Probability Density Function
above to enhance the performance of the security. For 31. PET Privacy Enhancing Technology
increased complexity of the network, beamforming tactic has 32. PLA Physical Layer Authentication
been adopted. Various mechanisms of intrusion detection are 33. PLS Physical Layer Security
34. PRS Physical Relay Selection
proposed such as Adaptive Intrusion Detection system (AIDS) 35. PSM Password Strength Meters
using HMM (Hidden Markov Model) for bandwidth spoofing, 36. PUE Primary User Emulation
Intrusion Detection for using Adaptive Neural Fuzzy Inference 37. PUF Physical Unclonable Functions
System, handover management schemes such as TPA, RBA for 38. QoS Quality of Service
the security and protection of most feasible attacking zone of 39. RBA Region Based Algorithm
40. RRC Radio Resource Control
the network. 41. RTC Request To Cooperate
42. RSC Random Selection Combining
V. CONCLUSION 43. SAFH Secret Adaptive Frequency Hopping
44. SCA Small Cell Access
This paper unifies the comprehensive survey based on a 45. SINR Signal to Interference plus Noise Ratio
security perspective in different technologies of 5G NR, 46. SLS Sector Level Sweep
followed by the investigation of their security solutions. The 47. SNR Signal to Noise Ratio
generalized security model for the technologies of 5G NR is 48. SOP Secrecy Outage Probability
49. SR Secrecy Rate
expressed in a systemized manner. Moreover, the integration 50. SSID Service Set Identifier
of these technologies raises additional security issues, 51. SVM Support Vector Machines
specifically in terms of authentication, confidentiality, privacy, 52. SWIPT Simultaneous Wireless Information and
Power Transfer
and integrity. Further, a schematized attacking scenario is 53. TAS Transmit Antenna Selection
proposed by operating the concept of artificial rain or artificial 54. TPA Traffic Pattern Analysis
dust, for which the security model is derived on account of the 55. TSD Threshold based Switched Diversity
secrecy rate parameter. This implicates the scenario for HD 56. TTI Transition Time Interval
57. UDN Ultra Dense Network
attack where an intruder, instead of targeting both uplink and 58. UE User Equipment
downlink, attacks the downlink exclusively to obtain the 59. WCN Wireless Communication Network
allocated resources that were meant to be allocated to the 60. ZF Zero Forcing
authenticated user. Therefore, adaptive and novel mechanisms
22
 TABLE V
CURRENT SECURITY PROJECTS IN WCN
[Link]. Project name Aim of research Area of research HTTP location
1. NetGuard Security To secure and protect physical, Network integrity [Link]
Management virtual, and hybrid mobile management
networks.
2. 5G ensure To achieve trustworthy, viable and Authentication, privacy, [Link]
secure 5G network trust, security monitoring,
network management, and
virtualization
3. ANASTACIA(Advance to identify cyber-security Security Orchestration [Link]
d Networked Agents for pertaining to researching, intelligently Leveraging and
Security and Trust developing and demonstrating a SDN and NFV
Assessment holistic solution capable of trust
in CPS (Cyber-Physical and security by design for CPS
Systems) / IOT based on IoT and Cloud
Architectures) architectures.
4. Nokia Threat To detect and identify malware Detection and identification [Link]
Intelligence Center infections of malware on the basis of intelligence
detection rules, command-
and-control communication,
and other network behavior
5. ECLEXYS To oppose attacks and analyze Data security by modular [Link]
Cybersecurity intrusion detection firewall and intrusion
detection system
6. Cybersecurity Locate vulnerabilities and access Protection of the [Link]
points into an organization’s cyber cyberinfrastructure security/
networks, web application server,
WIFI networks
7. Radware To protect against the advanced Protection mechanism using [Link]
threats machine learning based on
cybersecurity
8. CyberSec4Europe To address cyber threats and Cybersecurity competence [Link]
security problems network e
9. SMOG- To provide trusted interaction Fog security [Link]
Security Mechanisms between the entities present in a
for fog cOmputinG fog ecosystem
10. DISS-IIoT To create innovative security Internet of things and critical [Link]
Design and services to determine probable infrastructure
Implementation of security gaps and cover illegal
Security Services for the access
Industrial Internet of
Things

[8] Luong, N. C., Wang, P., Niyato, D., Liang, Y. C., Han, Z., & Hou, F.
REFERENCES (2019). Applications of economic and pricing models for resource
[1] Shafi, M., Molisch, A. F., Smith, P. J., Haustein, T., Zhu, P., De Silva, management in 5G wireless networks: A survey. IEEE Communications
P., & Wunder, G. (2017). 5G: A tutorial overview of standards, trials, Surveys & Tutorials, 21(4), 3298-3339
challenges, deployment, and practice. IEEE journal on selected areas in [9] Hossain, E., & Hasan, M. (2015). 5G cellular: key enabling technologies
communications, 35(6), 1201-1221. and research challenges. IEEE Instrumentation & Measurement
[2] Agiwal, M., Roy, A., & Saxena, N. (2016). Next generation 5G wireless Magazine, 18(3), 11-21
networks: A comprehensive survey. IEEE Communications Surveys & [10] Parvez, I., Rahmati, A., Guvenc, I., Sarwat, A. I., & Dai, H. (2018). A
Tutorials, 18(3), 1617-1655. survey on low latency towards 5G: RAN, core network and caching
[3] Dai, Y., Xu, D., Maharjan, S., & Zhang, Y. (2018). Joint computation solutions. IEEE Communications Surveys & Tutorials, 20(4), 3098-3130
offloading and user association in multi-task mobile edge [11] Gope, P., Lee, J., Hsu, R. H., & Quek, T. Q. (2019). Anonymous
computing. IEEE Transactions on Vehicular Technology, 67(12), 12313- Communications for Secure Device-to-Device-Aided Fog Computing:
12325. Architecture, Challenges, and Solutions. IEEE Consumer Electronics
[4] Dai, L., Wang, B., Ding, Z., Wang, Z., Chen, S., & Hanzo, L. (2018). A Magazine, 8(3), 10-16.
