Security in 5G-Enabled Internet of Things Communication: Issues, Challenges, and Future Research Roadmap

5G mobile communication systems promote the mobile network to not only interconnect people, but also interconnect and control the machine and other devices. 5G-enabled Internet of Things (IoT) communication environment supports a wide-variety of applications, such as remote surgery, self-driving car, virtual reality, flying IoT drones, security and surveillance and many more. These applications help and assist the routine works of the community. In such communication environment, all the devices and users communicate through the Internet. Therefore, this communication agonizes from different types of security and privacy issues. It is also vulnerable to different types of possible attacks (for example, replay, impersonation, password reckoning, physical device stealing, session key computation, privileged-insider, malware, man-in-the-middle, malicious routing, and so on). It is then very crucial to protect the infrastructure of 5G-enabled IoT communication environment against these attacks. This necessitates the researchers working in this domain to propose various types of security protocols under different types of categories, like key management, user authentication/device authentication, access control/user access control and intrusion detection. In this survey paper, the details of various system models (i.e., network model and threat model) required for 5G-enabled IoT communication environment are provided. The details of security requirements and attacks possible in this communication environment are further added. The different types of security protocols are also provided. The analysis and comparison of the existing security protocols in 5G-enabled IoT communication environment are conducted. Some of the future research challenges and directions in the security of 5G-enabled IoT environment are displayed. The motivation of this work is to bring the details of different types of security protocols in 5G-enabled IoT under one roof so that the future researchers will be benefited with the conducted work.


I. INTRODUCTION
5G is the fifth generation mobile communication systems which promotes the mobile network to not only interconnect The associate editor coordinating the review of this manuscript and approving it for publication was Alessandro Pozzebon.
people, but also interconnect and control the machine and other devices (i.e., smart devices). It improves the performance and efficiency which capacitate the user experiences. 5G delivers peak rates up to 10 Gbps, ''ultra-low latency'' and ''massive capacity'' which provides the consistence and uniform service to the user. It is a wireless networking The evolution of mobile communication system is also highlighted in Fig. 1 which shows the progress over the last few decades. It also includes the details of functionality features supported by various generations.
5G is designed with the forgoing four mega trends (for example, increased number of devices, traffic growth, high dependency on cloud and different 5G convergence services). Some suggestions were given by some interested companies for selecting the key performance indicators which should be used in 5G. On the basis of these suggestions, the International Telecommunication Union-Radiocommunication Sector (ITU-R) [10] selected some of the important parameters that are listed in Table 1.
• Delivers ultra low latency (1 ms) which allows users to experience less delay when requesting data from the network.
• Capacity increases as the the network expands.
• Manageable with the previous generations of mobile communication technologies.
• Effective and supportive for heterogeneous services (i.e., private network).
• Provides uniform, uninterrupted, and consistent connectivity for the various applications (i.e., communication of smart vehicles) across the world.
IoT is a communication environment of connected physical objects having unique address (i.e., IP address) which are accessible using the Internet. ''Thing'' in IoT can be a patient with smart health monitoring devices or a smart vehicle with built-in-sensors. These objects (i.e., smart devices) have assigned an IP address and using that they are able to collect and transfer data over a network without manual assistance [11]- [15]. Fifth Generation (5G) communication system has a great impact over the communication happens in IoT. 5G activates innovation across many industries (i.e., automobile sector, health sector) and provides a platform to enable the emerging technologies (for example, IoT) to become an essential part of the economy and people's daily lives.
was not possible in the past, that happens because of the advancement of communication technology (i.e., evolution of IoT) [16], [19].
-Automation: Automation is biggest requirement of current time which manages day-to-day activities without any human involvement. In business, it helps to boost the quality of services and further scale downs the human involvement.
• Disadvantages of IoT -Privacy and security: In current tech driven time each and every device which an individual uses is connected to the Internet. This further increases the risk of leakage of sensitive information. This is the major drawback of such kind of communication. If information is not handled properly may lead to the disclosure of confidential information to the third party. Therefore, some strong authentication, access control, intrusion detection and privacy preservation protocols for IoT communication to protect against any kind of data leakage attack are required [7]- [9], [13], [14], [23]- [26]. -Complexity and compatibility issues: IoT consists of different types of devices and networking protocols which are used to connect them. A single mistake in that system can affect the entire performance [27]. -Reduction in number of jobs: The need for human labor reduced drastically with the task automation as machine replaces the human (labor). This further impact the employ-ability. With the emergence of IoT applications in future, there will be decline in the number of jobs available to the professionals [28]- [30]. -Dependability: There is a drastic change in the technology and its applications. Technology is dominating our lifestyle and further increasing our dependability on technology (i.e., IoT applications). But at the same time it also has some drawbacks for example, if there is some vulnerability in a system then there are the possibilities that device does not work properly and can cause other serious consequences [28]- [30].

D. CONVERGENCE OF 5G AND IoT
It is an estimate that twenty billion IoT devices will be connected to a worldwide network up to 2020 which will produce enormous amount of data. Researchers have created an architecture called the 5G I-IoT paradigm, which is a merger of 5G cellular network, artificial intelligence (AI) and IoT that creates an Intelligent IoT (I-IoT) environment. Such kind of merger of three technologies provide efficient access to the users for the data generated by IoT devices which is further useful to make some sense from this generated data. The evolution of 5G networks becomes a major driving force for the growth of IoT as 5G provides extended coverage, faster speeds along with massive bandwidth as compared to other cellular network communication system. 5G I-IoT paradigm connects IoT devices (i.e., IoT sensors) to a cloud server, where the sensory data is gathered, processed and analyzed using AI algorithms. This analysis helps the users for further decision making. It makes the users capable to take appropriate decisions relevant to their field of work (i.e., smart transportation, smart agriculture, and smart healthcare) [13], [23], [27], [36]- [38].

