**1. Introduction**

Technology has advanced, and as a result, the world of today has seen a number of ground-breaking developments. These outcomes have been demonstrated to be more trustworthy, approachable, and economical in our everyday lives. In addition, people now engage with one another in novel ways in their social circles. Additionally, unmanned aerial vehicles (UAVs) are employed for both commercial and private purposes in addition to being heavily utilized in military contexts. The market potential for medium-sized drones has been estimated by the China Unmanned Aerial Vehicle Industry (CUAVI) to reach CNY 80 billion by 2025 [1], whereas the Federal Aviation Administration (FAA) concluded that there are currently 3 million drones flying in the US sky and that number will increase by four times by the end of 2022 [2]. Drone use is increasing because of its value in a variety of jobs, including through the live broadcasting of events, aerial video shoots, the mobility to move packages from one location to another, and simple navigation as shown

**Citation:** Abro, G.E.M.; Zulkifli, S.A.B.M.; Masood, R.J.; Asirvadam, V.S.; Laouti, A. Comprehensive Review of UAV Detection, Security, and Communication Advancements to Prevent Threats. *Drones* **2022**, *6*, 284. https://doi.org/10.3390/ drones6100284 4.0/). *drones*

Academic Editor: Diego González-Aguilera

Received: 22 August 2022 Accepted: 11 September 2022 Published: 1 October 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/

*Drones* **2022**, *6*, x FOR PEER REVIEW 2 of 21

that number will increase by four times by the end of 2022 [2]. Drone use is increasing because of its value in a variety of jobs, including through the live broadcasting of events, aerial video shoots, the mobility to move packages from one location to another, and simple navigation as shown in figure 1. These drones are commonly employed for transportation purposes due to their cheap maintenance requirements, ability to take-off and land vertically, ability to hover, and high degree of mobility. These drones are frequently out-

swarming of drones [3,4], and they have proven to be an effective choice for surveillance

in Figure 1. These drones are commonly employed for transportation purposes due to their cheap maintenance requirements, ability to take-off and land vertically, ability to hover, and high degree of mobility. These drones are frequently outfitted with computer vision and internet of things (IoT)-like features, particularly for the swarming of drones [3,4], and they have proven to be an effective choice for surveillance and rescue-related missions [5]. There are, however, a few important elements that are connected to UAV security worries. This collection contains the story of the Iranian military jamming an American drone's control signals [6]; nonetheless, it is still difficult to create a security control module for UAVs that is completely foolproof. Additionally, the first drones were unmanned balloons loaded with explosives that were used to assault Venice in Italy [7]. Later, in 1915, the British troops employed these unmanned balloons for photographic-based surveillance during the renowned Battle of Neuve Chapelle [8]. During this time, cameras were not as advanced; hence, this strategy was suggested to improve visibility [8]. In order to find various terrorists, several of them were also used during the Afghan War [9,10]. Prior to today, these drones were usually utilized for military operations, but they are now also chosen for the majority of domestic applications, to the point that Amazon began using drones to carry packages in 2014 [11]. They have also been utilized in fields including agriculture [1,2], for checking building sites [3], and to greatly assist law enforcement organizations with emergency rescue operations. The United States of America started making pilotless aircrafts that could maneuver for roughly a kilometer in the early 1910s. During World War II, the US started developing advanced UAV programs, such as the N2C-2 drone and the OQ-2 communications plane [9], but these endeavors were both expensive and unreliable. In the late 1980s, the US started developing sophisticated drones, and they already have some top-notch micro unmanned aerial vehicles. Drones are also being used in the media business for aerial photography and filmmaking. Drone use is expanding quickly, and at the same time, security and privacy issues have grown more complicated and serious. and rescue-related missions [5]. There are, however, a few important elements that are connected to UAV security worries. This collection contains the story of the Iranian military jamming an American drone's control signals [6]; nonetheless, it is still difficult to create a security control module for UAVs that is completely foolproof. Additionally, the first drones were unmanned balloons loaded with explosives that were used to assault Venice in Italy [7]. Later, in 1915, the British troops employed these unmanned balloons for photographic-based surveillance during the renowned Battle of Neuve Chapelle [8]. During this time, cameras were not as advanced; hence, this strategy was suggested to improve visibility [8]. In order to find various terrorists, several of them were also used during the Afghan War [9,10]. Prior to today, these drones were usually utilized for military operations, but they are now also chosen for the majority of domestic applications, to the point that Amazon began using drones to carry packages in 2014 [11]. They have also been utilized in fields including agriculture [1,2], for checking building sites [3], and to greatly assist law enforcement organizations with emergency rescue operations. The United States of America started making pilotless aircrafts that could maneuver for roughly a kilometer in the early 1910s. During World War II, the US started developing advanced UAV programs, such as the N2C-2 drone and the OQ-2 communications plane [9], but these endeavors were both expensive and unreliable. In the late 1980s, the US started developing sophisticated drones, and they already have some top-notch micro unmanned aerial vehicles. Drones are also being used in the media business for aerial photography and filmmaking. Drone use is expanding quickly, and at the same time, security and privacy issues have grown more complicated and serious.

