I. INTRODUCTION

AI and big data techniques make it possible to build complex applications. 6G adds new technical standards and performance indicators to meet the needs of these applications, which need fast data transfer, minimum delay, and a reliable network, as illustrated in Table I [9]. Given the usage of terahertz and optical wireless bands, the existing 5G networks are capable of delivering up to 1-10 Tbps of data. AI can enhance network management, increasing area traffic capacity to over 1Gb/s/$m^2$, spectrum efficiency by 3.5 times, and energy efficiency by up to 10 times compared to 5G technology [10]. Heterogeneous networks, also known as Het-nets, are expected to significantly increase connection density by ten to one hundred, enabling a wide range of communication scenarios and utilizing large bandwidths in high-frequency bands. Mobility will be supported at speeds exceeding $1000km/h$ through ultra-high-speed trains, UAVs, and satellites. Performance criteria such as cost effectiveness, safety capabilities, coverage, and intelligence are also crucial for 6G network evaluation [11].

Table 1. Comparison of Key Performance Requirements.

Table 1. Comparison of Key Performance Requirements.

Performance attributes	5G	B5G	6G
Data rate	1Gb/s	100Gb/s	1Tb/s
E2E delay	5ms	1ms	<1ms
Radio-only delay	100ns	100ns	10ns
Processing delay	100ns	50ns	10ns
E2E reliability requirement	99.999%	99.9999%	99.99999%

Table 2. Existing Literature Works.

Table 2. Existing Literature Works.

Ref. no	Year	Research title	Technology used	Algorithm used	Performance Indicators	Inference
[13]	2018	Hap-SliceR-A radio resource slicing framework for 5G.	Virtualised radio resource slicing strategy.	Heuristic algorithm for resource allocation with RL.	Slice Reservation Index(SRI) Round Robin(RR) and Best channel quality indicator (BCQI) for UL/DL data rates.	A feasible 5G network solution that is implemented in a standard cloud computing infrastructure near the RAN edge.
[14]	2019	Intelligent resource scheduling strategy (iRSS) for 5G RAN slicing.	The framework is a combination of Deep Learning and Reinforcement Learning.	LSTM-RNN is used as a DL algorithm and A3C as an RL algorithm.	DL for large-scale resource allocation. RL for online resource scheduling to manage small-time-scale network dynamics.	Resource scheduling for improving multiplexing gain, meeting RAN slicing requirements, and addressing issues like inaccurate predictions and unexpected state changes.
[15]	2020	ML for resource management in cellular IoT networks	ML-DL in Hetnets, MIMO, and D2D communications.	ML and DL-based techniques for resource management.	Cellular, Low power, cognitive and mobile IoT networks. Optimization, heuristic and game theoretic approaches.	It covers resource management techniques such as scheduling, duty cycling, clustering, data aggregation, traffic classification, prediction, power allocation, and interference management.
[16]	2021	Intelligent resource slicing for eMBB and URLLC for 5G/B5G	Resource slicing in dynamic multiplexing scenarios of URLLC and eMBB services of 5G.	DRRA-PGACL is utilized for efficient resource allocation in eMBB and URLLC scenarios.	DRRA in eMBB resource allocation for low computational complexity. URLLC scheduling employs DRL as PDACL to maintain latency and reliability.	The eMBB focuses on maintaining a high data rate while adhering to URLLC reliability constraints, with the DRL approach facilitating a good convergence rate.
[17]	2021	Multi-agent reinforcement learning (MARL)	MARL in Markov, stochastic games and extensive forms of games.	Dynamic programming, game theory, optimization theory and statistics.	MARL is used in applications like autonomous driving and robotics.	MARL frameworks can be centralized, decentralized with networked agents, or fully decentralized.
[18]	2021	ML-based physical layer security	ML is used to reduce the increasing complexity of wireless networks.	ML applied to wireless physical layer security (WPLS)	Relay node selection, antenna selection and authentication with ML	ML is utilized in various WPLS networks such as IoT, D2D, cognitive radio, and UAV.
[19]	2022	PSARE: Online participation selection scheme for area coverage ratio	Area coverage ratio in Mobile Crowdsensing using RL	Q-Learning for online participation selection	Coverage degree, coverage quality and time complexity.	The framework utilizes RL with a Q-learning-based online participation selection scheme to enhance coverage ratio and degree.
[20]	2022	Optimised packet forwarding in multi-band relay networks	Enhancing network coverage using relay packet forwarding.	Q-learning and Markov decision process (MDP)	Packet transmission time, buffer overflow and effective throughput	The framework utilizes multiple relay nodes as MDP to create an efficient channel allocation algorithm.
[21]	2023	Adversarial attacks and defences in ML network	Counter-attack detection and network robustness enhancement using ML	ML and deep neural network (DNN)	New types of attacks include search-based, decision-based, physical world attacks, and hierarchical classification attacks.	The latest defence methods balance training costs with performance, maintain clean accuracy, overcome gradient masking, and ensure method transferability.
This paper	2024	Network slicing enabled security concerns in 6G	Intrusion detection system (IDS) using game theory	Game theory integrated with Network slicing and ML	IDS implementation using game theory in path loss models for different user scenarios. Network slicing and ML for intrusion detection in large and complex networks.	The IoT significantly impacts data security, necessitating updates to B5G and 6G wireless networks and changes in communication systems for robust authentication and trust mechanisms.

II. Security Concerns in 6G Network

III. Security Aspects to be considered

Security is essential when using mobile IoT devices (like UAVs and bot devices) in highly interactive environments, such as the Tactile interfacing environment. Some of the important security aspects that must be taken into consideration are listed below and are effectively illustrated in Figure 2 using a layerwise IoT interfacing architecture. The security of the IoT perception layer is susceptible to several types of security attacks, such as node capture, code injection, side channel assaults, eavesdropping, interference, and booting. These security vulnerabilities provide substantial risks to the overall security of the IoT network. The networking layer is vulnerable to several forms of attacks, including routing assaults such as sinkhole and wormhole attacks, attacks on data transfer, Denial of Service (DoS) and Distributed Denial of Service (DDoS) attacks and Phishing site attacks IoT middleware provides powerful computation and storage by abstracting the network and application layers. It consists of brokers, persistent data storage, queuing systems, and machine learning. Middleware security issues fall mostly on database and cloud security [37]. Despite its usefulness in providing reliable IoT applications, middleware is susceptible to attacks that can control the entire application, which include flooding attacks, cloud malware attacks, man-in-the-middle attacks, SQL injection attacks, and signature wrapping attacks. While, the application layer encompasses various types of threats, such as data theft and access control attacks, service interruption attacks, sniffing attacks, and reprogramming attacks.