E. APPLICATIONS OF 5G-ENABLED IoT COMMUNICATION ENVIRONMENT
It can be convinced to everybody that 5G will be the next revolution. Some of the applications of 5G-enabled IoT communication environment are discussed below.
• Remote surgery: One of the tremendous advantage of 5G is its low latency feature. Because it has short time lag between a device pinging the network and getting the response whereas it was the problem with 4G LTE. Because of such characteristics of 5G network a surgeon can now perform a remote surgery and his/her physical presence is not required in the same operation theater. For example, the doctors at King's College London has demonstrated the surgery procedure in which they used a dummy patient by the help of ''virtual reality headset'' and ''special glove''. Through that they have performed remote surgery by the help of remote robotic arm [39], [40].
• Self-driving car: A autonomous vehicle (AV) or selfdriving car is a kind of smart vehicle which has ability to sens its surroundings and safety travelling with negligible human involvement. Self-driving cars are installed with different types of sensors to get information about their surroundings. These vehicle are also inbuilt with some other important functionalities such as inertial measuring units, odometry, radar and GPS system. The equipped control systems can interpret the sensed data to identify the required navigation paths along with the existing obstacles. AVs are also capable to communicate among each other for proper decision making and to select the appropriate path to reach to a destination. All such communication facilitates through 5G mobile communication system [41]- [43].
• Virtual reality (VR): Its a kind of artificial environment which is created using the software and other related tools. It is presented to the user in such a way that the user suspends the originality and agrees to accept it like an actual environment. Mostly it is felt through senses for example, sound and sight. In a recently held ''mobile world congress'', it was demonstrated that ''how 5G could enhance the user's experience''. It permitted the user to do chatting in the real time with live-streaming virtual worlds [44]- [46].
• Flying IoT (Drones): Internet of Drones (IoD) is a ''layered network control architecture'' which is implemented to coordinate the access of unmanned aerial vehicles (i.e., drones). It is used to control the airspace and carries out navigation services between the different locations. The various services of flying IoT are like traffic surveillance, search & rescue and package delivery. These drones can communicate to the base station (server) through the Internet connection. They capture (monitor) the surroundings and transmit the data to the base station using the Internet. In order to make such kind of communication a very efficient mobile communication network is required. Therefore, 5G will be a suitable match for this [47], [48].
• Security and surveillance: Surveillance and analytics is another application which will be very successful with 5G connectivity. Due to the increasing threats to public safety in recent years, many governments and security agencies installing the public surveillance and security systems. Majority of public video surveillance systems still rely on wired networks. However adoption for wireless communications such as WiFi or even more the 5G system is also gaining popularity due to the easy and fast set-up with low cost. Closed circuit television (CCTV) systems (for example, cameras installed in vehicles (i.e., police cars), public transport and surveillance drones are getting popularity. The adoption of 5G communication system boost up the performance which is required for sophisticated video content analysis in realtime and the deployment of massive numbers of cameras in the targeted regions [49]- [51].
• Transforming healthcare: Most of the time when somebody has some illness and he/she needs some medical treatment and attention travel to a doctor's clinic or hospital. But it is very difficult for the people living in the rural area where that kind of medical facilities are not available. Moreover, travelling in illness is challenging and time-consuming. However, with the advancement of information and communications technology (ICT), latest tools and technologies for example, telehealth and remote home monitoring systems are available through which care can be received within the comfort of the homes. Remote home health monitoring system consists of health monitoring devices (i.e., implantable or wearable health devices) and other video & imaging facilities in which the communication technologies such as 5G and IoT can be utilized. These devices sens and transmit the medical information of a patient (user) to a central authority (i.e., cloud server) through which a health expert (i.e., a doctor) can also access the medical data of the patient. Doctors can suggest medicine (treatment) prescription after a short video call remotely. Such kind of remote monitoring is equipped with sophisticated imaging equipment and produces enormous amount of health sensing data which causes additional strain on the networks. This often increases the congestion and slows down the data transfer speeds, especially in a large healthcare system which may interface with huge number of patients in VOLUME 9, 2021 a single day. Therefore, 5G mobile communication system which has very high data transfer speed along with low latency will be very helpful for such kind of healthcare system [2], [23], [52], [53].
Here, the details of the existing surveys on ''security protocols in 5G-enabled IoT environment'' are provided. This will help the readers how our work is different than the other existing works. Granjal et al. (2015) [17] performed an analysis on the security protocols available to protect the communications in IoT. They also discussed the existing research proposals and challenges in IoT security for the future researchers.
Khan et al. [31] (2018) presented the major security issues in IoT communication. They categorized the famous security issues as per the IoT architecture. Further they discussed the security requirements, existing attacks, threats and possible solutions in IoT communication. Moreover, they explained how blockchain technology could be utilized to resolve the IoT security problems. Some of the open research problems and challenges in IoT security were also highlighted.
Das et al. [32] (2018) presented a survey on security protocols in IoT. They have provided the details of various types of attacks possible in IoT communication. A taxonomy of security protocols under different categories such as ''key management'', ''user authentication'', ''identity management'' and ''access control'' is provided. Few emerging research directions of the future are also highlighted. Khan et al. (2019) [1] presented a survey work contained the details of core technologies which are used to implement a 5G security model. They discussed the security monitoring and management of 5G networks. The related security measures and standards of core 5G technologies were also added. Moreover, some key projects of international significance in 5G were discussed. Some future research directions were also highlighted.
Zhou et al. (2019) [33] identified some reasons for threats and challenges in IoT security. The eight IoT features which had the most impact on security and privacy of IoT communication were discussed. The current trends of IoT security along with the reasons and explanations were highlighted.
Noor et al. (2019) [34] presented the analysis of recent research in IoT security i.e., the trends and open issues. The details of the current state of IoT security research, the relevant tools, ''IoT modellers'' and ''IoT simulators'' were also provided. Sengupta et al. (2020) [35] classified the attacks on the basis of objects of vulnerability. These attacks were also mapped according to the layers of generalized IoT architecture. Some countermeasures for these attacks were also provided. The features of blockchain along with its integration into the IoT applications were discussed. A taxonomy of IoT and Industrial IoT security research areas along with the countermeasures were presented. Some of the research directions of this domain were also highlighted.
The summary of the existing surveys on security in 5G-enabled IoT environment is also tabulated in Table 2.

G. MOTIVATION OF THIS WORK
5G-enabled IoT environment, suffers from different types of security and privacy related issues as it is vulnerable to various types of attacks such as ''replay'', ''man-in-the-middle'', ''impersonation'', ''password guessing'', ''physical device stealing'', ''illegal session key computation'', ''privilegedinsider'', ''malware'', ''malicious routing'', many more. Hence it becomes essential to protect the infrastructure of 5G-enabled IoT environment against the different types of possible attacks. Therefore, time to time researchers proposed different types of security protocols under different types of categories like ''key management'', ''user authentication/device authentication'', ''access control/user access control'' and ''intrusion detection''. In this reviewed paper our motivation is to bring all these work under the one roof. So that the future readers will be much benefited with this work. Henceforth, the details of different types of security protocols for 5G-enabled IoT environment are provided.