**Figure 1.** Security and privacy threats of UAVs. **Figure 1.** Security and privacy threats of UAVs.

The major goal of this review article is to give readers a comprehensive understanding of the new advancements that have led to the issues surrounding unmanned aerial vehicles (UAVs), including security threats, privacy concerns, and other limits that are important and cannot be disregarded. To give readers a thorough understanding of the subject, the manuscript has been organized into 10 sections. The major goal is to identify these issues and give all scholars access to a single resource that will allow them to fully understand the most recent trends and work to advance their field.

Section 2 contains the regulatory standards, whereas Section 3 describes the classification of various unmanned aerial vehicles (UAVs). Section 4 of the document discusses the structures and techniques of communication. In Section 5, it is specifically mentioned how and why drones are used. Section 6 covers the key security challenges and weaknesses, whereas Section 7 covers the present constraints. The most recent methods to address these restrictions are also discussed in Section 8, along with open research areas and recommendations in Section 9. The thorough conclusion to this work can be found last, but by no means least, in Section 10.

#### **2. Study Related to Regulations**

Many countries have been following the standard regulations to ensure the security and privacy implications of drones. Many of them have started to propose several step-bystep procedures to license their UAVs [12,13]. If these regulations are not followed, then unlicensed drones are taken under custody and proper legal action is taken against the pilot [12]. As per the media news broadcasted by the British Broadcasting Corporation (BBC), the CAA and FAA have declared some standard operating procedures (SOPs) to maneuver the UAVs at a low altitude [14], which are mentioned below:

	- In the case of a UAV crash, legal action can be taken against any harmful actions or the damage that occurred from UAV failure.
	- UAVs with computer vision or camera surveillance are not allowed to maneuver within 50 m of people or any crowd.
	- UAVs will be summoned if they are not flown within the operator's line of sight.
	- UAVs will be summoned if they are flown at night without proper lighting.

The above standard rules and SOPs ensure the secure operation of drones.

It is also noted that with a dramatic rise in the drone industry, various countries have inducted their own rules as well [15]. Mainly, to operate a UAV, there are three fundamental components: the first is the ground control room (GCR); the second one is the communication method, for example, satellite, radio frequency, etc., as illustrated in Figure 2; and last, but certainly not the least, is the UAV itself. There are three different methods to communicate with a drone, i.e., satellite, radio signal, and internet, as shown in Figure 2 [15]. The essential license-exempted radio equipment, along with the frequencies, is mentioned in Table 1 [16].

The standard bandwidth through which a communication is established between a UAV and the ground control room (GCR) is mentioned in Table 1, whereas the other standards are still in progress for the safe operations of a UAV in any vicinity [17].

**Figure 2.** Communication channels mostly used to control UAVs. **Figure 2.** Communication channels mostly used to control UAVs.

**Table 1.** Frequencies through which GCR communicates with UAVs. **Table 1.** Frequencies through which GCR communicates with UAVs.


## 4 5030 to 5091 MHz **3. Classification of UAVs**

**Purpose:** Used in such operations where data sharing is important with ground control room (GCR). One must understand the real sense of calling any drone a UAV. Not all drones can be classified as UAVs. A UAV can be controlled autonomously without a pilot and can be controlled remotely [17].

communication Union (ITU) and, therefore, cannot be used for communication with drones.

#### The standard bandwidth through which a communication is established between a *3.1. Classification of Drones*

controlled remotely [17].