survey of non-orthogonal multiple access for 5G. IEEE Communications [12] Gupta, A., Jha, R. K., Gandotra, P., & Jain, S. (2017). Bandwidth
Surveys & Tutorials, 20(3), 2294-2323. spoofing and intrusion detection system for multistage 5G wireless
[5] Chen, M., & Hao, Y. (2018). Task offloading for mobile edge computing communication network. IEEE Transactions on Vehicular
in software defined ultra-dense network. IEEE Journal on Selected Areas Technology, 67(1), 618-632.
in Communications, 36(3), 587-597. [13] Gupta, A., Jha, R. K., & Jain, S. (2017). Attack modeling and intrusion
[6] Dai, Y., Xu, D., Maharjan, S., Chen, Z., He, Q., & Zhang, Y. (2019). detection system for 5G wireless communication network. International
Blockchain and deep reinforcement learning empowered intelligent 5g Journal of Communication Systems, 30(10), e3237.
beyond. IEEE Network, 33(3), 10-17. [14] Devi, R., Jha, R. K., Gupta, A., Jain, S., & Kumar, P. (2017).
[7] Abrol, A., & Jha, R. K. (2016). Power optimization in 5G networks: A Implementation of intrusion detection system using adaptive neuro-fuzzy
step towards GrEEn communication. IEEE Access, 4, 1355-1374.

23
 inference system for 5G wireless communication network. AEU- [37] Zhu, J., Schober, R., & Bhargava, V. K. (2015). Linear precoding of data
International Journal of Electronics and Communications, 74, 94-106. and artificial noise in secure massive MIMO systems. IEEE Transactions
[15] Rupprecht, D., Dabrowski, A., Holz, T., Weippl, E., & Pöpper, C. (2018). on Wireless Communications, 15(3), 2245-2261.
On security research towards future mobile network generations. IEEE [38] Mukherjee, P., & Ulukus, S. (2017). Secrecy in MIMO networks with no
Communications Surveys & Tutorials, 20(3), 2518-2542. eavesdropper CSIT. IEEE Transactions on Communications, 65(10),
[16] Chopra, G., Jain, S., & Jha, R. K. (2017). Possible security attack 4382-4391
modeling in ultradense networks using high-speed handover [39] Zhu, F., Gao, F., Lin, H., Jin, S., Zhao, J., & Qian, G. (2018). Robust
management. IEEE Transactions on Vehicular Technology, 67(3), 2178- beamforming for physical layer security in BDMA massive MIMO. IEEE
2192. Journal on Selected Areas in Communications, 36(4), 775-787.
[17] Yang, N., Wang, L., Geraci, G., Elkashlan, M., Yuan, J., & Di Renzo, M. [40] Venkategowda, N. K., Tandon, N., & Jagannatham, A. K. (2014).
(2015). Safeguarding 5G wireless communication networks using MVDR-based multicell cooperative beamforming techniques for
physical layer security. IEEE Communications Magazine, 53(4), 20-27. unicast/multicast MIMO networks with perfect/imperfect CSI. IEEE
[18] Chen, X., Ng, D. W. K., Gerstacker, W. H., & Chen, H. H. (2016). A Transactions on Vehicular Technology, 64(11), 5160-5176.
survey on multiple-antenna techniques for physical layer security. IEEE [41] Hoang, T. M., Duong, T. Q., Suraweera, H. A., Tellambura, C., & Poor,
Communications Surveys & Tutorials, 19(2), 1027-1053. H. V. (2015). Cooperative beamforming and user selection for improving
[19] Wang, C., & Wang, H. M. (2016). Physical layer security in millimeter the security of relay-aided systems. IEEE Transactions on
wave cellular networks. IEEE Transactions on Wireless Communications, 63(12), 5039-5051.
Communications, 15(8), 5569-5585. [42] Zhu, F., & Yao, M. (2015). Improving physical-layer security for CRNs
[20] Ahmad, I., Kumar, T., Liyanage, M., Okwuibe, J., Ylianttila, M., & using SINR-based cooperative beamforming. IEEE Transactions on
Gurtov, A. (2018). Overview of 5G security challenges and Vehicular Technology, 65(3), 1835-1841.
solutions. IEEE Communications Standards Magazine, 2(1), 36-43. [43] Wang, L., Wong, K. K., Elkashlan, M., Nallanathan, A., & Lambotharan,
[21] Wu, Y., Khisti, A., Xiao, C., Caire, G., Wong, K. K., & Gao, X. (2018). S. (2016). Secrecy and energy efficiency in massive MIMO aided
A survey of physical layer security techniques for 5G wireless networks heterogeneous C-RAN: A new look at interference. IEEE Journal of
and challenges ahead. IEEE Journal on Selected Areas in Selected Topics in Signal Processing, 10(8), 1375-1389
Communications, 36(4), 679-695. [44] Hoang, T. M., Duong, T. Q., Tuan, H. D., & Poor, H. V. (2017). Secure
[22] Wang, S., Li, W., & Lei, J. (2018). Physical-layer encryption in massive massive MIMO relaying systems in a poisson field of
MIMO systems with spatial modulation. China Communications, 15(10), eavesdroppers. IEEE Transactions on Communications, 65(11), 4857-
159-171. 4870.