H. MAIN CONTRIBUTIONS
The contributions of this work are provided below.
• The details of various types of system models i.e., network model and threat model require for 5G-enabled IoT environment are provided.
• Security requirements along with the potential attacks in 5G-enabled IoT environment are further added.
• The analysis and comparisons of the existing security protocols in 5G-enabled IoT environment are conducted.
• Finally, some of the future research challenges and directions in the security of 5G-enabled IoT environment are highlighted.

I. ORGANIZATION OF THE PAPER
Remaining part of the paper is presented as follows. Various system models i.e., network model and threat model require for 5G-enabled IoT environment are presented in Section II. Section III consists of details of security requirements and potential attacks in 5G-enabled IoT environment. The numerous categories of security protocols i.e., ''key management'', ''user authentication/device authentication'', ''access control/user access control'' and ''intrusion detection'' in 5G-enabled IoT environment are given in Section IV. Section V consists of the details of analysis and comparisons of the existing security protocols in 5G-enabled IoT environment. Section VI contains the details of research challenges for the future in the security of 5G-enabled IoT environment and some directions to resolve them. The details of lessons that has been learnt from this work are available in Section VII. Section VIII contains some concluding remarks of this work.

II. SYSTEM MODELS
In an 5G-enabled IoT environment following models can be utilized.

A. NETWORK MODEL
The network model of the 5G-enabled IoT environment is given in Fig. 2. In this figure there are different types of scenarios i.e., smart farming, smart home, smart manufacturing, smart transportation system, smart healthcare and smart grid. These scenarios consists of various types of smart IoT devices such as smart vehicles, smart healthcare device (for example, connected inhalers), smart home appliances (for example, smart AC controller), smart meters, smart soil VOLUME 9, 2021 The widely-accepted ''Dolev-Yao (DY) threat model'' [57] can be accepted in the designing of ''security protocols'' in 5G-enabled IoT environment. As information provided in Section IV, the security protocols are the collection of different types of protocols such as ''key management'', ''user authentication/device authentication'', ''access control/user access control'' and ''intrusion detection''.  [58], also the current de facto standard model for the designing and modeling of an ''authenticated key agreement'' security mechanism. In this model, A can have same capabilities like the DY model. Moreover, A can also accommodate the secret credentials along with the ''session states & session keys'' utilized in a session. Furthermore, A has the ability to physical capture some of the smart IoT devices to deduce the stored information from the devices by the application of steps of ''power analysis attack'' [59]. The obtained data can be used to launch other types of severe attacks like computation of ''identities'', ''passwords'' and ''session key''. This helps A to launch other attacks in the network for example, ''device impersonation attack'', ''replay attack'', ''man-in-the-middle attack'' and ''privileged-insider attack''. Apart form that A can reproduce new malicious devices with other attack launching functions (i.e., wormhole, blackhole) by the help of extracted information. After fabricating these malicious devices (i.e., blackhole attacker nodes) A directly deploys them in the target area [60]- [62]. Under the effect of discussed attacks the data packets (messages) can be lost, dropped, delayed or updated. Again it affects the performance of the communication in an 5G-enabled IoT environment.

III. SECURITY REQUIREMENTS AND POTENTIAL ATTACKS IN 5G-ENABLED IoT COMMUNICATION
Here, various ''security and privacy'' issues of 5G-enabled IoT communication are discussed. Apart from that the security requirements and the possible attacks are discussed.

A. SECURITY AND PRIVACY ISSUES IN 5G-ENABLED IoT COMMUNICATIONS
Although IoT is rapidly growing and researchers discovered the new techniques to fix the issues in IoT communication.
• Lack of robust security schemes: IoT devices are connected with the system i.e., desktop or smartphone.
In such an environment the lack of security mechanism improves the threat of leakage of personal data. The collected and transmitted data of IoT devices may be disclosed (i.e., health data collected through smart healthcare devices).
• Openness of the network: It is necessary to connect IoT devices with a consumer network which have connection with the other the systems. It is also possible that IoT devices comprehend with some security vulnerabilities, which may be harmful for the network of the consumers. Because it may become the entry point for the attacker to get entry into the system.
• Privacy of sensitive data: IoT environment consists of different types of devices, which have various types of hardware and software. Some of them may be vulnerable to different attacks i.e., replay, MITM, impersonation, password guessing, etc. Therefore, the sensitive data may be leaked through unauthorized access and manipulations. Some of these devices transmit user's personal information such as name, address, contact number, date of birth, health data and credit card. Hence it is always required to protect the communication of IoT against the possible attacks.

B. SECURITY REQUIREMENTS IN 5G-ENABLED IoT ENVIRONMENT
The essential security requirements in 5G-enabled IoT environment along with the ''general security requirements'' are provided here [68]: • Authentication procedure: This procedure is used to validate the identities of the communicating devices (i.e., IoT device). Mutual verification of identities is mandatory to start a secure communication and it should be conducted in advance. In a 5G-enabled IoT environment, authentication may be happened among the smart IoT devices, various kind of servers (i.e., cloud, fog, edge servers), various kind of users, the service providers and gateway nodes.
• Integrity property: It provides assurance of real and accurate data. The composition of the received message should not accommodate illegal inject, unapproved modification and removal. Data should be safeguarded against unauthorized modification.
• Confidentiality property: This property provides protection of information against any kind of unauthorized access. In another way, it is known as ''privacy'' which protects the exchanged messages against any kind of disclosure attacks.
• Non-repudiation property: It assures any entity should not deny the validity of something for example, a message. It is one of important service in ''information security'' which provides proof regarding the ''origin of the message'' and ''integrity of the data'' in that message. Therefore, it is very difficult for the illegal entities to deny the ''origin of the message'' and ''authenticity of the message''. Digital signatures methods are useful for ''non-repudiation''. For example, in case of a online transactions, where it is important to assure that a party to whom the contract was made (communicated party) should not deny the originality of his/her signature on the report. It has following categories: -Source's non-repudiation: It provides assurance of genuineness of the sender. This concludes the message was transmitted by the genuine source. -Destination's non-repudiation: It provides assurance of genuineness of the receiver. This concludes the message was received by the genuine receiver.
• Authorization procedure: Authorization procedure is used to regulate device or user privileges (i.e., access restrictions) for network or system resources (i.e., files, data applications and services). Generally, it is anticipated by authentication procedure for the verification of identities of device or user. In general a authority (for example, a administrator) designs and implements the access rules for all available resources.
• Freshness property: This property provides assurance of ''data freshness''. Hence the illegal entity will not be able to re-transmit the previously exchanged messages.
• Availability property: It provides assurance that the information is only available to the genuine entities. If an adversary is not able to produce the attack on the confidentiality and the integrity then he/she may contend for other hazardous attacks such as distributed denialof-service (DDoS) attack on a web server which brings down the associated websites.
• Forward secrecy property: If an entity (for example, smart home user, smart IoT device) leaves the communication network then it must not have any ability to access the future messages.
• Backward secrecy property: If an entity (for example, smart home user, smart IoT device) is just deployed in the communication network then this entity should not be able to access the previously exchanged messages. VOLUME 9, 2021