UAV and the ground control room (GCR) is mentioned in Table 1, whereas the other standards are still in progress for the safe operations of a UAV in any vicinity [17]. **3. Classification of UAVs** Drone is a very generic term and can refer to intelligent or autonomous vehicles such that there are unmanned aerial vehicles of different types. This can be hexarotors, quadrotors, multirotors or wing-based air vehicles. Mainly discussing the flying drones, they can be classified into three main categories as follows:

One must understand the real sense of calling any drone a UAV. Not all drones can be classified as UAVs. A UAV can be controlled autonomously without a pilot and can be


The drones with a vertical take-off and landing (VTOL) feature and that can hover at a high rate are known as rotary-wing drones. The most common example is a quadrotor unmanned aerial vehicle that has four brushless DC motors. Drones with the capability to fly aggressively and glide even with heavy payloads are known as fixed-wing drones. They perform a horizontal take-off and landing (HTOL). Lastly, the drones that have both fixed and rotary wings are known as hybrid-wing drones. They are designed to have both features of rotary- and fixed-wing drones so that they can perform HTOL and VTOL along with high-rate hovering. Some of the robots are designed to exhibit the flying motion of a fly [18], and here they proposed a 5% more power-efficient wing by changing the wing stiffener pattern parametrically. An experimental aerodynamic analysis found that this could relate to increased wing stiffness, as well as indications of vortex generation during the flap cycle. The experiments reported an improved generated lift, allowing the DelFly to be outfitted with a yaw-rate gyro, pressure sensor, and microprocessor. These flapping-wings were later scaled to the micro level and are known as flapping-wing micro air vehicles (FWMAVs), due to inspiration taken from microscopic insects. These FWMAVs have the ability to perform activities in urban and interior environments. However, there are many hurdles for the successful flight of these vehicles that are replicating insect flight, including their design, manufacture, control, and propulsion [19,20].

#### *3.2. Classification of UAVs Based on Ground Command and Control*

It is already shown in Figure 2 that any UAV can be controlled remotely using a ground command and control mechanism, either by mobile phone, radio channel frequency, or the internet of things [21]. Therefore, these UAVs are classified based on their ability to fly over long distances without any intervention. These types are mentioned below as:


The above-referred classification is summarized in Table 2 [22] along with the pros and cons of the UAVs.


**Table 2.** Classification of UAVs based on wing type and altitude.

In the above, the reader might be confused with the term of flexible deployment of communication. This states the proper coverage at which the drone can be controlled and stabilized for any sort of task. There are some research contributions that have classified the drones based on their altitude, as demonstrated in Figure 3 along with their examples [23,24]. stabilized for any sort of task. There are some research contributions that have classified the drones based on their altitude, as demonstrated in Figure 3 along with their examples [23,24].

In the above, the reader might be confused with the term of flexible deployment of communication. This states the proper coverage at which the drone can be controlled and

*Drones* **2022**, *6*, x FOR PEER REVIEW 6 of 21

**Table 2.** Classification of UAVs based on wing type and altitude.

**Rotary-Wing Type**

**Fixed-Wing Type**

Transport goods Yes Low

**Factors**

Battery time (in

Flexible deployment of communication

Maneuver speed High

**Based on Wing Type Based on Altitude** 

weight Yes Yes No

Speed Low speed High speed High Speed High Speed

No No No Yes No

**Low Altitude Below 400 ft**  **High Altitude Above 400 ft**

**Hybrid-Wing Type**

Hovering No Yes Yes Yes Yes Small Size No Yes Yes Yes No

hour) >1 h 1 h >1 h >1 h >1 h

Cost effective Expensive Cheap Expensive Cheap Cheap Endurance High Low Medium Low High


**Figure 3.** Classification of UAVs in terms of altitude. **Figure 3.** Classification of UAVs in terms of altitude.

#### **4. Communication Methods and Architecture 4. Communication Methods and Architecture**

Today, one may see the variety of several drones being opted for commercial and domestic use. This is because they are cost effective and are controlled remotely from anywhere. In military operations, mostly the micro or miniature-type UAVs are used, but there are some major limitations in terms of size and weight. In Figure 3, tier II and III UAVs have several requirements such as being able to deploy sensors, and having a global positioning system (GPS), communication module, and efficient batteries. This is illustrated further in Figure 4. Although there are huge advancements being noticed in the field of drones or unmanned aerial vehicles (UAVs), at the same time, there are some limitations associated with the software and hardware support. Today, one may see the variety of several drones being opted for commercial and domestic use. This is because they are cost effective and are controlled remotely from anywhere. In military operations, mostly the micro or miniature-type UAVs are used, but there are some major limitations in terms of size and weight. In Figure 3, tier II and III UAVs have several requirements such as being able to deploy sensors, and having a global positioning system (GPS), communication module, and efficient batteries. This is illustrated further in Figure 4. Although there are huge advancements being noticed in the field of drones or unmanned aerial vehicles (UAVs), at the same time, there are some limitations associated with the software and hardware support. *Drones* **2022**, *6*, x FOR PEER REVIEW 7 of 21

**Figure 4.** Components of an unmanned aerial vehicle. **Figure 4.** Components of an unmanned aerial vehicle.

major components as follows:

Communication system;

 Drone airframe; Onboard controller; Payload capability;

Efficient batteries.