[23] Dean, T. R., & Goldsmith, A. J. (2017). Physical-layer cryptography [45] Akhlaghpasand, H., Razavizadeh, S. M., Björnson, E., & Do, T. T.
through massive MIMO. IEEE Transactions on Information (2017). Jamming detection in massive MIMO systems. IEEE Wireless
Theory, 63(8), 5419-5436. Communications Letters, 7(2), 242-245.
[24] You, L., Wang, J., Wang, W., & Gao, X. (2019). Secure Multicast [46] Akhlaghpasand, H., Bjornson, E., & Razavizadeh, S. M. (2020).
Transmission for Massive MIMO with Statistical Channel State Jamming Suppression in Massive MIMO Systems. IEEE Transactions on
Information. IEEE Signal Processing Letters, 26(6), 803-807. Circuits and Systems II: Express Briefs, 67(1), 182-186.
[25] Xiong, Q., Liang, Y. C., Li, K. H., Gong, Y., & Han, S. (2016). Secure [47] Ahmed, M., & Bai, L. (2018). Secrecy capacity of artificial noise aided
transmission against pilot spoofing attack: A two-way training-based secure communication in MIMO Rician channels. IEEE Access, 6, 7921-
scheme. IEEE Transactions on Information Forensics and 7929.
Security, 11(5), 1017-1026. [48] Li, Q., & Yang, L. (2018). Artificial noise aided secure precoding for
[26] Xiao, L., Chen, T., Han, G., Zhuang, W., & Sun, L. (2017). Game MIMO untrusted two-way relay systems with perfect and imperfect
theoretic study on channel-based authentication in MIMO systems. IEEE channel state information. IEEE Transactions on Information Forensics
Transactions on Vehicular Technology, 66(8), 7474-7484 and Security, 13(10), 2628-2638.
[27] Zhang, X., Guo, D., An, K., Liang, X., & Ma, W. (2018). Secure [49] Song, C. (2019). Leakage Rate Analysis for Artificial Noise Assisted
Transmission in Multi-Pair AF Relaying Massive MIMO Networks Massive MIMO with Non-coherent Passive Eavesdropper in Block-
Against Active Pilot Spoofing Attacks. IEEE Access, 7, 3547-3560. fading. IEEE Transactions on Wireless Communications. 18(4), 2111-
[28] Akgun, B., Krunz, M., & Koyluoglu, O. O. (2018). Vulnerabilities of 2124.
massive MIMO systems to pilot contamination attacks. IEEE [50] Timilsina, S., Baduge, G. A. A., & Schaefer, R. F. (2018). Secure
Transactions on Information Forensics and Security, 14(5), 1251-1263. communication in spectrum-sharing massive MIMO systems with active
[29] Zhang, W., Lin, H., & Zhang, R. (2018). Detection of pilot contamination eavesdropping. IEEE Transactions on Cognitive Communications and
attack based on uncoordinated frequency shifts. IEEE Transactions on Networking, 4(2), 390-405.
Communications, 66(6), 2658-2670. [51] Nguyen, N. P., Ngo, H. Q., Duong, T. Q., Tuan, H. D., & Tourki, K.
[30] Yang, T., Zhang, R., Cheng, X., & Yang, L. (2018). Secure massive (2018). Secure massive MIMO with the artificial noise-aided downlink
MIMO under imperfect CSI: Performance analysis and channel training. IEEE Journal on Selected Areas in Communications, 36(4),
prediction. IEEE Transactions on Information Forensics and 802-816
Security, 14(6), 1610-1623 [52] Wu, W., Gao, X., Wu, Y., & Xiao, C. (2018). Beam domain secure
[31] Chen, J., Chen, X., Gerstacker, W. H., & Ng, D. W. K. (2016). Resource transmission for massive MIMO communications. IEEE Transactions on
allocation for a massive MIMO relay aided secure communication. IEEE Vehicular Technology, 67(8), 7113-7127.
Transactions on Information Forensics and Security, 11(8), 1700-1711 [53] Zhang, X., Guo, D.,K., & B. Zhang, B. (2019). Secure Communications
[32] Li, L., Chen, Z., Petropulu, A. P., & Fang, J. (2017). Linear precoder Over Cell-Free Massive MIMO Networks With Hardware
design for an MIMO Gaussian wiretap channel with full-duplex source Impairments. IEEE Systems Journal. (To be published) doi:
and destination nodes. IEEE Transactions on Information Forensics and 10.1109/JSYST.2019.2919584.
Security, 13(2), 421-436. [54] Zhang, X., Guo, D., & An, K. (2018). Secure communication in
[33] Li, L., Petropulu, A. P., & Chen, Z. (2017). MIMO secret multigroup multicasting cell-free massive MIMO networks with active
communications against an active eavesdropper. IEEE Transactions on spoofing attack. Electronics Letters, 55(2), 96-98.
Information Forensics and Security, 12(10), 2387-2401. [55] Xu, J., Xu, W., Zhu, J., Ng, D. W. K., & Swindlehurst, A. L. (2019).
[34] Li, L., Chen, Z., Zhang, D., & Fang, J. (2015). A full-duplex Bob in the Secure Massive MIMO Communication with Low-resolution
MIMO Gaussian wiretap channel: Scheme and performance. IEEE Signal DACs. IEEE Transactions on Communications, 67(5), 3265-3278.