C. POSSIBLE ATTACKS IN 5G-ENABLED IoT COMMUNICATION
An 5G-enabled IoT communication can have following potential attacks which may be performed by a passive or an active adversary [69]: • Eavesdropping: It is also called as sniffing or snooping attack. It happens in case when attacker eavesdrops the exchanged messages among the communicating parties. It is one of the potential attacks in 5G-enabled IoT communication as this will help the attacker to launch further attacks.
• Traffic analysis: It is a another form of passive attack in which attacker does the interception and an examination of the exchanged messages to figure out what's going on there.
• Replay attack: It happens when a attacker intercepts the exchanged messages and then deceitfully delays or re transmits it to confound the receiving entity.
• Man-in-the-middle attack (MITM): In this malign activity, first attacker expropriates the transmitted messages and then attempts to update or delete the messages before forwarding them to the receiver.
• Impersonation attack: In this malign activity, a attacker successfully determines the identity of a genuine communicating party and then creates a message and sends that to the recipient on the behalf of the ''genuine communicating party''.
• Denial-of-Service attack: In this malign activity, an adversary conducts some vengeful tasks to prevent the legitimate parties from accessing the resources of the network or system (for instance, some data resource or some IoT device). There is also a variant DoS of attack, called as the ''Distributed DoS (DDoS)'' attack which is conducted through the multiple attacker systems simultaneously. Some of the examples are UDP, HTTP, TCP SYN flooding attacks. The flooded packets under these attacks consume the resources (for example, bandwidth) of the targeted system (i.e., web servers) very quickly [70]- [72]. In an 5G-enabled IoT communication the DoS attack can also preformed through other types of routing attacks for instance, sinkhole, wormhole, greyhole, blackhole and misdirection attacks. In such attacks the physically deployed attacker nodes disturb the ongoing routing process to drop, delay or modify the exchanged packets. In presence of such attacks data messages (packets) may be altered, delayed or dropped before reaching to their intended recipient. This causes increment in the ''endto-end delay'', diminution in the ''throughput'' and also affects the other network performance parameters [60]- [62], [73]. • Insider attack: In this malign activity, ''a privilegedinsider user'' of the trusted authority exploits the stored information to introduce other severe attacks such as session key computation, password guessing attack.
• Physical capturing of deployed devices: The physical monitoring of IoT devices is not possible for 24 × 7 hours. Therefore, sometimes adversary gets chance to capture these device physically and then tries to extract the sensitive information (i.e., identities, secret keys, etc.) from the memories of these devices.
Further adversary may utilize this information to conduct other unwanted activities (for example, impersonation, password guessing, session key computation and man-in-the-middle (MITM) attacks) in an 5G-enabled IoT environment [59].

IV. CATEGORIES OF SECURITY PROTOCOLS IN 5G-ENABLED IoT COMMUNICATIONS ENVIRONMENT
As discussed earlier 5G-enabled IoT communications environment suffers from various attacks. Therefore, researcher working in this domain proposed various security protocols which can be divided into different categories for example, ''key management protocols'', ''authentication/user authentication protocols'', ''access control/user access control protocols'' and ''intrusion detection protocols'' [1], [17], [31]- [35]. In this section, some of existing security protocols of this domain under the discussed categories are briefed. The pictorial view of classification of security protocols in 5G-enabled IoT communications environment is also provided in Fig. 3.

A. KEY MANAGEMENT
A key management protocol does the generation, distribution, establishment and management of cryptographic keys among the communicating parties in an IoT environment. The defined procedure contains various steps such as ''generation of key'', ''exchange of key'', ''usage of key'' and ''revocation of key'' according to the demand. Key management protocol utilizes a ''cryptographic procedure'' that keeps the details of key servers (the trusted authority), various types of users (i.e., static or mobile) and devices involved (for example, smart IoT device). To provide secure communication strong key management practice should be followed [56], [86]- [89]. Ordinarily, a key management protocol contains some phases like ''pre-deployment phase'', ''key generation and distribution phase'', ''key establishment phase'' and ''key revocation & dynamic device addition phase'' [56], [90]- [104]. The details of different phases of key management protocols are provided in Fig. 4.

B. USER AUTHENTICATION/DEVICE AUTHENTICATION
User/device authentication is a procedure of identification and verification of the identities of the communicating entities i.e., user or device. Generally, the communicating entities user, smart IoT devices verify their identities among each other which is also known as mutual authentication. After the successful completion of ''mutual authentication'', the disseminating entities establish a session key to secure their communication. Device authentication also happens in the same way. In order to simplify this, the details of user authentication process are given. A user authentication scheme for an 5G-enabled IoT communications contains phases such as ''system setup and pre-deployment phases'', ''user registration phase'', ''login phase'', ''authentication & key agreement phase'', ''password & biometric update phase'' and ''dynamic device addition phase''. The details of these phases are provided in Fig. 5. The ''two factor'' and ''three factor'' user authentication mechanisms are very common in practice which administer security on the basis of the available factors. The three factors used in authentication methods are like user's credentials (i.e., information of used username and password), user's used device (i.e., smart card or smartphone) and user's biometrics information (i.e., his/her iris scan or fingerprints) [14], [22], [40], [48], [55], [104]- [109].