*Communication Methods*

The UAVs which are proposed for military operations have some advanced sensors

Discussing the airframe of a UAV, one must consider aspects such as aerodynamics,

When one discusses the communication aspect of UAVs, one is directed toward several integral subcomponents such as the communication protocols, the network type, and the UAV model itself. This means that with the change of communication method, one may induct the number of components and this will change the architecture of the system [27]. Many researchers have suggested several topologies and designs. This has been illustrated in Figure 4 along with the altitude range. One may opt for a different communication module and protocol as per the mission and the nature of task type [28]. Moreover, with the advent of 5G networks, several constraints such as data rate, latency, and

a lightweight structure, and stability. These can be some of the constraints to designing the UAV airframe. Moreover, the onboard controller is the main thing that maneuvers the drone. Therefore, it must be equipped with all essential sensors such as the accelerometer, gyro sensor, pressure sensor, GPS, and camera. While designing the drone, one should consider the factor of payload variation. In this way, the drone may carry some nominal amount of weight from one place to another [25]. Another important component is the communication system where the drone requires some communication equipment such as a sitcom, modem, or radio channel-based equipment. This will ensure the communication and control between a UAV and the ground control room (GCR). Lastly, the UAV must have a reliable power source that can help it to fly for a specific time to fulfil the task. Mostly, lithium batteries are considered as the main power source for these UAVs [26].

The UAVs which are proposed for military operations have some advanced sensors which are not accessible to ordinary people. These sensors enable drones to carry additional payloads. After studying different UAVs, one may subdivide UAVs into their five major components as follows:


Discussing the airframe of a UAV, one must consider aspects such as aerodynamics, a lightweight structure, and stability. These can be some of the constraints to designing the UAV airframe. Moreover, the onboard controller is the main thing that maneuvers the drone. Therefore, it must be equipped with all essential sensors such as the accelerometer, gyro sensor, pressure sensor, GPS, and camera. While designing the drone, one should consider the factor of payload variation. In this way, the drone may carry some nominal amount of weight from one place to another [25]. Another important component is the communication system where the drone requires some communication equipment such as a sitcom, modem, or radio channel-based equipment. This will ensure the communication and control between a UAV and the ground control room (GCR). Lastly, the UAV must have a reliable power source that can help it to fly for a specific time to fulfil the task. Mostly, lithium batteries are considered as the main power source for these UAVs [26].

## *Communication Methods*

When one discusses the communication aspect of UAVs, one is directed toward several integral subcomponents such as the communication protocols, the network type, and the UAV model itself. This means that with the change of communication method, one may induct the number of components and this will change the architecture of the system [27]. Many researchers have suggested several topologies and designs. This has been illustrated in Figure 4 along with the altitude range. One may opt for a different communication module and protocol as per the mission and the nature of task type [28]. Moreover, with the advent of 5G networks, several constraints such as data rate, latency, and coverage have been resolved. This advancement in communication technology not only improved these areas, but also helps to improvise the positioning and control of drones in several critical rescue and surveillance missions; for that purpose, people have used advanced flight controls and multiple sensors as well, along with camera placement, controlled and communicated in different ways as per their altitude as shown in Figure 5. These technologies are summarized later in Table 3 [29]. *Drones* **2022**, *6*, x FOR PEER REVIEW 8 of 21 coverage have been resolved. This advancement in communication technology not only improved these areas, but also helps to improvise the positioning and control of drones in several critical rescue and surveillance missions; for that purpose, people have used advanced flight controls and multiple sensors as well, along with camera placement, controlled and communicated in different ways as per their altitude as shown in Figure 5. These technologies are summarized later in Table 3 [29].

niques in brief [30].

**Technique Channel** 

**LTE-A** Up to 100

**Type of Communication**

Geostationary Earth orbit (GEO)

Medium Earth orbit (MEO)

MHz

**Table 4.** Communication based on satellite type.

**Elevation in Km**

> Up to 36,000

5000– 15,000

Due to high security concerns, the modern UAVs are controlled using the line-ofsight method at a low altitude, whereas for high altitudes, researchers have given preference to GPS and the beyond-line-of-sight (BLoS) technique. Table 4 describes these tech-