Processing Letters, 23(1), 107-111. [56] Wang, J., Lee, J., Wang, F., & Quek, T. Q. (2015). Jamming-aided secure
[35] Fatema, N., Hua, G., Xiang, Y., Peng, D., & Natgunanathan, I. (2018). communication in massive MIMO Rician channels. IEEE Transactions
Massive MIMO linear precoding: A survey. IEEE Systems Journal, on Wireless Communications, 14(12), 6854-6868.
12(4), 3920-3931. [57] Zhu, J., & Xu, W. (2016). Securing massive MIMO via power
[36] Kapetanovic, D., Zheng, G., & Rusek, F. (2015). Physical layer security scaling. IEEE Communications Letters, 20(5), 1014-1017.
for massive MIMO: An overview on passive eavesdropping and active [58] Zhu, J., Xu, W., & Wang, N. (2016). Secure massive MIMO systems with
attacks, IEEE Communications Magazine, 53(6), 21-27. limited RF chains. IEEE Transactions on Vehicular Technology, 66(6),
5455-5460.

24
[59] Alageli, M., Ikhlef, A., & Chambers, J. (2018). SWIPT massive MIMO [82] Ahmed, I., Khammari, H., Shahid, A., Musa, A., Kim, K. S., De Poorter,
systems with active eavesdropping. IEEE Journal on Selected Areas in E., & Moerman, I. (2018). A survey on hybrid beamforming techniques
Communications, 37(1), 233-247 in 5G: Architecture and system model perspectives. IEEE
[60] Zheng, Z., Haas, Z. J. & Kieburg, M. (2018). Secrecy Rate of Cooperative Communications Surveys & Tutorials, 20(4), 3060-3097.
MIMO in the Presence of a Location Constrained Eavesdropper. IEEE [83] Alavi, F., Cumanan, K., Ding, Z., & Burr, A. G. (2018). Beamforming
Transactions on Communications, 67(2), 1356-1370. techniques for nonorthogonal multiple access in 5G cellular
[61] Zhu, J., Schober, R., & Bhargava, V. K. (2014). Secure transmission in networks. IEEE Transactions on Vehicular Technology, 67(10), 9474-
multicell massive MIMO systems. IEEE Transactions on Wireless 9487.
Communications, 13(9), 4766-4781 [84] Zhou, B., Liu, A., & Lau, V. (2019). Successive Localization and
[62] Basciftci, Y. O., Koksal, C. E., & Ashikhmin, A. (2018). Physical-Layer Beamforming in 5G mmWave MIMO Communication Systems. IEEE
Security in TDD Massive MIMO. IEEE Transactions on Information Transactions on Signal Processing, 67(6), 1620-1635.
Theory, 64(11), 7359-7380 [85] Jayaprakasam, S., Rahim, S. K. A., & Leow, C. Y. (2017). Distributed
[63] Li, X., Zhang, Y., & Cadeau, W. (2017). Hybrid massive MIMO for and collaborative beamforming in wireless sensor networks:
secure transmissions against stealthy eavesdroppers. IEEE classifications, trends, and research directions. IEEE Communications
Communications Letters, 22(1), 81-84. Surveys & Tutorials, 19(4), 2092-2116.
[64] Zhu, X., Dai, L., & Wang, Z. (2015). Graph coloring based pilot [86] Kutty, S., & Sen, D. (2015). Beamforming for millimeter wave
allocation to mitigate pilot contamination for multi-cell massive MIMO communications: An inclusive survey. IEEE communications surveys &
systems. IEEE Communications Letters, 19(10), 1842-1845 tutorials, 18(2), 949-973.
[65] Upadhya, K., Vorobyov, S. A., & Vehkapera, M. (2017). Superimposed [87] Xu, Q., Jiang, C., Han, Y., Wang, B., & Liu, K. R. (2017). Waveforming:
pilots are superior for mitigating pilot contamination in massive An overview with beamforming. IEEE Communications Surveys &
MIMO. IEEE Transactions on Signal Processing, 65(11), 2917-2932. Tutorials, 20(1), 132-149.
[66] Ahmadi, H., Farhang, A., Marchetti, N., & MacKenzie, A. (2015). A [88] Hamoud, O. N., Kenaza, T., & Challal, Y. (2017). Security in device-to-
game theoretic approach for pilot contamination avoidance in massive device communications: a survey. IET Networks, 7(1), 14-22.
MIMO. IEEE Wireless Communications Letters, 5(1), 12-15. [89] Haus, M., Waqas, M., Ding, A. Y., Li, Y., Tarkoma, S., & Ott, J. (2017).
[67] Wang, W., Teh, K. C., Luo, S., & Li, K. H. (2018). Physical Layer Security and privacy in device-to-device (D2D) communication: A
Security in Heterogeneous Networks With Pilot Attack: A Stochastic review. IEEE Communications Surveys & Tutorials, 19(2), 1054-1079.
Geometry Approach. IEEE Transactions on Communications, 66(12), [90] Alibeigi, M., & Taherpour, A. (2018). Optimisation of secrecy rate in
6437-6449. cooperative device to device communications underlaying cellular
[68] Muppirisetty, L. S., Charalambous, T., Karout, J., Fodor, G., & networks. IET Communications, 13(5), 512-519
Wymeersch, H. (2018). Location-aided pilot contamination avoidance [91] Xu, J., Chen, L., Liu, K., & Shen, C. (2018). Designing Security-Aware
for massive MIMO systems. IEEE Transactions on Wireless Incentives for Computation Offloading via Device-to-Device
Communications, 17(4), 2662-2674. Communication. IEEE Transactions on Wireless
[69] Hu, D., He, L., & Wang, X. (2015). Semi-blind pilot decontamination for Communications, 17(9), 6053-6066.
massive MIMO systems. IEEE Transactions on Wireless [92] Wang, J., Huang, Y., Jin, S., Schober, R., You, X., & Zhao, C. (2018).