Table 3. Machine Learning Implementations in Real-Time Wireless IoT Networks.

Table 3. Machine Learning Implementations in Real-Time Wireless IoT Networks.

Source	Machine Learning Algorithms	Network Optimisation Entity	Use cases in IoT network
[32]	Classification	Data Prediction, Time Prediction, Labelling data detection	Smart Traffic, Smart Cities
[33]	Clustering	Prediction and data segregation, Passenger data pattern	Smart Traffic, Smart Health
[34]	Anomaly detection	Finding anomalies in power datasets, Fault detection	Smart Traffic, Smart Environment
[35]	Support Vector Machine(SVM)	Data Forecasting	Smart Weather prediction
[36]	Linear Regression	Real-time prediction and data reduction	Economics, Market analysis, Energy usage

IV. Proposed Network Scenario

A. Proposed System

Figure 3 presents an intrusion detection system divided into physical and network domains, comprising the personal area domain, immediate area domain and a wide area domain. The personal area domain here involves direct communication between user terminals and peripheral devices, while the immediate area domain covers communication between devices within the user environment. The wide area domain includes communication between devices via network operator infrastructure, covering a range of a few meters to several tens of meters. The network is sliced into different domains using network slicing, with each layer having its own independent yet interconnected network. All of these layers are susceptible to security attacks. Therefore, the suggested system model proposes to use game theory to address intrusion detection in a low-complexity host network. However, to address security breaches in a highly complex network domain, network-slicing and game theory are proposed. Moreover, to train the concerned network against similar breaches in the future, ML approaches must be corroborated using the network complexity to prevent future breaches.

Table 4. Channel Parameters in 6G Wireless Network.

Table 4. Channel Parameters in 6G Wireless Network.

S.no
	$\sum {\mathcal{B}}_{\begin{array}{c}(More bandwidth \\ access)\end{array}}$	Millimetre wave (mm-wave)[54]	Terahertz (THz) network[55]	Optical wireless communication[56]	Cognitive Radio Network (CRN)[57]	Full Duplex (FD) and multi-standard systems[58]
	$\sum {\mathcal{P}}_{\begin{array}{c}(More Power \\ and efficiency)\end{array}}$	Small cells and cell-free networks [59]	Green communication[60]	Advanced modulation and Network coding[61]	Polar codes, Turbo codes[62]	Low-density parity check code (LDPC)
	${\mathcal{N}}_{\begin{array}{c}noise \\ and filtering\end{array}}+{\mathbb{I}}_{\begin{array}{c}interference\\ \end{array}}$	Interference randomisation[63]	Interference coordination [64]		Interference cancellation[65]	Interference alignment[66]
	Applications	Industrial IoT[67]	Augmented, Virtual and Mixed reality (AR/VR/MR)[68]		Wearable displays	Robots and Drones

B. System Model

We consider a wireless channel with $\mathbb{N}$ deployed nodes, where the majority of connected nodes are legitimate users. The reason for such an assumption is to propose a model to check the effect of a few eavesdroppers or attackers on the majority of the legitimate users that are connected to the same base station. Whereas some of the nodes in the network might be either compromised nodes or intruders trying to access the host network. Let there be ‘$i$’ number of legitimate users and ‘$\mathbb{N}-i=j$’ number of attackers accessing the host network, with $P_n$ being the transmission power by the base station (BS) to each connected device, either user or attacker. Here, as per the proposed scenario, both the legitimate user and the attacker compete with each other to try to gain access to the same host network.

We consider $(x_{BS},y_{BS})$ to be coordinates of the connected BS. Let us represent $\left(x_U^i,y_U^i\right)$ and $(x_A^j,y_A^j)$ as the coordinates of the corresponding $i^{th}$ user and $j^{th}$ attacker located at a certain distance $\underset{BS-i}{d}$ and $\underset{BS-j}{d}$ from the BS serving the host network. The distance between the BS and the user/attacker is computed by

$$d_{BS,i}=\sqrt{{(x_{BS}-x_U^i)}^2+{(y_{BS}-y_U^i)}^2}$$

(1)

and

$$d_{BS,j}=\sqrt{{(x_{BS}-x_A^j)}^2+{(y_{BS}-y_A^j)}^2}$$

(2)

However, it is assumed that the connected BSs are well equipped with omnidirectional antennas, with known CSI at both the transmitter and receiver sides. Here the connected user or attacker is denoted by $U$ and $A$ respectively in a Rayleigh Flat fading channel. Furthermore, a constant noise power is assumed at every instant for the $i^{th}$ user and $j^{th}$ attacker. Based on the given assumptions, the concerned BS transmits the signal ${\mathit{x}}_i$ to the $i^{th}$ user, and the signal received at the receiving end is given by

$$y_i={\mathit{h}}_i {\mathit{x}}_i+n_i$$

(3)

here $n_i~\mathsf{{\rm N}}(0,{\sigma }_m^2)$, denotes the complex additive white Gaussian noise (AWGN), having zero mean and variance ${\sigma }_m^2$ at the receiving node, and ${\mathit{h}}_i$ is the Rayleigh fading coefficient associated with the channel between the host BS and the valid user. $\alpha$ denotes the path loss exponent of the main channel. Likewise, (3) can be further written as

$$y_i=\sqrt{P_n{d_{BS,i}}^{-\alpha }}{\mathit{h}}_i{\mathit{x}}_i+n_i$$

(4)

The attacker on the other hand pings the host network and tries to eavesdrop on the signal between host BS and the users and receives the following signal:

$$z_j={\mathit{h}}_j {\mathit{x}}_j+n_j$$

(5)

where ${\mathit{h}}_j$ is the channel coefficient between $j^{th}$ attacker node and host BS, and $n_j$ denotes the complex Gaussian noise with zero mean and ${\sigma }_m^2$ variance.

Following this the received signal at $i^{th}$ user is further expressed as a combination of actual signal components, AWGN noise and interference from neighbouring nodes which may either be compromised nodes or intruders.

$${\left|y_i\right|}^2=P_n{d_{BS,i}}^{-\alpha }{\left|{\mathit{h}}_i\right|}^2{\left|{\mathit{x}}_i\right|}^2+{\sigma }_m^2+\sum _{j=1,j\ne i}^n{P}_j{d_{BS,j}}^{-\alpha }{\left|{\mathit{h}}_j\right|}^2{\left|{\mathit{x}}_j\right|}^2$$

(6)

here $P_j$ represents the transmission power allocated to the interfering $j^{th}$ node, which may be either an attacker or a failed node.