C. ACCESS CONTROL/USER ACCESS CONTROL
The access control methods put restriction on the access of the user or device to the resources of a network or system. The followed mechanism grants access and privileges to different users or devices for the various available resources.
To enhance the overall lifetime of the IoT communication environment, it is required to add new devices i.e., smart IoT device in the network. That may occur in case if a device stop its working due to battery depletion or some physical stealing [59]. There are also the chances that an adversary tries to install his or her malicious device in the target area [60], [61]. For this reason, it is essential to differentiate among malicious devices and legitimate devices. Accordingly, secure access control methods to restrict the entry of pernicious entities in 5G-enabled IoT environment should be designed. Access control protocols has two categories ''certificate-based access control'' and ''Certificate-less access control''. These protocols contain phases such as ''device authentication phase'' and ''key establishment phase'' [24], [54], [110], [111]. The details of different categories and phases of access control protocols are provided in Fig. 6. However, It is important to notice that to provide the access to the genuine users for certain resources and services user access control protocols are utilized.

D. INTRUSION DETECTION
Intrusion detection protocols are deployed for the monitoring and analysis of pernicious exertions inside a system or a network. A system which does this work of ''intrusion detection'' is named as ''intrusion detection system (IDS)''. An IDS defends various devices i.e., smart IoT device against the potential attacks. The used intrusion detection technique in an 5G-enabled IoT environment monitors and verifies the different types of traffic (may be normal or malicious), and then predicts the sign of intrusions. If an intrusion is detected then the linked tool takes the required action for example, blocks the IP of malicious source or sends the information of intrusion to the administrator. Apart from that there also some possibilities that an adversary may physically steal few IoT devices. Then adversary may try to deduce the secret information (i.e., credentials, secret keys) from the stolen device through the steps of ''power analysis attacks'' [59]. Further, the adversary tries to set up his/her pernicious nodes (devices) by storing the deduced information in these nodes. These ''malicious devices'' are capable to launch some hazardous attacks i.e., routing attacks (for example, sinkhole, blackhole, misdirection, wormhole, etc.) [60]- [62], [73]. Under the effect of these potential attacks, the exchanged information may be leaked, modified, delayed or dropped  before forwarding them to the required destination. That causes serious deterioration in the performance of the network i.e., improvement in ''end-to-end delay'', decrement in ''network throughput'' and ''packet delivery ratio'' [61], [73]. Moreover, IoT botnets may also attack such kind of communication through which the bots make efforts to install malware in the operating system or in the memory of IoT devices. Under the existence of malware attacks, the IoT devices may work inaccurately or stop completely. Hence it is crucial to provide protection to the 5G-enabled IoT environment against the intrusions. Therefore, it is mandatory to study intrusion detection protocols for such an environment [12], [13], [74], [112]- [117]. IDS can be divided into two classes on the basis of the deployment: i) ''network based intrusion detection system (NIDS)'' which recognizes interventions inside a network (for example, Suricata) and ii) ''host based intrusion detection system (HIDS)'' which recognizes interventions inside a system (for example, OSSEC). Furthermore, ''intrusion detection techniques'' can be categorised as: i) ''misuse based detection'', ii) ''anomaly based detection'' and iii) ''specification based detection'' [118], [119]. Their features are available in Fig. 7.

V. EXISTING SECURITY PROTOCOLS IN 5G-ENABLED IoT COMMUNICATIONS ENVIRONMENT
As discussed in Section IV, the security protocols in 5G-enabled IoT communications environment can be divided into four categories. Here the details of some of the existing protocols belong to these categories are provided.

A. EXISTING KEY MANAGEMENT PROTOCOLS IN IoT ENVIRONMENT
The details of authentication and key management schemes is provided in the following part of this section. Li et al. [120] presented a identity-based hierarchical model for cloud computing. It has associated encryption and signature techniques which were used in the designing of identity-based authentication schemes for cloud computing services. It was proved that the presented method was efficient than ''secure socket layer (SSL) authentication protocol''. Li et al. [121] used the ''elliptic curve cryptography (ECC)'' for the designing of an authentication protocol suitable for cloud computing environment. Hu et al. [122] proposed methods for identity authentication, data encryption and data integrity checking. The methods were proposed to achieve confidentiality, integrity and availability properties. Mishra et al. [123] also proposed a authentication and key management scheme for secure multimedia communications in IoT-enabled WSNs. Later on it was observed that some of the existing schemes lacking in security and functionality features as various attacks such as ''smart card stolen'', ''impersonation'', ''password guessing'', ''replay'' and ''man-in-the middle'' attacks were possible [109]. Sharif et al. [124] proposed ''a lightweight authentication and key agreement scheme'' for IoT based WSNs. The formal security verification by employing the automated formal security verification software toll named as ''Automated Validation of Internet Security Protocols and Applications (AVISPA)'' [125] is also provided for [124]. Wazid et al. [56] proposed a ''key management and authentication'' technique for the secure communication in fog computing environment. The key management procedure was utilized for the ''smart devices and fog servers'' & ''fog servers and cloud servers'', ''pairwise key establishment'' for the secure communication.

1) COMPARATIVE STUDY OF EXISTING KEY MANAGEMENT PROTOCOLS IN IoT ENVIRONMENT
The comparative study of existing key management protocols in IoT environment is provided in Table 5. The protocols of Li et al. [120], Li et al. [121], Hu et al. [122], Mishra et al. [123], Sharif et al. [124] and Wazid et al. [56] were compared and analyzed. The protocols of Li et al. [120], Li et al. [121], Hu et al. [122] did not support most of the required security and functionality features whereas the protocols of Mishra et al. [123], Sharif et al. [124] did not provide the support for multi-server environment. Moreover, the formal security analysis using the widely accepted Real-Or-Random (ROR) model [126] were not provided for most of the protocols. The protocol of Wazid et al. [56] supported most of the security and functionality features and they have also provided different types of security analysis for their protocol.