**UMTS** 5 MHz 700 to 2600 MHz 2 Mbps 10 Km 20–70 ms High

**Width Band Bit Rate Range Latency Mobility** 

GHz 6–54 Mbps 100 m 10 ms Low

MHz 50 bps - 10 ms Higher

**Frequency Doppler Gateway** 

15+ NA Low Very ex-

global 10–15 Low Medium Expensive High

Mbps 30 Km

Up to 1

**Handoff** 

Gbps 50 m - Ultra-

Gbps - Very High

**Cost**

pensive Highest

**Support**

High

**Propagation Path Loss**

10 ms Very High

**Figure 5.** Communication methods for high- and low-altitude levels.

**Figure 5.** Communication methods for high- and low-altitude levels.

**Table 3.** Classification of UAVs based on communication channels.

**5G** 2.16 GHz 57 to 64 GHz Up to 4

**LTE** 20 MHz 700 to 2690 MHz Up to 300

**Number of Satellites**

3, no polar coverage

8–20

450 MHz to 4.99 GHz

> **Satellite Life**

**Wi-Fi** 20 MHz 2.4 GHz to 5.2

**GPS** 2 MHz 1176 to 1576


**Table 3.** Classification of UAVs based on communication channels.

Due to high security concerns, the modern UAVs are controlled using the line-of-sight method at a low altitude, whereas for high altitudes, researchers have given preference to GPS and the beyond-line-of-sight (BLoS) technique. Table 4 describes these techniques in brief [30].

**Table 4.** Communication based on satellite type.


Tables 3 and 4 are very important for the readers to understand the significance of the several channels based on different bandwidths and satellites, respectively. Discussing Table 3, it communicates the different wireless communication methods, but at the same time, it shares that the communication will have a latency rate as well. The table also shares the channel width, band interval, and most importantly, the mobility support for the readers to design their drones accordingly.

Discussing Table 4, it communicates the type of communication based on satellite type. This will help the reader to see the elevation in kilometers, number of satellites, satellite life, handoff frequency doppler gateway cost, and most importantly, propagation path loss of each satellite communication [31].

#### **5. Utilization of UAVs in Different Domains**

The potential of UAVs and drones has been proved already, and this domain covers every type of utilization from personal usage to military purpose, as illustrated in Figure 6.

These UAVs can be more efficient in performing several missions if they are equipped with a camera, smart sensors, and processors. With these essential components, one may see 100 plus applications of drones mentioned by several researchers, such as in [32].

Please note that Figure 6 shows the areas where UAVs are utilized mostly in general, whereas Figure 7 shows the benevolent usage where UAVs are commonly used. The term malevolent usage shows the areas and specific domains where people have witnessed an incremental increase in utilizing the drones over the last few decades. Factors such as diligence, cost, mobility in the areas where humans are unable to reach, payload options, and risk compel everyone to use drones/UAVs. Now, depending on the type of drone, they may be used in a better way. Commonly, it is seen that the design of drones is dependent on the type of mission they perform in the field [33]. Thus, categorizing them all with respect to their domains may lead to understanding their architecture in a better way. This has been illustrated in Figure 7, where every domain has its own privacy and security needs [33,34].

global 3–7 High High Cheap Least

Tables 3 and 4 are very important for the readers to understand the significance of the several channels based on different bandwidths and satellites, respectively. Discussing Table 3, it communicates the different wireless communication methods, but at the same time, it shares that the communication will have a latency rate as well. The table also shares the channel width, band interval, and most importantly, the mobility support for

Discussing Table 4, it communicates the type of communication based on satellite type. This will help the reader to see the elevation in kilometers, number of satellites, satellite life, handoff frequency doppler gateway cost, and most importantly, propagation

The potential of UAVs and drones has been proved already, and this domain covers every type of utilization from personal usage to military purpose, as illustrated in Figure

**Figure 6.** Utilization of drones in several domains. **Figure 6.** Utilization of drones in several domains.

Low Earth or-

6.

bit (LEO) 500–1500 40–800

the readers to design their drones accordingly.

path loss of each satellite communication [31].

**5. Utilization of UAVs in Different Domains**

**Figure 7.** Malevolent and benevolent usages of a drone. **Figure 7.** Malevolent and benevolent usages of a drone.

#### **6. Security Threats Related to UAVs 6. Security Threats Related to UAVs**

[38].

UAVs offer several perks as the technology advances, but still, there are some constraints associated with privacy, security, and safety concerns [35,36]. Regularization and some important measures to license the utilization of drones is a very significant aspect. This limits unnecessary aerial photography. Most of the authorities in the world ensure this aspect and provide strict policies over uninformed aerial photography. If one discusses the network security point and the risk analysis, it is an admitted fact that the coverage is quite different as compared to any sort of wireless sensor network (WSN) or any mobile ad hoc networks (MANETs) [37]. The reason is because of the resource constraints, UAVs offer several perks as the technology advances, but still, there are some constraints associated with privacy, security, and safety concerns [35,36]. Regularization and some important measures to license the utilization of drones is a very significant aspect. This limits unnecessary aerial photography. Most of the authorities in the world ensure this aspect and provide strict policies over uninformed aerial photography. If one discusses the network security point and the risk analysis, it is an admitted fact that the coverage is quite different as compared to any sort of wireless sensor network (WSN) or any mobile ad hoc networks (MANETs) [37]. The reason is because of the resource constraints, as UAV-related coverage is broader and wider than WSNs or MANETs.