The proposed scenario involves the attacker preparing to attack the host network by analyzing the SINR values in the interfering network regions. Therefore the signal-to-noise plus interference ratio (SINR), is crucial in evaluating the intrusion level in the received signal, as shown

$$SINR=\frac{Signal power\left(P_s\right)}{\begin{array}{c}Noise power\left(\mathcal{N}\right)+Interference between nodes\left(\mathbb{I}\right)+Interference due to\\ intrusion on host network\left(\delta \right)\end{array}}$$

(7)

The overall SINR comprises the ratio of the transmission power by the host network, which is denoted as signal power $P_s$, to the combined noise power $\mathcal{N}$ of all the connected entities in the host network including both legitimate users and intruders, with ‘$\mathbb{I}$’ as the nodal interference between connected entities in the host network. However, due to an intrusion attack, the SINR taken into consideration also comprises the interference occurring due to an intrusion attack on the concerned host network. The received instantaneous SINR at the $i^{th}$ user can be mathematically expressed as:

$${{\gamma }_i}^t=\frac{P_s }{{\sigma }_m^2+\sum I}$$

(8)

where $P_s$=$P_n{d_{BS,i}}^{-\alpha }{\left|{\mathit{h}}_i\right|}^2{\left|{\mathit{x}}_i\right|}^2$, thus (8) can be further rewritten as

$${{\gamma }_i}^t=\frac{P_n{d_{BS,i}}^{-\alpha }{\left|{\mathit{h}}_i\right|}^2{\left|{\mathit{x}}_i\right|}^2}{{\sigma }_m^2+\sum _{j=1, j\ne i}^n{P}_j{d_{BS,j}}^{-\alpha }{\left|{\mathit{h}}_j\right|}^2{\left|{\mathit{x}}_j\right|}^2}$$

(9)

Here ${\sigma }_m^2$ is the noise power and all channels are assumed to be available including those for the attacker. Therefore the average channel capacity of the $i^{th}$ user can be expressed by

$${Ć}_i={\log }_2(1+{{\gamma }_i}^t)\mathrm{b}\mathrm{p}\mathrm{s}/\mathrm{H}\mathrm{z}$$

(10)

Table 5. List of Symbols used.

Table 5. List of Symbols used.

Symbol	Description
$\underset{BS,i}{d}$	Base station-User link
$\underset{BS,j}{d}$	Base station-Attacker link
$P_n$	Reference channel power allocated by the BS to each connected user/attacker.
${\mathit{h}}_i$, ${\mathit{h}}_j$	Channel coefficient vector from BS to the $i^{th}$ user and $j^{th}$ attacker respectively.
${\gamma }_i$	SINR at receiver signal.
$\alpha$	Pathloss exponent of the main channel.
${\sigma }_m^2$	Noise power
$B_n^t$	Total bandwidth allocated to $n$ users at instant $t$
${\mathcal{T}}_i^t$	Average throughput of $i^{th}$user accessing BS at instant $t$
${Ć}_{sec}$	Secrecy capacity
${Ć}_i$	Channel capacity before intrusion attack
${Ć}_j$	Channel capacity after intrusion attack
$\delta$	Signal interference due to intrusion
$\mathcal{N}+\mathbb{I}$	Overall noise and interference in the established channel
$\underset{User}{Ű}$	A valid or legitimate user
$\underset{Attacker}{Ű}$	Attacker or intruder gaining host bandwidth access
${\mho }_{user }$	Set of strategies adopted by the valid user.
${\mho }_{attacker }$	Set of strategies by attacker accessing the host bandwidth
${\varphi }_i^1$	Strategy played by the user
${\varphi }_j^2$	Strategy played by the attacker element
(${\mathcal{a}\mathcal{\text{'}}}_{ji} , {\mathcal{a}}_{ji})$	Rewards earned by user and attacker during the game played.

As per the required scenario, the total bandwidth $B_{Total}$ is allocated to each connected node in the considered network. Initially, equal bandwidth access is granted to both the users and attackers in a suggested host network, where bandwidth obtained by each connected device at $t$ time slot is estimated by:

$$B_n^t=B_{total}f\left(n^t\right)$$

(11)

The total bandwidth accessed is, hence a function of the total $n$ number of connected entities (users and attackers) at instant $t$, as per the network deployment.

In due course, both the user and the attacker try to gain access to the majority of the host bandwidth, each of them corroborating their set of strategies while playing a game. The game theory concept is therefore implemented in the proposed scenario for network monitoring, detecting intruder elements, and removing malicious attacker devices from the network, thereby preventing their access to the host network [69].

The primary goal is to safeguard host network data from unauthorized access and prevent future incidents by analyzing channel capacity in real-time scenarios, as illustrated in Table IV. Table V provides a comprehensive description of the notations utilized in the system model.

V. Flow Process and Implementation

A. Game Theory Approach

Game theory is a field that studies strategic interactions among rational players, focusing on optimizing their benefits. It offers efficient and robust distributed algorithms, making it useful in wireless networks for modelling, analyzing, and designing distributed schemes. Hence in this aspect, game theory is utilized to study and forecast the future behaviour of the complicated attacks in present and future networks. Furthermore, it aids in the characterization of the intricate nature of these attacks and facilitates the prediction of their future patterns, hence enhancing the ability to assess and prepare for effective responses.

The wireless network components commonly seen in contemporary settings encompass vehicle ad hoc networks (VANETS)[72], drone networks, and high-speed trains (HST)[73]. These components consist of a combination of mobile and stationary nodes, each equipped with a range of communication technologies and infrastructure. Intrusion detection systems (IDS) are used to safeguard networks from both internal and external cyber threats. The IDS employs anomaly detection policies to identify malicious activities perpetrated by attackers. However, these policies are vulnerable to many forms of attacks, such as location spoofing. The accurate prediction and detection of zero-day threats, which encompass one or two intricate attacks, can be achieved through the utilization of Game Theory and Machine Learning (ML) methodologies [74], some of which are illustrated in Table III.

A game consists of three elements: the set of players $N$, the strategy space for each player ${\varphi }_i$, and the payoff ${\mathcal{U}}_i$, which is the reward players receive at the end of the game based on other players’ actions. The wireless communication network involves players like UAVs, end devices, ground users, or base stations, where the strategies may include beaconing, offloading, channel assignment, and intruder evasion. While the payoffs comprise throughput, SINR, delays, and node coverage based on the real applications. A game is static if all players make decisions simultaneously without knowing each other’s strategies, while it is dynamic if players make decisions sequentially or repeatedly[69].