B. EXISTING USER AUTHENTICATION PROTOCOLS IN IoT ENVIRONMENT
The details of user authentication protocols in IoT environment is given as follows. Porambage et al. [134] proposed a ''certificate-based authentication scheme'' in WSNs for distributed IoT applications. The developed scheme had ''two-phase authentication protocol'' which allowed the sensor nodes and the end-users to authenticate each other and to maintain a secure connection. Porambage et al. [135] presented two group key establishment protocols for secure IoT communication. The deployment conditions and requirements of the protocols were also described in terms of IoT application scenarios. The proposed protocols were also analyzed and justified in terms of performance analysis, scalability and security. Turkanovic et al. [136] proposed ''a user authentication and key agreement scheme'' for heterogeneous WSNs applicable to IoT. The presented scheme helped the remote user to establish a secure session key with a sensor node by the help of a lightweight key agreement mechanism. The presented method also ensured the ''mutual authentication'' among the user, sensor node and the gateway node. Their presented scheme could be adapted for the resourceconstrained WSN as it only utilized ''simple hash'' and ''XOR computations''. Farash et al. [137] proposed a protocol for ''user authentication and key agreement'' for heterogeneous WSN applicable for IoT environment. Challa et al. [11] presented a user authentication protocol for IoT applications. The presented scheme was based on ''elliptic curve cryptography (ECC)'' and depended on ''ECC-based digital signature''. The presented scheme supported user anonymity and untraceability properties. However, this scheme needed more computation and communication costs. Wazid et al. [138] proposed ''a lightweight authentication protocol'' in cloudbased IoT environment (LAM-CIoT). Through LAM-CIoT, an authenticated user could access the data of a remote IoT sensors securely. LAM-CIoT utilized efficient ''one-way cryptographic hash functions'' along with ''bitwise XOR operations''. LAM-CIoT was also analyzed for security part by the help of formal security analysis using the widelyused ''Real-Or-Random (ROR)'' model, formal security verification through ''Automated Validation of Internet Security Protocols and Applications (AVISPA)'' tool as well as the informal security analysis. It proved the resilience of the presented protocol against possible attacks.

1) COMPARATIVE STUDY OF EXISTING USER AUTHENTICATION PROTOCOLS IN IoT ENVIRONMENT
The comparisons of functionality and security features of user authentication protocols in IoT environment is provided in Table 8. The schemes of Porambage et al. [134], Porambage et al. [135], Turkanovic et al. [136] and Farash et al. [137] did not support most of the security and functionality features. However, the scheme of VOLUME 9, 2021 Challa et al. [11] and Wazid et al. [138] provided most of security and functionality features. But the scheme of Challa et al. [11] produced little more computational and communication overhead. Hence from the security, functionality features and computation & communication costs point of views the scheme of Wazid et al. [138] seems much better.

C. EXISTING ACCESS CONTROL PROTOCOLS IN IoT ENVIRONMENT
The facts of access control protocols in IoT environment are presented in the following part of this section. Li et al. [139] proposed an access control technique for wireless sensor networks applicable to IoT environment. They have utilized a heterogeneous signcryption scheme which allowed the party to send a message to another party in  an identity-based cryptography environment. Their protocol is costly in terms of computation cost due to utilization of identity-based cryptography (IBC) and bilinear pairing mechanisms. Braeken et al. [140] proposed an efficient and distributed authentication protocol (eDAAAS) for the accessing of end nodes in an IoT environment of smart home. Their protocol was based on symmetric key cryptographic technique and one-way cryptographic hash function. Though this protocol had low computation cost but it had high communication cost. Furthermore, this protocol involved gateway node in the access control mechanism between the two IoT devices. Luo et al. [141] presented a access control protocol for WSNs applicable to IoT environment. It facilitated the communication of Internet user in an certificate less cryptography environment with the smart device in an IBC environment. However, their protocol was costly in terms of computation cost because of the utilization of IBC and bilinear pairing mechanisms. Ding et al. [142] presented a attribute-based access control mechanism for IoT environment. The blockchain technology was utilized to record the distribution of attributes to avoid the single point failure and for data security. Riad et al. [143] proposed access control protocol for managing and securing the cloud-hosted electronic health records. Their protocol ensured the secrecy of patient's data in which only authorized users were able to edit or review the health data of patients. Fan et al. [144] presented a privacy preserving multi-authority access control protocol for fog-based IoT environment. In their protocol verifiable outsourced decryption was used to reduce the computation costs for the end user devices. A secure user revocation method was also designed. The discussed protocols [139]- [141], [143], [144] are vulnerable to the session key leakage and other attacks [145]. Das et al. [145] presented a certificate-based lightweight access control and key agreement protocol for IoT environment. It utilized mechanisms such as elliptic curve cryptography (ECC) and collision-resistant one-way cryptographic hash function. Various analysis i.e., formal security analysis under ''Real-Or-Random (ROR) model'', informal (non-mathematical) security analysis and formal security verification through ''Automated Validation of Internet Security Protocols and Applications (AVISPA)'' tool were provided. The conducted analysis proved the resilience of their protocol against possible attacks.

1) COMPARATIVE STUDY OF EXISTING ACCESS CONTROL PROTOCOLS IN IoT ENVIRONMENT
The comparison of functionality and security features of access control protocols in IoT environment is provided in Table 11. The protocols of Luo et al. [141], Li et al. [139], Braeken et al. [140] and Das et al. [145] (2019) were compared and analysed. The protocols of Luo et al. [141], Li et al. [139], Braeken et al. [140] did not provide the required security and functionality features whereas the protocol of Das et al. [145] provided the most of the desired security and functionality features along with less computation and communications costs. The protocol of Das et al. [145] will be preferable for the access control mechanism in IoT environment.
The comparison of computation costs are provided in Table 9. The protocols of Luo et al. [141], Li et al. [139], Braeken et al. [140] and Das et al. [145] are compared and analyzed. The similar assumptions are taken as the information available in Section V-A1. However, T bp and T me are time needed for bilinear paring operation and modular exponentiation operation. As reported in [145], time required for T bp ≈ 32.71 ms and for T me ≈ 2.25 ms. The computation costs of protocols of Luo et al. [141], Li et al. [139], Braeken et al. [140] and Das et al. [145] [140] and Das et al. [145] are 3040, 3488, 3552 and 3296, respectively (as shown in Table 10. The protocols of Braeken Luo et al. [141] and Das et al. [145] performed better from the communication cost point of view.