Communications, 15(1), 525-536. Resource management for device-to-device communication: A physical
[70] Gong, Z., Li, C., & Jiang, F. (2019). Pilot Decontamination in layer security perspective. IEEE Journal on Selected Areas in
Noncooperative Massive MIMO Cellular Networks Based on Spatial Communications, 36(4), 946-960.
Filtering. IEEE Transactions on Wireless Communications, 18(2), 1419- [93] Shen, W., Yin, B., Cao, X., Cai, L. X., & Cheng, Y. (2016). Secure
1433. device-to-device communications over WiFi direct. IEEE
[71] Chen, Z., & Yang, C. (2016). Pilot decontamination in wideband massive Network, 30(5), 4-9.
MIMO systems by exploiting channel sparsity. IEEE Transactions on [94] Kwon, Y., Suh, H., Oh, J., & Hwang, T. (2017). Energy efficient
Wireless Communications, 15(7), 5087-5100. communication for secure D2D underlaid cellular networks. IEEE
[72] Elijah, O., Leow, C. Y., Rahman, T. A., Nunoo, S., & Iliya, S. Z. (2015). Transactions on Vehicular Technology, 66(10), 9110-9123.
A comprehensive survey of pilot contamination in massive MIMO—5G [95] Waqas, M., Ahmed, M., Li, Y., Jin, D., & Chen, S. (2018). Social-aware
system. IEEE Communications Surveys & Tutorials, 18(2), 905-923. secret key generation for secure device-to-device communication via
[73] Jiang, X., Decurninge, A., Gopala, K., Kaltenberger, F., Guillaud, M., trusted and non-trusted relays. IEEE Transactions on Wireless
Slock, D., & Deneire, L. (2018). A framework for over-the-air reciprocity Communications, 17(6), 3918-3930.
calibration for TDD massive MIMO systems. IEEE Transactions on [96] Liu, Y., Wang, L., Zaidi, S. A. R., Elkashlan, M., & Duong, T. Q. (2015).
Wireless Communications, 17(9), 5975-5990. Secure D2D communication in large-scale cognitive cellular networks:
[74] Vieira, J., Rusek, F., Edfors, O., Malkowsky, S., Liu, L., & Tufvesson, F. A wireless power transfer model. IEEE Transactions on
(2017). Reciprocity calibration for massive MIMO: Proposal, modeling, Communications, 64(1), 329-342.
and validation. IEEE Transactions on Wireless Communications, 16(5), [97] Zhang, A., Chen, J., Hu, R. Q., & Qian, Y. (2015). SeDS: Secure data
3042-3056. sharing strategy for D2D communication in LTE-Advanced
[75] Neupane, K., & Haddad, R. J. (2019). Secrecy sum-rate analysis of networks. IEEE Transactions on Vehicular Technology, 65(4), 2659-
massive MIMO systems under dual-threat attacks using normalization 2672.
methods. Digital Communications and Networks, 5(4), 237-244. [98] Kamel, M., Hamouda, W., & Youssef, A. (2016). Ultra-dense networks:
[76] Tehrani, R. H., Vahid, S., Triantafyllopoulou, D., Lee, H., & Moessner, A survey. IEEE Communications Surveys & Tutorials, 18(4), 2522-2545.
K. (2016). Licensed spectrum sharing schemes for mobile operators: A [99] Chopra, G., Jha, R. K., & Jain, S. (2017). A survey on ultra-dense
survey and outlook. IEEE Communications Surveys & Tutorials, 18(4), network and emerging technologies: Security challenges and possible
2591-2623. solutions. Journal of Network and Computer Applications, 95, 54-78.
[77] Voicu, A. M., Simić, L., & Petrova, M. (2018). Survey of spectrum [100] Chopra, G., Jha, R. K., & Jain, S. (2018). TPA: Prediction of Spoofing
sharing for inter-technology coexistence. IEEE Communications Surveys Attack Using Thermal Pattern Analysis in Ultra Dense Network for High
& Tutorials, 21(2), 1112-1144. Speed Handover Scenario. IEEE Access, 6, 66268-66284.
[78] Park, J. M., Reed, J. H., Beex, A. A., Clancy, T. C., Kumar, V., & Bahrak, [101] Chopra, G., Jha, R. K., & Jain, S. (2019). RBA: Region Based Algorithm
B. (2014). Security and enforcement in spectrum sharing. Proceedings of for secure harvesting in Ultra Dense Network. Journal of Network and
the IEEE, 102(3), 270-281. Computer Applications, 125, 179-189.
[79] Kour, H., Jha, R. K., & Jain, S. (2018). A comprehensive survey on [102] Sun, L., & Du, Q. (2017). Physical layer security with its applications in
spectrum sharing: Architecture, energy efficiency, and security 5G networks: A review. IEEE China Communications, 14(12), 1-14.
issues. Journal of Network and Computer Applications, 103, 29-57. [103] Chen, Z., Chen, S., Xu, H., & Hu, B. (2018). A security authentication
[80] Zou, Y. (2016). Physical-layer security for spectrum sharing scheme of 5G ultra-dense network based on block chain. IEEE Access, 6,
systems. IEEE Transactions on Wireless Communications, 16(2), 1319- 55372-55379.