Games can be categorized into complete and incomplete information games based on their known information structure. The information games are perfect or imperfect based on whether all players know each other’s historical actions. Cooperative games are those where players cooperate to optimize a common goal, while non-cooperative games are those where players don’t cooperate. Furthermore, the use of a distributed architecture that incorporates prediction and detection algorithms allows for the optimisation of processes by using both mobile nodes and centralised infrastructure to enhance efficiency and effectiveness.

Game theory is commonly recognised as a mathematical framework that can be employed to analyze the dynamics of conflict between a defender (IDS) and attackers, with the aim of determining the optimal strategy for the defence in accurately identifying a suspicious target as malevolent [75]. Therefore, to safeguard the integrity of the prevailing intelligent network against significant and perhaps fatal cyber threats, a number of cyber games have been developed. These games predominantly employ non-cooperative game theory as a means to address issues pertaining to prediction and decision-making[76]. In the context of such a particular game, participants strive to optimise their own payoffs while simultaneously decreasing the payoffs of their adversaries.

The primary aim of a non-cooperative game is to establish a state of Nash Equilibrium (NE) between the attacker and the Intrusion Detection System (IDS)[77]. In this condition, the attacker deploys malicious software targeting networking and IoT nodes, while the IDS employs its detection mechanism to safeguard legitimate users. This article therefore proposes utilising the intruder attack pattern through channel modelling to facilitate the analysis of intrusions into host networks by employing game theory and devising counter-protection strategies.[78]. The security aspects of using different ML techniques in an adversarial system between intruders and authorised devices and users are also explored further in this article.

The game theory approach, which focuses on the interaction between users and attackers, along with its PD matrix, is briefly discussed.

Let us denote$\underset{User}{Ű}$and$\underset{Attacker}{Ű}$as the valid user and the intruder in the network. These players in the game possess a set of strategies ${\mho }_{user=} \left\{{\varphi }_i^1\right|i=\mathrm{1,2},...,m\}$ and ${\mho }_{attacker=}\left\{{\varphi }_j^2\right|j=\mathrm{1,2},...,n\}$, respectively, where $m$ and $n$ are the maximum number of strategies that the user and attacker can carry out respectively. Table VI depicts a multidimensional payoff matrix or the prisoner’s dilemma (PD) matrix between these two players where ${\mathcal{a}}_{ji}$and ${\mathcal{a}\mathcal{\text{'}}}_{ji}$ are the rewards earned by$\underset{Attacker}{Ű}$and$\underset{User}{Ű}$respectively. The rewards earned by both players are defined by the strategies ${\varphi }_i^1$ and ${\varphi }_j^2$ performed in the given instances illustrated below:

Table 6. Multidimensional Payoff Reward Matrix.

Table 6. Multidimensional Payoff Reward Matrix.

• The rewards (${\mathcal{a}\mathcal{\text{'}}}_{ji} , {\mathcal{a}}_{ji})$ are earned by the players$\underset{User}{Ű}$and$\underset{Attacker}{Ű}$when IDS has not carried out the monitoring process yet the attacker has launched a malicious attack on the host network.
• The rewards (${\mathcal{a}\mathcal{\text{'}}}_{ji} , {\mathcal{a}}_{ji})$ are earned by the players $\underset{User}{Ű}$and$\underset{Attacker}{Ű}$when IDS monitors the network while the attacker launches an attack on the host network.
• The rewards (${\mathcal{a}\mathcal{\text{'}}}_{ji} , {\mathcal{a}}_{ji})$ are earned by the players$\underset{User}{Ű}$and$\underset{Attacker}{Ű}$when IDS runs a monitoring process and the attacker behaves like a normal node.
• The rewards (${\mathcal{a}\mathcal{\text{'}}}_{ji},{\mathcal{a}}_{ji})$ are earned by the players$\underset{User}{\mathrm{Ű}}$and$\underset{Attacker}{\mathrm{Ű}}$when IDS does not monitor the network and the attacker behaves like a normal node.

The matrix depicts the reward allocation for both the user and attacker in a game to gain access to the host bandwidth. The reward allocation under the applied game theory is done under various combinations of interactions between the user and the attacker and is categorised into four instances, as illustrated. The first instance takes place when the attacker has access to the host network and gains more rewards than the user when IDS fails to monitor the host network. The second instance illustrates the rewards earned by both attackers and users when IDS starts monitoring the host network immediately after the attacker launches its attack.

Furthermore, there can also be a scenario when an attacker behaves like a normal node after launching an attack, causing the host IDS to fail to detect its presence. Thus, rewards for this instance are allocated accordingly to the user and the attacker. The host network can thus, enhance its security by implementing effective countermeasures by thoroughly examining user and attacker situations. Pseudocode 1 effectively presents the attacking model, strategy setup, and protection mechanism for the proposed scenario using game theory, as illustrated in the flow process and shown in Figure 4.

The proposed intrusion detection scenario further aims to identify and combat security challenges in the IoT framework by focusing on perception, network, middleware, and application layers. It therefore suggests integrating network slicing, machine learning techniques, and a game theory approach to effectively address these challenges.

VI. Integration with Network Slicing and ML

VII. Simulation Results and Discussion

The decision tree in the proposed system model checks if a given data point is an intruder, compromised node, or valid user. It then moves to the next decision node, checking the value for another feature, and ends at the leaf node, determining the node status value, and helping distinguish between intruders and valid users, as illustrated in Figure 15. Here, Gini impurity, also known as the Gini index, is a measure of data disorder in decision trees used to split nodes during training, indicating the degree of data impurity. This is calculated for a binary classification problem involving two classes (valid user and intruder or compromised node), in our context, using the following formula:

Table 7. Simulation Parameters.

Table 7. Simulation Parameters.