D. EXISTING INTRUSION DETECTION PROTOCOLS IN IoT ENVIRONMENT
The details of different intrusion detection protocols in IoT environment are provided. Wazid et al. [61] designed a intrusion detection mechanism for detecting and preventing VOLUME 9, 2021 sinkhole attacker nodes in a cluster based wireless sensor network. The operations were conducted in two phases, in first phase it discovered the existence of mischievous nodes through various network performance parameters. If a node came under suspicious category then the steps of second phase confirmed the node as the sinkhole attacker node along with its types i.e., ''sinkhole message modification node'', ''sinkhole message dropping node'' and ''sinkhole message delay node''. Selvakumar et al. [146] presented a intrusion detection protocol on the basis of temporal reasoning method. It had utilized ''multi-class classification'' through self designed mechanism called as ''fuzzy and rough set based nearest neighborhood algorithm (FRNN)''. Wazid et al. [60], [62] presented a intrusion detection method for detecting blackhole attacker nodes and for ''hybrid anomaly'' in a cluster based wireless sensor network. They detected attacker nodes by the help of resource rich cluster head nodes. These proposed protocols are also applicable for the detection of intrusions in IoT environment. Jan et al. [147] designed a light weight intrusion detection technique for the mitigation of common type DOS attacks in IoT. From the packet transmission rate, two-three features were extracted which they were utilized for the reduction of the overall time required to classify the various flows of traffic. There were reduction in the complexity and time required by ''support vector machine (SVM)'' for the classification which results in an efficient mitigation DOS attack. Sharma et al. [148] designed a light weight detection technique based on behaviour rule specification for ''IoT-embedded cyber-physical systems''. The existence of intruder was detected via misbehaviour feature of an existing node. In the designed mechanism they utilized a ''profiler'' which read the module and forward this data to fuzzy analysis segment to verify the validity of behavior rules. Pajouh et al. [149] presented a intrusion detection mechanism for different attacks happened in IoT environment. The presented protocols used two mechanism for the reduction of dimensions and for minimizing the number of features required. The classification algorithm for example, ''KNN'' and ''naive Bayes'' were used for the detection of malign nodes. Mudgerikar et al. [150] presented a client system based intrusion detection protocol which utilized anomalies based detection method. It was deployed with three layers of security. But some drawbacks were there as with the increasing level of security the increment in the overhead occurred i.e., computation and communication costs. In the first component of the detection module a white list was prepared during the learning process. It separated the legitimate processes from the malicious processes. The comparisons was done on the basis of IDs of the processes. In the second module classifier was trained using the generated logs obtained through learning phase. However, it was very expensive procedure because of the use of machine learning algorithms in the resource constrained devices. Saeed et al. [151] presented a intrusion detection technique consisted of two phases to maintain the security in the system. The neural network was utilized in the first phase for the anomaly detection purpose. Though in the second phase a tag system was deployed for the anomaly detection. Then the anomaly detection was conducted through tag-checking method. Wazid et al. [73] designed an intrusion detection method to mitigate routing attacks in an edge-based IoT environment. In this technique they have used a resource rich edge node for the detection of intrusion in the network.

1) COMPARATIVE STUDY OF EXISTING INTRUSION DETECTION PROTOCOLS IN IoT ENVIRONMENT
The achievements of different intrusion detection method in IoT environment are compared. For the performance analysis purpose, there are certain parameters which need to be compared are explained as follows. The first parameter is ''detection rate (DR)'' (which is also called as ''true positive rate (TPR) or sensitivity or hit rate'') and the second one is ''false positive rate or fall out (FPR)''. DR can be formulated as ''the number of attackers detected by a intrusion detection scheme divided by the total number of attackers present in the test sample'' which is formulated as [60], [62], However, FPR is estimated as ''the number of nodes falsely detected as attacker nodes'' which is computed as [60], [62], The results of various intrusion detection protocols in IoT environment such as Wazid et al. [61], Wazid et al. [73], Selvakumar et al. [146], Wazid et al. [60], Jan et al. [147], Sharma et al. [148], Pajouh et al. [149], Mudgerikar et al. [150] and Saeed et al. [151] are provided.

VI. FUTURE RESEARCH CHALLENGES
5G-enabled IoT communications environment supports various types of applications, such as smart home, smart transportation, smart healthcare, smart grid and smart manufacturing. This environment requires solitary requirements for example, live processing and accessing of data (i.e., environmental monitoring of an industrial plant in real-time). Such environment generates very huge data (i.e., big data) hence big data analytic process should be used on this data to identify some specific patterns from this (e.g., possibilities of an accident in a coal mine). In this environment all the devices and users communicate through the Internet. Hence it undergoes with some traditional security, privacy, and other kind of challenges. In this part of the paper some of the current challenges of this domain are discussed and further some directions for research work are highlighted.

A. SECURITY OF THE EXISTING PROTOCOLS
Most of the security protocols proposed for IoT environment insecure as they do not furnish the complete security against the possible attacks. Furthermore, some of the existing protocols works for a particular attack and do not work for multiple attacks concurrently. Hence, it is mandatory to design some security protocols which should concurrently protect multiple attacks. Therefore, designing of such kind of protocols for this domain is a challenging task which should be solved by the future researchers [27].

B. EFFICIENT DESIGN OF SECURITY PROTOCOLS
5G-enabled IoT environment contains the resource constrained devices i.e., IoT sensors which have limited computation capability, limited storage size and small battery unit. These devices are not able to execute communication, computation and storage demanding tasks which require more strength in terms of these parameters. Hence it is desirable to design security protocols to such a degree that they should need low computation power, low communication cost and small storage size without negotiating the security of the system [27], [60], [61].

C. SCALABILITY OF SECURITY PROTOCOLS
5G-enabled IoT environment is a combination of heterogeneous network of different communication mechanism and applications. These applications have their own abilities and concerns. In such situations designing of a security protocol for this kind of communication environment will be complex task. For example, in an smart healthcare communication environment, patient's electronic health records which need to be stored over a cloud server for further processing and decision making. In an body area network (BANs), there are different types of devices which generate data and then forward it to the cloud server. This constitutes a ''heterogeneous network'' of different types of communicating devices. Accordingly, some special types of security protocols which can be employed to protect various types of devices of 5G-enabled IoT environment should be designed. Therefore, more research work is needed in this area.

D. PRIVACY OF STORED DATA
The privacy of data concerns with the proper handling of information over the various resource i.e., consent, notice, and regulatory obligations. 5G-enabled IoT environment is also utilized in ''information sensitive'' operations (like smart healthcare). In this privacy-demanding environment, the health monitoring smart devices are used around/inside the body of a patient to monitor his/her health conditions. Then the sensed and collected health data is sent to cloud server(s) for further storage and processing. Usually, this type of communication environment can be affected by different potential attackers [23], [60], [62]. This may cause the leakage of data at transit and the data maintained over the servers. Consequently, it is important to sustain the privacy of the data at transit and also to the stored data. Hence new efficient protocols are much needed to maintain the privacy of data. Thenceforth, more research work is required in this direction.