1329. [104] Wang, L., Wong, K. K., Jin, S., Zheng, G., & Heath, R. W. (2018). A
[81] Nguyen, N. P., Kundu, C., Ngo, H. Q., Duong, T. Q., & Canberk, B. new look at physical layer security, caching, and wireless energy
(2016). Secure full-duplex small-cell networks in a spectrum sharing harvesting for heterogeneous ultra-dense networks. IEEE
environment. IEEE Access, 4, 3087-3099. Communications Magazine, 56(6), 49-55.
25
[105] Wang, Y., Miao, Z., & Jiao, L. (2016). Safeguarding the ultra-dense [128] Li, C., & Palanisamy, B. (2018). Privacy in Internet of Things: from
networks with the aid of physical layer security: A review and a case Principles to Technologies. IEEE Internet of Things Journal, 6(1), 488-
study. IEEE Access, 4, 9082-9092 505.
[106] Mukherjee, A., Fakoorian, S. A. A., Huang, J., & Swindlehurst, A. L. [129] Chatterjee, B., Das, D., Maity, S., & Sen, S. (2018). RF-PUF: Enhancing
(2014). Principles of physical layer security in multiuser wireless IoT security through authentication of wireless nodes using in-situ
networks: A survey. IEEE Communications Surveys & Tutorials, 16(3), machine learning. IEEE Internet of Things Journal, 6(1), 388-398.
1550-1573. [130] Xiao, L., Wan, X., & Han, Z. (2018). PHY-layer authentication with
[107] Kamel, M., Hamouda, W., & Youssef, A. (2017). Physical layer security multiple landmarks with reduced overhead. IEEE Transactions on
in ultra-dense networks. IEEE Wireless Communications Letters, 6(5), Wireless Communications, 17(3), 1676-1687
690-693. [131] Kenneally, E. (2018). Privacy and Security. IEEE Internet of Things
[108] Chen, S., Qin, F., Hu, B., Li, X., & Chen, Z. (2016). User-centric ultra- Magazine, 1(1), 8-10.
dense networks for 5G: challenges, methodologies, and directions. IEEE [132] Xiao, L., Wan, X., Lu, X., Zhang, Y., & Wu, D. (2018). IoT security
Wireless Communications, 23(2), 78-85. techniques based on machine learning: How do IoT devices use AI to
[109] Feng, M., & Mao, S. (2019). Dealing with Limited Backhaul Capacity in enhance security?. IEEE Signal Processing Magazine, 35(5), 41-49.
Millimeter-Wave Systems: A Deep Reinforcement Learning [133] Restuccia, F., D’Oro, S., & Melodia, T. (2018). Securing the internet of
Approach. IEEE Communications Magazine, 57(3), 50-55. things in the age of machine learning and software-defined
[110] Chen, S., Ma, R., Chen, H. H., Zhang, H., Meng, W., & Liu, J. (2017). networking. IEEE Internet of Things Journal, 5(6), 4829-4842.
Machine-to-machine communications in ultra-dense networks-A [134] He, D., Ye, R., Chan, S., Guizani, M., & Xu, Y. (2018). Privacy in the
survey. IEEE Communications Surveys & Tutorials, 19(3), 1478-1503. internet of things for smart healthcare. IEEE Communications
[111] Rupprecht, D., Dabrowski, A., Holz, T., Weippl, E., & Pöpper, C. (2018). Magazine, 56(4), 38-44.
On security research towards future mobile network generations. IEEE [135] Zhang, T., Cai, Y., Huang, Y., Duong, T. Q., & Yang, W. (2017). Secure
Communications Surveys & Tutorials, 20(3), 2518-2542. full-duplex spectrum-sharing wiretap networks with different antenna
[112] Koufos, K., & Dettmann, C. P. (2019). Boundaries as an Enhancement reception schemes. IEEE Transactions on Communications, 65(1), 335-
Technique for Physical Layer Security. IEEE Transactions on 346
Information Forensics and Security, 14(1), 61-74 [136] Gong, S., Xing, C., Fei, Z., & Ma, S. (2017). Millimeter-wave secrecy
[113] Xiao, L., Wan, X., Dai, C., Du, X., Chen, X., & Guizani, M. (2018). beamforming designs for two-way amplify-and-forward MIMO relaying
Security in mobile edge caching with reinforcement learning. IEEE networks. IEEE Transactions on Vehicular Technology, 66(3), 2059-
Wireless Communications, 25(3), 116-122. 2071
[114] Chopra, G., Jha, R. K., & Jain, S. (2018). Novel Beamforming Approach [137] Hu, J., Yang, N., & Cai, Y. (2018). Secure Downlink Transmission in the
for Secure Communication in UDN to Maximize Secrecy Rate and Internet of Things: How Many Antennas Are Needed?. IEEE Journal on
Fairness Security Assessment. IEEE Internet of Things Journal. 6(4), Selected Areas in Communications, 36(7), 1622-1634.
5935-5947. [138] Hu, L., Wen, H., Wu, B., Pan, F., Liao, R. F., Song, H., & Wang, X.
[115] Wang, H. M., Zheng, T. X., Yuan, J., Towsley, D., & Lee, M. H. (2016). (2017). Cooperative jamming for physical layer security enhancement in
Physical layer security in heterogeneous cellular networks. IEEE Internet of Things. IEEE Internet of Things Journal, 5(1), 219-228.