Parameter	Value
Frequency(f)	28 GHz
Bandwidth	800 MHz
Noise figure	-106dBm
Maximum coverage $d$	500m
Maximum transmission power	30dBm
Minimum transmission power	18dBm
Pathloss exponent ($\alpha$)	2 to 4
Pathloss of a neighbourhood area network	$13.54+20{log}_{10}\left(f\right)+39.08{log}_{10}\left(d\right) dB$
Pathloss of the urban user area	$28.0+20{log}_{10}\left(f\right)+40{log}_{10}\left(d\right)-24.6792 dB$
Pathloss of indoor area network	$17.30+24.9{log}_{10}\left(f\right)+38.3{log}_{10}\left(d\right)dB$
Path loss of remote user area	$148.1+37.6{log}_{10}\left(d\right) dB$
Channel gain formula for a given path loss	${10}^{\left(\frac{-Path Loss}{10}\right)}$

$$Gini impurity=1-\sum _{i=1}^k{p}_i^2$$

(20)

where $k$ is the number of classes, and $p_i$ is the probability of belonging to class $i$. Gini impurity is used to measure class separation in decision trees. It determines the best split at each node during training, as lower Gini values indicate more homogeneous nodes. The Gini index, therefore, represents the impurity at each decision point in a decision tree plot.

Table IX presents the hyperparameters utilized in executing the aforementioned algorithms following our specific scenario. As discussed previously, the flow process for our proposed system illustrates the use of the host network database to check the presence of intruders. Therefore, we make use of some of the machine-learning classification techniques discussed previously to predict and detect the intruder's presence in the given host network, where the intruder is most likely to behave as a normal user to gain access to the host bandwidth.

Table 8. Comparative Analysis of Channel Parameters Before, During and After the Host Network Intrusion.

Table 8. Comparative Analysis of Channel Parameters Before, During and After the Host Network Intrusion.

User Scenarios	Parameters	Before Intrusion			During Intrusion			After Intrusion
		SNR	Channel Capacity (bits/s/Hz)	Throughput (bps)	SNR	Channel Capacity (bits/s/Hz)	Throughput (bps)	SNR	Channel Capacity (bits/s/Hz)	Throughput (bps)
	Urban user coverage (d=500m)	0.23	0.3075	246020000	0.029	0.042	34014000	0.09	0.1304	104300000
	Remote user coverage (d=500m)	0.51	0.59	479230000	0.06	0.09	72463000	0.20	0.26	215140000
	Urban micro: Neighbourhood area network (NaN)(d=250m)	0.27	0.35	280100000	0.034	0.049	39235000	0.10	0.149	119770000
	Indoor Scenario (d=50m)	35.92	5.20	4165400000	4.52	2.46	1972400000	14.30	3.93	3148700000

Table 9. Hyperparameters for Applied ML Algorithms.

Table 9. Hyperparameters for Applied ML Algorithms.

Description	Value
Frequency	28GHz
Bandwidth	800MHz
No. of nodes deployed ($N$)	10 to 50
No. of intruder nodes considered $\left(I\right)$	$1/5\left(N\right)$
No. of failed or compromised nodes $\left(F\right)$	$1/10\left(N\right)$
No. of valid users $\left(U\right)$	$N-(I+F)$