The framework that sets the rules to operate drones in any vicinity is known as au-

These drones are low cost and easily available in markets nowadays, and therefore, they are easy to use for any sort of criminal activity. Their ability to carry a wide range of external payloads make them more dangerous as it could lead to drones carrying any harmful chemical or explosive thing. Moreover, their ability to reach places where normal human beings cannot makes them more lethal because they can deliver anything without coming under anyone's notice [39]. It should be noted that security is not the only concern, but one may also see a safety concern if drones are flying over any populous place and, due to any number of faults, may crash, which can lead to several types of tragedies [40].

controller of a drone to operate as per the mentioned administrative rights, whereas it also shares some of the rigid authentication procedures for drones to protect the control of a drone so that it may not be diverted to any other unknown entity. Moreover, in case of any uncertainty or illegal activity by drone, one may easily track down the operator. This is done to limit illegal surveillance, cyberattacks, and privacy threats. Thus, several mechatronic engineering solutions have been presented to overcome these malicious activities

as UAV-related coverage is broader and wider than WSNs or MANETs.

The framework that sets the rules to operate drones in any vicinity is known as authentication, authorization, and accounting (AAA), which states several privileges to the controller of a drone to operate as per the mentioned administrative rights, whereas it also shares some of the rigid authentication procedures for drones to protect the control of a drone so that it may not be diverted to any other unknown entity. Moreover, in case of any uncertainty or illegal activity by drone, one may easily track down the operator. This is done to limit illegal surveillance, cyberattacks, and privacy threats. Thus, several mechatronic engineering solutions have been presented to overcome these malicious activities [38]. *Drones* **2022**, *6*, x FOR PEER REVIEW 11 of 21

These drones are low cost and easily available in markets nowadays, and therefore, they are easy to use for any sort of criminal activity. Their ability to carry a wide range of external payloads make them more dangerous as it could lead to drones carrying any harmful chemical or explosive thing. Moreover, their ability to reach places where normal human beings cannot makes them more lethal because they can deliver anything without coming under anyone's notice [39]. It should be noted that security is not the only concern, but one may also see a safety concern if drones are flying over any populous place and, due to any number of faults, may crash, which can lead to several types of tragedies [40]. These sorts of incidents have been reported often. One of the examples is when a UAV faced a collision with British Airways BA727, which was a passenger aircraft in April 2016. After looking over these incidents and issues, one may ensure below the mentioned public safety measures: *Drones* **2022**, *6*, x FOR PEER REVIEW 11 of 21 These sorts of incidents have been reported often. One of the examples is when a UAV faced a collision with British Airways BA727, which was a passenger aircraft in April 2016. After looking over these incidents and issues, one may ensure below the mentioned public safety measures: It is a high probability that a drone can be hacked or may deviate from its path due to heavy wind disturbance. Thus, there should be a reset option available which may turn the drone to a hovering condition only and help to gain the control back. These sorts of incidents have been reported often. One of the examples is when a UAV faced a collision with British Airways BA727, which was a passenger aircraft in April 2016. After looking over these incidents and issues, one may ensure below the mentioned public safety measures: It is a high probability that a drone can be hacked or may deviate from its path due to heavy wind disturbance. Thus, there should be a reset option available which may turn the drone to a hovering condition only and help to gain the control back. There are certain areas where drones may face signal jammers and, later, can be con-


The third safety measure is related to its design, as most drones have open propellers as shown in Figure 8. In case of uncertainty, these propellers may go off and may harm anyone nearby; thus, the safety design as shown in Figure 9 is necessary to avoid any harm during a crash. anyone nearby; thus, the safety design as shown in Figure 9 is necessary to avoid any harm during a crash. harm during a crash.

**Figure 8.** UAVs with open propellers. **Figure 8.** UAVs with open propellers. **Figure 8.** UAVs with open propellers.

**Figure 9.** UAVs with closed and safety propellers. **Figure 9.** UAVs with closed and safety propellers.

**7. Current Vulnerability Issues of UAVs**

**7. Current Vulnerability Issues of UAVs**

agreed permission [41].

agreed permission [41].