Transactions on Communications, 64(3), 1204-1219. [139] Mosenia, A., & Jha, N. K. (2016). A comprehensive study of security of
[116] Eltayeb, M. E., Choi, J., Al-Naffouri, T. Y., & Heath, R. W. (2017). internet-of-things. IEEE Transactions on Emerging Topics in
Enhancing secrecy with multiantenna transmission in millimeter wave Computing, 5(4), 586-602.
vehicular communication systems. IEEE Transactions on Vehicular [140] Lee, J. H., & Kim, H. (2017). Security and privacy challenges in the
Technology, 66(9), 8139-8151. internet of things [security and privacy matters]. IEEE Consumer
[117] Chu, Z., Zhu, Z., Johnston, M., & Le Goff, S. Y. (2015). Simultaneous Electronics Magazine, 6(3), 134-136.
wireless information power transfer for MISO secrecy channel. IEEE [141] Chou, R. A., & Bloch, M. R. (2016). Polar coding for the broadcast
Transactions on Vehicular Technology, 65(9), 6913-6925. channel with confidential messages: A random binning analogy. IEEE
[118] Sun, X., Yang, W., Cai, Y., Tao, L., Liu, Y., & Huang, Y. (2018). Secure Transactions on Information Theory, 62(5), 2410-2429
Transmissions in Wireless Information and Power Transfer Millimeter- [142] Yang, Y., Wu, L., Yin, G., Li, L., & Zhao, H. (2017). A survey on security
Wave Ultra-Dense Networks. IEEE Transactions on Information and privacy issues in Internet-of-Things. IEEE Internet of Things
Forensics and Security, 14(7), 1817-1829. Journal, 4(5), 1250-1258.
[119] Vuppala, S., & Abreu, G. (2015). Secrecy outage analysis in cellular [143] Yu, Y., Li, Y., Tian, J., & Liu, J. (2018). Blockchain-Based Solutions to
networks. IEEE Wireless Communications Letters, 4(2), 157-160. Security and Privacy Issues in the Internet of Things. IEEE Wireless
[120] Pérez, M. G., Celdrán, A. H., Ippoliti, F., Giardina, P. G., Bernini, G., Communications, 25(6), 12-18.
Alaez, R. M. & Carrozzo, G. (2017). Dynamic reconfiguration in 5G [144] Siegel, J. E., Kumar, S., & Sarma, S. E. (2017). The future internet of
mobile networks to proactively detect and mitigate botnets. IEEE Internet things: secure, efficient, and model-based. IEEE Internet of Things
Computing, 21(5), 28-36. Journal, 5(4), 2386-2398.
[121] Yang, W., Wang, S., Zheng, G., Yang, J., & Valli, C. (2019). A Privacy- [145] Abedi, M. R., Mokari, N., Saeedi, H., & Yanikomeroglu, H. (2017).
Preserving Lightweight Biometric System for Internet of Things Robust resource allocation to enhance physical layer security in systems
Security. IEEE Communications Magazine, 57(3), 84-89. with full-duplex receivers: Active adversary. IEEE Transactions on
[122] Sofi, I. B., & Gupta, A. (2018). A survey on energy efficient 5G green Wireless Communications, 16(2), 885-899.
network with a planned multi-tier architecture. Journal of Network and [146] Zou, Y., Zhu, J., Wang, X., & Leung, V. C. (2015). Improving physical-
Computer Applications, 118, 1-28. layer security in wireless communications using diversity
[123] Schinianakis, D. (2019). Lightweight Security for the Internet of Things: techniques. IEEE Network, 29(1), 42-48
A Soft Introduction to Physical Unclonable Functions. IEEE [147] Pechetti, S. V., Jindal, A., & Bose, R. (2018). Exploiting Mapping
Potentials, 38(2), 21-28. Diversity for Enhancing Security at Physical Layer in the Internet of
[124] Yu, H., Kim, T., & Jafarkhani, H. (2018). Wireless secure Things. IEEE Internet of Things Journal, 6(1), 532-544.
communication with beamforming and jamming in time-varying wiretap [148] Sofi, I. B., Gupta, A., & Jha, R. K. (2019). Power and energy optimization
channels. IEEE Transactions on Information Forensics and with reduced complexity in different deployment scenarios of massive
Security, 13(8), 2087-2100. MIMO network. International Journal of Communication
[125] Paul, L. Y., Verma, G., & Sadler, B. M. (2015). Wireless physical layer Systems, 32(6), e3907.
authentication via fingerprint embedding. IEEE Communications [149] Mehmood, Y., Haider, N., Imran, M., Timm-Giel, A., & Guizani, M.
Magazine, 53(6), 48-53 (2017). M2M communications in 5G: state-of-the-art architecture, recent
[126] Ssettumba, T., El-Malek, A. H. A., Elsabrouty, M., & Abo-Zahhad, M. advances, and research challenges. IEEE Communications
(2019). Physical Layer Security Enhancement for Internet of Things in Magazine, 55(9), 194-201
the Presence of Co-channel Interference and Multiple [150] Mishra, A. K., Tripathy, A. K., Puthal, D., & Yang, L. T. (2018).
Eavesdroppers. IEEE Internet of Things Journal, 6(4), 6441-6452. Analytical model for sybil attack phases in internet of things. IEEE
[127] Ferdowsi, A., & Saad, W. (2018). Deep learning for signal authentication Internet of Things Journal, 6(1), 379-387.
and security in massive internet-of-things systems. IEEE Transactions on
Communications, 67(2), 1371-1387.