VIII. Conclusion

References

1. G. Gui, M. Liu, F. Tang, N. Kato, and F. Adachi, ‘6G: Opening New Horizons for Integration of Comfort, Security, and Intelligence’, IEEE Wirel. Commun., vol. 27, no. 5, pp. 126–132, 2020. [CrossRef]
2. M. Laroui, B. Nour, H. Moungla, M. A. Cherif, H. Afifi, and M. Guizani, ‘Edge and fog computing for IoT: A survey on current research activities & future directions’, Comput. Commun., vol. 180, pp. 210–231, 2021. [CrossRef]
3. Z. Zhang et al., ‘6G wireless networks: Vision, requirements, architecture, and key technologies’, IEEE Veh. Technol. Mag., vol. 14, no. 3, pp. 28–41, 2019.
4. K. David and H. Berndt, ‘6G vision and requirements: Is there any need for beyond 5G?’, IEEE Veh. Technol. Mag., vol. 13, no. 3, pp. 72–80, 2018.
5. C. D. Alwis et al., ‘Survey on 6G Frontiers: Trends, Applications, Requirements, Technologies and Future Research’, IEEE Open J. Commun. Soc., vol. 2, pp. 836–886, 2021. [CrossRef]
6. G. Geraci et al., ‘What Will the Future of UAV Cellular Communications Be? A Flight From 5G to 6G’, IEEE Commun. Surv. Tutor., vol. 24, no. 3, pp. 1304–1335, 2022. [CrossRef]
7. E. Omolara et al., ‘The internet of things security: A survey encompassing unexplored areas and new insights’, Comput. Secur., vol. 112, p. 102494, 2022. [CrossRef]
8. S. Dang, O. Amin, B. Shihada, and M.-S. Alouini, ‘What should 6G be?’, Nat. Electron., vol. 3, no. 1, pp. 20–29, Jan. 2020. [CrossRef]
9. F. Guo, F. R. Yu, H. Zhang, X. Li, H. Ji, and V. C. M. Leung, ‘Enabling Massive IoT Toward 6G: A Comprehensive Survey’, IEEE Internet Things J., vol. 8, no. 15, pp. 11891–11915, Aug. 2021. [CrossRef]
10. D. C. Nguyen et al., ‘6G Internet of Things: A Comprehensive Survey’, IEEE Internet Things J., vol. 9, no. 1, pp. 359–383, Jan. 2022. [CrossRef]
11. M. Vaezi et al., ‘Cellular, Wide-Area, and Non-Terrestrial IoT: A Survey on 5G Advances and the Road Toward 6G’, IEEE Commun. Surv. Tutor., vol. 24, no. 2, pp. 1117–1174, 2022. [CrossRef]
12. V. Hassija, V. Chamola, V. Saxena, D. Jain, P. Goyal, and B. Sikdar, ‘A survey on IoT security: application areas, security threats, and solution architectures’, IEEE Access, vol. 7, pp. 82721–82743, 2019.
13. Aijaz, ‘Hap-SliceR : A Radio Resource Slicing Framework for 5G Networks With Haptic Communications’, IEEE Syst. J., vol. 12, no. 3, pp. 2285–2296, Sep. 2018.
14. M. Yan, G. Feng, J. Zhou, Y. Sun, and Y.-C. Liang, ‘Intelligent Resource Scheduling for 5G Radio Access Network Slicing’, IEEE Trans Veh Technol, vol. 68, no. 8, pp. 7691–7703, 2019. [CrossRef]
15. F. Hussain, S. A. Hassan, R. Hussain, and E. Hossain, ‘Machine Learning for Resource Management in Cellular and IoT Networks: Potentials, Current Solutions, and Open Challenges’, IEEE Commun. Surv. Tutor., vol. 22, no. 2, pp. 1251–1275, 2020. [CrossRef]
16. M. Alsenwi, N. H. Tran, M. Bennis, S. R. Pandey, A. K. Bairagi, and C. S. Hong, ‘Intelligent Resource Slicing for eMBB and URLLC Coexistence in 5G and Beyond: A Deep Reinforcement Learning Based Approach’, IEEE Trans. Wirel. Commun., vol. 20, no. 7, pp. 4585–4600, 2021. [CrossRef]
17. K. Zhang, Z. Yang, and T. Başar, ‘Multi-Agent Reinforcement Learning: A Selective Overview of Theories and Algorithms’, in Handbook of Reinforcement Learning and Control, K. G. Vamvoudakis, Y. Wan, F. L. Lewis, and D. Cansever, Eds., Cham: Springer International Publishing, 2021, pp. 321–384. [CrossRef]
18. K. Kamboj, P. Jindal, and P. Verma, ‘Machine learning-based physical layer security: techniques, open challenges, and applications’, Wirel. Netw., vol. 27, no. 8, pp. 5351–5383, Nov. 2021. [CrossRef]
19. Y. Xu, Y. Wang, J. Ma, and Q. Jin, ‘PSARE: A RL-Based Online Participant Selection Scheme Incorporating Area Coverage Ratio and Degree in Mobile Crowdsensing’, IEEE Trans. Veh. Technol., vol. 71, no. 10, pp. 10923–10933, Oct. 2022. [CrossRef]
20. B. Mughal, Z. Md. Fadlullah, M. M. Fouda, and S. Ikki, ‘Optimizing Packet Forwarding Performance in Multiband Relay Networks via Customized Reinforcement Learning’, IEEE Open J. Commun. Soc., vol. 3, pp. 973–985, 2022. [CrossRef]
21. Y. Wang et al., ‘Adversarial Attacks and Defenses in Machine Learning-Empowered Communication Systems and Networks: A Contemporary Survey’, IEEE Commun. Surv. Tutor., pp. 1–1, 2023. [CrossRef]
22. M. Gupta, R. K. Jha, and S. Jain, ‘Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey’, IEEE Access, vol. 11, pp. 30639–30689, 2023. [CrossRef]
23. Gupta, R. K. Jha, P. Gandotra, and S. Jain, ‘Bandwidth Spoofing and Intrusion Detection System for Multistage 5G Wireless Communication Network’, IEEE Trans Veh Technol, 2018. [CrossRef]
24. P. Srividya, L. N. Devi, and A. N. Rao, ‘A trusted effective approach for forecasting the failure of data link and intrusion in wireless sensor networks’, Theor. Comput. Sci., vol. 941, pp. 1–13, 2023. [CrossRef]
25. V. Fanibhare, N. I. Sarkar, and A. Al-Anbuky, ‘A Survey of the Tactile Internet: Design Issues and Challenges, Applications, and Future Directions’, Electronics, vol. 10, no. 17, 2021. [CrossRef]
26. R. N. N and H. V. Nath, ‘Critical analysis of the layered and systematic approaches for understanding IoT security threats and challenges’, Comput. Electr. Eng., vol. 100, p. 107997, 2022. [CrossRef]
27. Budhiraja, S. Tyagi, S. Tanwar, N. Kumar, and J. J. Rodrigues, ‘Tactile internet for smart communities in 5g: An insight for noma-based solutions’, IEEE Trans. Ind. Inform., vol. 15, no. 5, pp. 3104–3112, 2019.
28. H. Xu, P. V. Klaine, O. Onireti, B. Cao, M. Imran, and L. Zhang, ‘Blockchain-enabled resource management and sharing for 6G communications’, Digit. Commun. Netw., vol. 6, no. 3, pp. 261–269, 2020.
29. Ometov, O. L. Molua, M. Komarov, and J. Nurmi, ‘A Survey of Security in Cloud, Edge, and Fog Computing’, Sensors, vol. 22, no. 3, 2022. [CrossRef]