**Figure 9.** UAVs with closed and safety propellers. Lastly, there are some serious privacy concerns as well. Since UAVs on the market Lastly, there are some serious privacy concerns as well. Since UAVs on the market Lastly, there are some serious privacy concerns as well. Since UAVs on the market can easily be procured with high-definition cameras, this may lead to the recording of any

can easily be procured with high-definition cameras, this may lead to the recording of any

can easily be procured with high-definition cameras, this may lead to the recording of any

For these UAVs, unfortunately, there is neither a standardization of policies nor the

availability of wireless security [42–44]. This leads to several threats, as highlighted in Table 5. There are researchers who have addressed different types of cyber-attacks associated with the several types of UAVs in a pre-controlled environment [45–50]. Such practical validations include the crashing of drones with many parallel requests and modifying the request packet known as the buffer-overflow attack, whereas some researchers went for the cache-poisoning approach that leads to the shutdown of communication between the drone and GCR. In all conditions, most attacks occur to target the operating system or, in other words, the microcontroller of the drone [51]. Since there are huge advancements in the technology, UAVs have a high probability of experiencing such attacks, as shown in Figure 9 [52–58]. From these attacks, the most common attack is GPS spoofing,

For these UAVs, unfortunately, there is neither a standardization of policies nor the

availability of wireless security [42–44]. This leads to several threats, as highlighted in Table 5. There are researchers who have addressed different types of cyber-attacks associated with the several types of UAVs in a pre-controlled environment [45–50]. Such practical validations include the crashing of drones with many parallel requests and modifying the request packet known as the buffer-overflow attack, whereas some researchers went for the cache-poisoning approach that leads to the shutdown of communication between the drone and GCR. In all conditions, most attacks occur to target the operating system or, in other words, the microcontroller of the drone [51]. Since there are huge advancements in the technology, UAVs have a high probability of experiencing such attacks, as shown in Figure 9 [52–58]. From these attacks, the most common attack is GPS spoofing,

such as signal jamming, de-authentication. and zero-day attacks.

such as signal jamming, de-authentication. and zero-day attacks.

private property without permission. Due to this reason, Canadian Public Safety (CPS) stated that these UAVs are prohibited from flying over any property without mutually agreed permission [41].

#### **7. Current Vulnerability Issues of UAVs**

For these UAVs, unfortunately, there is neither a standardization of policies nor the availability of wireless security [42–44]. This leads to several threats, as highlighted in Table 5. There are researchers who have addressed different types of cyber-attacks associated with the several types of UAVs in a pre-controlled environment [45–50]. Such practical validations include the crashing of drones with many parallel requests and modifying the request packet known as the buffer-overflow attack, whereas some researchers went for the cache-poisoning approach that leads to the shutdown of communication between the drone and GCR. In all conditions, most attacks occur to target the operating system or, in other words, the microcontroller of the drone [51]. Since there are huge advancements in the technology, UAVs have a high probability of experiencing such attacks, as shown in Figure 9 [52–58]. From these attacks, the most common attack is GPS spoofing, such as signal jamming, de-authentication. and zero-day attacks.


**Table 5.** Summary of all current vulnerability issues in UAVs.

#### **8. Current State-of-the-Art Solutions**

The very first and significant thing to resolve the threats is to identify them first. Thus, Table 5 and Figure 10 classify these attacks. Moreover, there are several contributions which address these sorts of attacks along with the suitable measures [58]. In recent times, researchers have utilized machine learning approaches to demonstrate the intrusion detection system (IDS). Thus, machine learning (ML)-based IDS is one of the areas where researchers are still working to improve the results [59,60]. Blockchain is also among the most effective approaches for UAV/drone security and privacy [12]. This ML-based IDS technique is from the robust technique, and it is categorized into three kinds as mentioned below: *Drones* **2022**, *6*, x FOR PEER REVIEW 13 of 21 Rule-based IDS;

	- Signature-based IDS;

**Figure 10.** UAV attack vector with reported incidents. **Figure 10.** UAV attack vector with reported incidents.

Above are the major approaches for detecting threats that intrusion detection systems utilize to inform the operator in the ground control room (GCR). Rule-based threat detection is a new approach enabled by artificial intelligence (AI) [61]. In comparison to others, it is more reliant on technology and less on manual interaction. Signature-based detection works well for recognizing known threats. It uses a pre-programmed list of known threats Above are the major approaches for detecting threats that intrusion detection systems utilize to inform the operator in the ground control room (GCR). Rule-based threat detection is a new approach enabled by artificial intelligence (AI) [61]. In comparison to others, it is more reliant on technology and less on manual interaction. Signature-based detection works well for recognizing known threats. It uses a pre-programmed list of known threats and their indicators of compromise to operate (IOCs). An indicator of compromise (IOC) could be a distinctive behavior that typically precedes a malicious network attack, such

unusual behavior to generate alarms.