26
[151] Wang, J., Tang, Q., Yang, C., Schober, R., & Li, J. (2016). Security Communication was accepted by ITU (International Telecommunication
enhancement via device-to-device communication in cellular Union) in 2010. He has received young scientist author award by ITU in Dec
networks. IEEE Signal Processing Letters, 23(11), 1622-1626. 2010. He has received an APAN fellowship in 2011, 2012, 2017, and 2018 and
[152] Ma, C., Liu, J., Tian, X., Yu, H., Cui, Y., & Wang, X. (2014). Interference student travel grant from COMSNET 2012. He is a senior member of IEEE,
exploitation in D2D-enabled cellular networks: A secrecy GISFI and SIAM, International Association of Engineers (IAENG), and ACCS
perspective. IEEE Transactions on Communications, 63(1), 229-242. (Advanced Computing and Communication Society). He is also member of
[153] Shayea, I., Rahman, T. A., Azmi, M. H., & Islam, M. R. (2018). Real
ACM and CSI, many patents, and more than 1765 Citations in his credit.
Measurement Study for Rain Rate and Rain Attenuation Conducted Over
26 GHz Microwave 5G Link System in Malaysia. IEEE Access, 6, 19044-
19064. Dr. Manish Sabraj is currently an Assistant
[154] Banday, Y., Rather, G. M., & Begh, G. R. (2018). Effect of atmospheric Professor in the School of Electronics and
absorption on millimetre wave frequencies for 5G cellular networks. IET Communication Engineering, Shri Shri Mata
Communications, 13(3), 265-270. Vaishno Devi University, Katra, Jammu and
[155] Nandi, D., & Maitra, A. (2018). Study of rain attenuation effects for 5G Kashmir, India. He has received his [Link] degree in
Mm-wave cellular communication in tropical location. IET Microwaves, Electronics and Communication from Govt. College
Antennas & Propagation, 12(9), 1504-1507. of Engineering and Technology Jammu and Kashmir,
[156] Zhang, Y. P., Wang, P., & Goldsmith, A. (2015). Rainfall effect on the India, in 2001 and [Link] degree in Electronics and
performance of millimeter-wave MIMO systems. IEEE Transactions on Communication from IIT Guwahati, Guwahati,
wireless communications, 14(9), 4857-4866. India, in 2003. He has completed his Ph.D. degree from Shri Mata Vaishno
[157] Dong, X. Y., Chen, H. Y., & Guo, D. H. (2011). Microwave and Devi University, Jammu and Kashmir, India in 2013. His research interest
millimeter-wave attenuation in sand and dust storms. IEEE Antennas and includes Digital Signal Processing, Wireless Sensor Networks, and Spectrum
wireless propagation letters, 10, 469-471. estimation and analysis.
[158] Chiou, M. M., & Kiang, J. F. (2016). Attenuation of millimeter-wave in
a sand and dust storm. IEEE Geoscience and remote sensing
letters, 13(8), 1094-1098.
[159] Chen, H. Y., & Ku, C. C. (2012). Calculation of wave attenuation in sand
and dust storms by the FDTD and turning bands methods at 10–100
GHz. IEEE Transactions on Antennas and Propagation, 60(6), 2951-
2960.
[160] Dong, Q. F., Li, Y. L., Xu, J. D., Zhang, H., & Wang, M. J. (2012). Effect
of sand and dust storms on microwave propagation. IEEE transactions
on antennas and propagation, 61(2), 910-916.
[161] Esmail, M. A., Fathallah, H., & Alouini, M. S. (2016). An experimental
study of FSO link performance in desert environment. IEEE
Communications Letters, 20(9), 1888-1891.
[162] Chiou, M. M., & Kiang, J. F. (2016). PWE-Based Radar Equation to
Predict Backscattering of Millimeter-Wave in a Sand and Dust
Storm. IEEE Transactions on Antennas and Propagation, 65(2), 785-
793.
[163] Dong, Q. F., Guo, L. X., Li, Y., Zhang, H., & Wang, M. J. (2016).
Weathering Sand and Dust Storms: Particle shapes, storm height, and
elevation angle sensitivity for microwave propagation in earth-satellite
links. IEEE Antennas and Propagation Magazine, 59(1), 58-65

Misbah Shafi received the B.E degree in electronics

and communication engineering from Islamic
University of science and technology, Jammu and
Kashmir, India, in 2016 and [Link] degree in
electronics and communication from Lovely
Professional University, Punjab, India, in 2018. She is
currently pursuing a Ph.D. degree in electronics and
communication engineering at Shri Mata Vaishno Devi
University, Jammu and Kashmir. Her research interest
includes network security in emerging technologies of 5G wireless
communication networks, cryptography, optical fibre communication
networks. Currently she is doing her research on security issues of 5G NR. She
is working on MATLAB for wireless communication.

Dr. Rakesh K Jha (S’10, M’13, SM 2015) is

currently an Associate Professor in the School of
Electronics and Communication Engineering, Shri
Mata Vaishno Devi University, Katra, Jammu and
Kashmir, India. He is carrying out his research in
wireless communication, power optimizations,
wireless security issues, and optical
communications. He has done [Link] in Electronics
and Communication Engineering from Bhopal,
India and M. Tech from NIT Jalandhar, India. He received his Ph.D. degree
from NIT Surat, India, in 2013. He has published more than 41 Science Citation
Index Journals Papers, including many IEEE Transactions, IEEE Journal, and
more than 25 International Conference papers. His area of interest is Wireless
communication, Optical Fiber Communication, Computer networks, and
Security issues. Dr. Jha’s one concept related to router of Wireless

27