30. K. K. Karmakar, V. Varadharajan, S. Nepal, and U. Tupakula, ‘SDN-Enabled Secure IoT Architecture’, IEEE Internet Things J., vol. 8, no. 8, pp. 6549–6564, 2020.
31. S. Shahzadi et al., ‘Machine Learning Empowered Security Management and Quality of Service Provision in SDN-NFV Environment’, CMC-Comput. Mater. Contin., vol. 66, no. 3, pp. 2723–2749, 2021.
32. S. M. Aldossari and K.-C. Chen, ‘Machine learning for wireless communication channel modeling: An overview’, Wirel. Pers. Commun., vol. 106, no. 1, pp. 41–70, 2019.
33. C. Jiang, H. Zhang, Y. Ren, Z. Han, K.-C. Chen, and L. Hanzo, ‘Machine Learning Paradigms for Next-Generation Wireless Networks’, IEEE Wirel. Commun., vol. 24, no. 2, pp. 98–105, Apr. 2017. [CrossRef]
34. Chen, Y. Fu, X. Zheng, and G. Lu, ‘An efficient network behavior anomaly detection using a hybrid DBN-LSTM network’, Comput. Secur., vol. 114, p. 102600, 2022. [CrossRef]
35. T. E. Bogale, X. Wang, and L. B. Le, ‘Machine Intelligence Techniques for Next-Generation Context-Aware Wireless Networks’, ITU Spec. Issue Impact Artif. Intell. AI Commun. Netw. Serv., 2018. [CrossRef]
36. D. Gündüz, P. de Kerret, N. D. Sidiropoulos, D. Gesbert, C. R. Murthy, and M. van der Schaar, ‘Machine Learning in the Air’, IEEE J. Sel. Areas Commun., vol. 37, no. 10, pp. 2184–2199, Oct. 2019. [CrossRef]
37. M. A. da Cruz, J. J. Rodrigues, A. K. Sangaiah, J. Al-Muhtadi, and V. Korotaev, ‘Performance evaluation of IoT middleware’, J. Netw. Comput. Appl., vol. 109, pp. 53–65, 2018.
38. B. Alzahrani, O. S. Oubbati, A. Barnawi, M. Atiquzzaman, and D. Alghazzawi, ‘UAV assistance paradigm: State-of-the-art in applications and challenges’, J. Netw. Comput. Appl., vol. 166, p. 102706, 2020. [CrossRef]
39. T. Taleb, S. Dutta, A. Ksentini, M. Iqbal, and H. Flinck, ‘Mobile Edge Computing Potential in Making Cities Smarter’, IEEE Commun. Mag., vol. 55, no. 3, pp. 38–43, Mar. 2017. [CrossRef]
40. Y. Xiao, G. Shi, Y. Li, W. Saad, and H. V. Poor, ‘Toward self-learning edge intelligence in 6G’, IEEE Commun. Mag., vol. 58, no. 12, pp. 34–40, 2020.
41. D. Fudenberg and J. Tirole, ‘Game theory, The MIT Press’, Camb. MA, vol. 86, 1991.
42. M. Samir, D. Ebrahimi, C. Assi, S. Sharafeddine, and A. Ghrayeb, ‘Leveraging Uavs for Coverage in Cell-Free Vehicular Networks: A Deep Reinforcement Learning Approach’, IEEE Trans. Mob. Comput., vol. 20, no. 09, pp. 2835–2847, Sep. 2021. [CrossRef]
43. S. Bi, R. Zhang, Z. Ding, and S. Cui, ‘Wireless communications in the era of big data’, IEEE Commun. Mag., vol. 53, no. 10, pp. 190–199, 2015.
44. Alnoman, S. K. Sharma, W. Ejaz, and A. Anpalagan, ‘Emerging edge computing technologies for distributed IoT systems’, IEEE Netw., vol. 33, no. 6, pp. 140–147, 2019.
45. Afolabi, A. Ksentini, M. Bagaa, T. Taleb, M. Corici, and A. Nakao, ‘Towards 5G network slicing over multiple-domains’, IEICE Trans. Commun., vol. 100, no. 11, pp. 1992–2006, 2017. [CrossRef]
46. L. U. Khan, I. Yaqoob, M. Imran, Z. Han, and C. S. Hong, ‘6G Wireless Systems: A Vision, Architectural Elements, and Future Directions’, IEEE Access, vol. 8, pp. 147029–147044, 2020. [CrossRef]
47. Al-Dulaimi, X. Wang, and I. Chih-Lin, 5G networks: fundamental requirements, enabling technologies, and operations management. John Wiley & Sons, 2018.
48. J. Liu et al., ‘Initial Access, Mobility, and User-Centric Multi-Beam Operation in 5G New Radio’, IEEE Commun Mag, vol. 56, no. 3, pp. 35–41, 2018. [CrossRef]
49. F. Al-Turjman, E. Ever, and H. Zahmatkesh, ‘Small Cells in the Forthcoming 5G/IoT: Traffic Modelling and Deployment Overview’, IEEE Commun. Surv. Tutor., vol. 21, no. 1, pp. 28–65, 2019. [CrossRef]
50. B. Zong, C. Fan, X. Wang, X. Duan, B. Wang, and J. Wang, ‘6G Technologies: Key Drivers, Core Requirements, System Architectures, and Enabling Technologies’, IEEE Veh. Technol. Mag., vol. 14, no. 3, pp. 18–27, 2019. [CrossRef]
51. J. Choi, ‘On Delay-Constrained Transmissions of a Finite Number of Short-Length Packets’, IEEE Access, vol. 9, pp. 1005–1015, 2021. [CrossRef]
52. M. Gupta, R. K. Jha, and M. Sabraj, ‘Touch-Interfacing Middleware Network Design in 6G’, in 2023 15th International Conference on COMmunication Systems & NETworkS (COMSNETS), Jan. 2023, pp. 345–349. [CrossRef]
53. S. Akbar, Y. Deng, A. Nallanathan, M. Elkashlan, and G. K. Karagiannidis, ‘Massive multiuser MIMO in heterogeneous cellular networks with full duplex small cells’, IEEE Trans. Commun., vol. 65, no. 11, pp. 4704–4719, 2017.
54. Y. Yuan et al., ‘A 3D Geometry-Based Reconfigurable Intelligent Surfaces-Assisted MmWave Channel Model for High-Speed Train Communications’, IEEE Trans. Veh. Technol., pp. 1–16, 2023. [CrossRef]
55. S. S. Omar, A. M. A. El-Haleem, I. I. Ibrahim, and A. M. Saleh, ‘Capacity Enhancement of Flying-IRS Assisted 6G THz Network Using Deep Reinforcement Learning’, IEEE Access, vol. 11, pp. 101616–101629, 2023. [CrossRef]
56. C. Su, J. Zhang, and Y. Ji, ‘Cyclic transmission window-based bandwidth allocation scheme for asynchronous time-sensitive industrial applications in TDM-PON’, J Opt Commun Netw, vol. 15, no. 11, pp. 820–829, Nov. 2023. [CrossRef]
57. Ali and W. Hamouda, ‘Advances on Spectrum Sensing for Cognitive Radio Networks: Theory and Applications’, IEEE Commun. Surv. Tutor., vol. 19, no. 2, pp. 1277–1304, 2017. [CrossRef]
58. T. Ji, M. Hua, C. Li, Y. Huang, and L. Yang, ‘Exploiting Intelligent Reflecting Surface for Enhancing Full-Duplex Wireless-Powered Communication Networks’, IEEE Trans. Commun., pp. 1–1, 2023. [CrossRef]
59. Mabrouk and R. Zayani, ‘Toward Energy-Efficient 6G Networks: Uplink Cell-Free Massive MIMO With NLD Cancellation Technique of Hardware Impairments’, IEEE Access, vol. 11, pp. 105314–105329, 2023. [CrossRef]