E. HETEROGENEITY OF USED DEVICES
5G-enabled IoT environment consists of various types of devices from powerful servers, laptops, personal digital VOLUME 9, 2021 assistants, desktops to resource restricted RFID tags and sensing devices. Besides that, these devices work with various types of communication techniques. Devices are also contrasting as per their communication strength, computation capacity, storage size and deployed system software (e.g., operating system). Hence security protocols should be designed in such fashion that it provides protection to different varieties of devices and associated technologies and mechanisms. Thenceforth, more research work is required in this area [27], [60].

F. BLOCKCHAIN BASED SECURITY PROTOCOLS DESIGNING
The operations of blockchain can also be utilized to secure the 5G-enabled IoT environment. Because blockchain operations are decentralized, efficient and transparent to all entities of the communication environment. To devise security protocols in 5G-enabled IoT environment ''blockchain operations'' are also applicable. To perform the required task a block consists of the data about the required functionality e.g., authentication message, data message can be created and added it to the blockchain. In view of that blockchain is provided to legal network entities, then these entities can access the data using the blocks of the blockchain. That work can be accomplished in an effective way using the operation of blockchain. Accordingly, designing of ''blockchain based security protocols'' seems to be a good problem for the future researchers [152], [153].

VII. LESSONS LEARNED
5G-enabled IoT environment provides delay efficient services. It has wide variety of applications, such as healthcare, home automation, remote surgery, autonomous vehicle (AV), virtual reality (VR), flying IoT drones, and security and surveillance. This communication environment facilities dayto-day activities of the people. However, it also agonizes from various ''security and privacy'' issues. To cope this, researchers working in this area proposed various types of security protocols under different categories, such as ''key management protocols'', ''authentication/ user authentication protocols'', ''access control/user access control protocols'' and ''intrusion detection''. The intrusion detection protocols are of three types like ''misuse based detection'', ''anomaly based detection'' and ''specification based detection''. ''Specification based detection'' scheme needs certain constraints and specifications for its deployment. It uses the advantages of the other two schemes. Its false positive rate is also less as compared to the other detection schemes. Inspite of that, it is bit inconvenient as it is time consuming, we need some time in the defining and construction of the required set. It detects ''zero-day attacks'' effectively. In this detection scheme the intrusion detection happens efficiently and effectively with less number of ''false negatives'' and ''false positives'' rates.
The key management protocols of Li et al. [120], Li et al. [121], Hu et al. [122], Mishra et al. [123], Sharif et al. [124] and Wazid et al. [56] were compared and analysed. The protocols of Li et al. [120], Li et al. [121], Hu et al. [122] did not support most of the desired funtionalties and security lineaments, whereas the schemes of Mishra et al. [123], Sharif et al. [124] did not provide the support for environment of multiple servers. Inspite of that the scheme of Wazid et al. [56] supported most of the functionalities and security lineaments.
The user authentication protocols of Challa et al. [11] and Wazid et al. [138] provided most of functionalities and security lineaments. But, the scheme of Challa et al. [11] produced little more computational and communication overheads. Hence, from the functionalities, secuirty lineaments and computation & communication costs point of view, the scheme of Wazid et al. [138] seems finer than Challa et al.'s method [11].
The access control mechanisms of Luo et al. [141], Li et al. [139], and Braeken et al. [140] did not provide the required functionalities and security lineaments, whereas the method of Das et al. [145] provided most of the desired functionalities and security lineaments along with less computation and communications costs.
The intrusion detection mechanisms of Wazid et al. [62], Selvakumar et al. [146] and Mudgerikar et al. [150] performed better with regard to false positive rate and true positive (detection) rate.
There are some future research directions of ''security in 5G-enabled IoT environment'' such as ''security of the existing protocols'', ''efficient design of security protocols'', ''scalability of security protocols'', ''privacy of stored data'', ''heterogeneous nature of devices'' and ''use of blockchain in the designing of security protocols''. These topics should be addressed in near future.

VIII. CONCLUSION
5G-enabled IoT environment suffers from various types of ''security and privacy'' issues as it is susceptible to various types of attacks. It becomes essential to protect the infrastructure of 5G-enabled IoT environment against these attacks. Therefore, different types of security protocols under different categories (for example, ''key management'', ''user authentication/device authentication'', ''access control/user access control'' and ''intrusion detection'') came into picture. In this survey article, the details of various system models (for example, network model and threat model) are provided for 5G-enabled IoT environment. Various security requirements and attacks possible in this communication environment are also outlined. The different types of categories of security protocols are then prepared. The different types of analysis of the existing security protocols in 5G-enabled IoT environment are also organized. Lastly, some challenging problem of future in the security of 5G-enabled IoT environment are displayed to help the researchers working in the same domain. Ashok Kumar Das (Senior Member, IEEE) received the M.Tech. degree in computer science and data processing, the M.Sc. degree in mathematics, and the Ph.D. degree in computer science and engineering from IIT Kharagpur, India. He is currently an Associate Professor with the Center for Security, Theory and Algorithmic Research, International Institute of Information Technology, Hyderabad, India. His research interests include cryptography, network security, blockchain, security in Internet of Things (IoT), Internet of Vehicles (IoV), Internet of Drones (IoD), smart grids, smart city, cloud/fog computing and industrial wireless sensor networks, intrusion detection, blockchain, and AI/ML security. He has authored over 245 papers in international journals and conferences in the above areas, including over 210 reputed journal papers. Some of his research findings are published in top cited journals, such as the IEEE TRANSACTIONS ON   He has authored or coauthored over 950 papers in refereed international journals and conferences, three books, two patents, and one ITU-T Recommendation. He is a member of the Internet Society and a Senior Member of ACM. He received several outstanding leadership and outstanding service awards from the IEEE Communications Society and several best papers awards. He is an IEEE Distinguished Lecturer, a Member Representative of the IEEE Communications Society on the IEEE Biometrics Council, and the President of the scientific council at ParkUrbis-Covilhã Science and Technology Park. He was Director of the Conference Development-IEEE ComSoc Board of Governors, the Technical Activities Committee Chair of the IEEE ComSoc Latin America Region Board, the past Chair of the IEEE ComSoc Technical Committee on eHealth and the IEEE ComSoc Technical Committee on Communications Software, a Steering Committee Member of the IEEE Life Sciences Technical Community, and the Publications Co-Chair. He is also the Editor-in-Chief of the International Journal of E-Health and Medical Communications and an Editorial Board Member of several high-reputed journals. He has been the General Chair and the TPC Chair of many international conferences, including IEEE ICC, IEEE GLOBECOM, IEEE HEALTHCOM, and IEEE LatinCom. VOLUME 9, 2021