60. S. Mukherjee, C. Beard, and S. Song, ‘Transformers for Green Semantic Communication: Less Energy, More Semantics’. 2023.
61. C. Zheng, F.-C. Zheng, J. Luo, P. Zhu, X. You, and D. Feng, ‘Differential Modulation for Short Packet Transmission in URLLC’. 2023.
62. S. Weithoffer, G. Aousaji, J. Nadal, and C. A. Nour, ‘Iteration Overlap for Low-Latency Turbo Decoding’, in 2023 12th International Symposium on Topics in Coding (ISTC), Sep. 2023, pp. 1–5. [CrossRef]
63. W. Nie, M. Liu, J. Chen, W. Tan, and C. Li, ‘Spectrum and Energy Efficiency of Massive MIMO for Hybrid Architectures With Phase Shifter and Switches in IoT Networks’, IEEE Internet Things J., pp. 1–1, 2023. [CrossRef]
64. D. Malak, F. V. Mutlu, J. Zhang, and E. M. Yeh, ‘Joint Power Control and Caching for Transmission Delay Minimization in Wireless HetNets’, IEEEACM Trans. Netw., pp. 1–16, 2023. [CrossRef]
65. Al-Dweik, E. Alsusa, O. A. Dobre, and R. Hamila, ‘Multi-Symbol Rate NOMA for Improving Connectivity in 6G Communications Networks’, IEEE Commun. Mag., pp. 1–7, 2023. [CrossRef]
66. J. C. M. Filho, T. Abrao, E. Hossain, and A. Mezghani, ‘Reconfigurable Intelligent Surfaces-Enabled Intra-Cell Pilot Reuse in Massive MIMO Systems’. 2023.
67. H. Wang, Y. Bai, and X. Xie, ‘Deep Reinforcement Learning Based Resource Allocation in Delay-Tolerance-Aware 5G Industrial IoT Systems’, IEEE Trans. Commun., pp. 1–1, 2023. [CrossRef]
68. B. Guo, Y. Chen, P. Cheng, M. Ding, J. Hu, and L. Hanzo, ‘Pareto-Optimal Multi-Agent Cooperative Caching Relying on Multi-Policy Reinforcement Learning’, IEEE Internet Things J., pp. 1–1, 2023. [CrossRef]
69. R. Carmona and F. Delarue, Probabilistic Theory of Mean Field Games with Applications II: Mean Field Games with Common Noise and Master Equations, vol. 84. Springer, 2018.
70. C.-X. Wang, J. Huang, H. Wang, X. Gao, X. You, and Y. Hao, ‘6G Wireless Channel Measurements and Models: Trends and Challenges’, IEEE Veh. Technol. Mag., vol. 15, no. 4, pp. 22–32, 2020. [CrossRef]
71. H. Sedjelmaci, S. M. Senouci, and N. Ansari, ‘Intrusion Detection and Ejection Framework Against Lethal Attacks in UAV-Aided Networks: A Bayesian Game-Theoretic Methodology’, IEEE Trans. Intell. Transp. Syst., vol. 18, no. 5, pp. 1143–1153, 2017. [CrossRef]
72. S. Sharma and A. Kaul, ‘VANETs Cloud: Architecture, Applications, Challenges, and Issues’, Arch. Comput. Methods Eng., vol. 28, no. 4, pp. 2081–2102, 2021. [CrossRef]
73. H. Liu, L. Yang, and H. Yang, ‘Cooperative Optimal Control of the Following Operation of High-Speed Trains’, IEEE Trans. Intell. Transp. Syst., vol. 23, no. 10, pp. 17744–17755, Oct. 2022. [CrossRef]
74. V. Kumar and D. Sinha, ‘A robust intelligent zero-day cyber-attack detection technique’, Complex Intell. Syst., vol. 7, no. 5, pp. 2211–2234, Oct. 2021. [CrossRef]
75. H. Sedjelmaci, M. Hadji, and N. Ansari, ‘Cyber security game for intelligent transportation systems’, IEEE Netw., vol. 33, no. 4, pp. 216–222, 2019.
76. H. Sedjelmaci, S. M. Senouci, and N. Ansari, ‘A hierarchical detection and response system to enhance security against lethal cyber-attacks in UAV networks’, IEEE Trans. Syst. Man Cybern. Syst., vol. 48, no. 9, pp. 1594–1606, 2018.
77. K. Lu, G. Jing, and L. Wang, ‘Distributed algorithms for searching generalized Nash equilibrium of noncooperative games’, IEEE Trans. Cybern., vol. 49, no. 6, pp. 2362–2371, 2018.
78. Gupta, R. K. Jha, and S. Jain, ‘Attack modeling and intrusion detection system for 5G wireless communication network’, Int. J. Commun. Syst., vol. 30, no. 10, p. e3237, 2017.
79. Charrada and A. Samet, ‘Joint interpolation for LTE downlink channel estimation in very high-mobility environments with support vector machine regression’, IET Commun., vol. 10, no. 17, pp. 2435–2444, 2016.
80. P. Kulkarni, N. Saxena, and A. Roy, ‘Efficient Video QoE Prediction in Intelligent O-RAN.’, Comput. Netw. Commun., vol. 1, no. 2, pp. 343–356, 2023. [CrossRef]
81. S. Zhang, X. Li, M. Zong, X. Zhu, and D. Cheng, ‘Learning k for KNN Classification’, ACM Trans Intell Syst Technol, vol. 8, no. 3, Jan. 2017. [CrossRef]
82. U. Challita, A. Ferdowsi, M. Chen, and W. Saad, ‘Machine Learning for Wireless Connectivity and Security of Cellular-Connected UAVs’, IEEE Wirel. Commun., vol. 26, no. 1, pp. 28–35, 2019. [CrossRef]
83. D. G. Riviello, F. Di Stasio, and R. Tuninato, ‘Performance Analysis of Multi-User MIMO Schemes under Realistic 3GPP 3-D Channel Model for 5G mmWave Cellular Networks’, Electronics, vol. 11, no. 3, 2022. [CrossRef]
84. S. Khan, ‘The 5G Network Backbone: A Guide to Small Cell Technology’, Telit. Accessed: Feb. 06, 2024. [Online]. Available: https://www.telit.
85. M. Dangana, S. Ansari, Q. H. Abbasi, S. Hussain, and M. A. Imran, ‘Suitability of NB-IoT for Indoor Industrial Environment: A Survey and Insights.’, Sensors, vol. 21, no. 16, Aug. 2021. [CrossRef]
86. T. F. Rahman, A. S. Abdalla, K. Powell, W. AlQwider, and V. Marojevic, ‘Network and Physical Layer Attacks and countermeasures to AI-Enabled 6G O-RAN’. 2022.
87. A. Barakabitze, A. Ahmad, R. Mijumbi, and A. Hines, ‘5G network slicing using SDN and NFV: A survey of taxonomy, architectures and future challenges’, Comput. Netw., vol. 167, p. 106984, 2020.
88. S. Sharma, R. Miller, and A. Francini, ‘A cloud-native approach to 5G network slicing’, IEEE Commun. Mag., vol. 55, no. 8, pp. 120–127, 2017.
89. M. A. Al-Garadi, A. Mohamed, A. K. Al-Ali, X. Du, I. Ali, and M. Guizani, ‘A survey of machine and deep learning methods for internet of things (IoT) security’, IEEE Commun. Surv. Tutor., vol. 22, no. 3, pp. 1646–1685, 2020.
90. K. Fan, Y. Ren, Y. Wang, H. Li, and Y. Yang, ‘Blockchain-based efficient privacy preserving and data sharing scheme of content-centric network in 5G’, IET Commun., vol. 12, no. 5, pp. 527–532, 2018.