and their indicators of compromise to operate (IOCs). An indicator of compromise (IOC) could be a distinctive behavior that typically precedes a malicious network attack, such as

tabase of known IOCs or attack signatures to detect any suspicious behavior. Anomalybased intrusion detection systems, on the other hand, can alert you to unusual behavior. An anomaly-based detection system uses machine learning to educate the detection system to recognize a normalized baseline rather than searching for known threats. All network activity is compared to the baseline, which represents how the system ordinarily performs. Rather than looking for known IOCs, anomaly-based IDS simply detects any

To identify the false data injection attacks, one may use the rule-based approach. This is used to target the signal strength in between the UAV and ground control room (GCR) and can be useful for any sort of known attack, pattern, or technique only. Some research papers have proposed a signature-based IDS over drones [62] where they addressed bioinspired cyberattacks associated with air-born networks. Last, but certainly not least, is as file hashes, malicious sites, known byte sequences, or even the content of email subject headings. A signature-based IDS examines network packets and compares them to a database of known IOCs or attack signatures to detect any suspicious behavior. Anomalybased intrusion detection systems, on the other hand, can alert you to unusual behavior. An anomaly-based detection system uses machine learning to educate the detection system to recognize a normalized baseline rather than searching for known threats. All network activity is compared to the baseline, which represents how the system ordinarily performs. Rather than looking for known IOCs, anomaly-based IDS simply detects any unusual behavior to generate alarms.

To identify the false data injection attacks, one may use the rule-based approach. This is used to target the signal strength in between the UAV and ground control room (GCR) and can be useful for any sort of known attack, pattern, or technique only. Some research papers have proposed a signature-based IDS over drones [62] where they addressed bioinspired cyberattacks associated with air-born networks. Last, but certainly not least, is the anomaly-based IDS scheme which is used against jamming attacks [62]. The only limitation of anomaly IDS is the huge resource requirement.

Similarly, there are some researchers who have suggested some algorithm schemes with forensic methods to address the advanced and complex attacks. They are complex and difficult to identify [63–69]. With the help of forensics, both perpetrator and method of attack can be identified. With the identification of the attack type, appropriate countermeasures can be implemented to avoid any future incident. As per the survey of [70], between 2014 and 2017 incidents among airplanes and drones amplified from 6 to 93, which makes it very important for the authorities to address security and privacy issues for UAVs. Due to an increase in cyberattacks on drones/UAVs, the government needs to introduce strict policies and standards to minimize these concerns. With the popularity of UAVs among the civilian population, attacks and the illegal use of UAVs will likely proliferate. Civilian or domestic UAV countermeasures are divided into physical and local countermeasures, which are already proposed but still can be improved.

There are several survey papers that address the latest integration of UAVs into cellular networks and discuss the inference issues [71], as well as those, like this paper, that address the significant concerns related to the standardization and regulation of drones and their privacy. In addition to this, the manuscript focuses on the issues related to addressing these limitations while communicating from the drone to the ground control room (GCR). Some of the researchers proposed survey papers also on the quantity and quality of service requirements and discussed network-relevant mission parameters [72], which is unlike this paper that discusses the safety, privacy, and adaptability features of drones.

#### **9. Open Research Areas and Recommendations**

After studying the previous sections, it is noted that there is still a need to improve some of the areas associated with unmanned aerial vehicles (UAVs). These areas are very significant and one may address these concerns to enhance the utilization of drones [73,74]. One of the important areas is path loss, where one needs to propose the channel model to hold on to the communication at higher carrier frequencies, and even in the presence of tall concrete buildings. This area is in regard to the latency in the communication, which should be less than 1 millisecond and remains as an area of concern [75,76].

In addition to these areas, one may work over the reliability aspect, where one may improve the drones with ultra-reliable communication so that even with the increase in UAVs in the sky, the communication can be performed easily. It is noted that these drones have not been operated in the sky for a long time, which is because of the battery life. Hence, battery optimization for drones is also one of the areas where researchers may engage themselves to increase the flight time.

Last, but certainly not least, is the amalgamation of artificial intelligence and computer vision algorithms in a drone to improve the mobility of the drone without any collision. This will protect the drone in terms of data logging and security [77–84]. This manuscript

also suggests some of the recommendations that may improve the privacy and security aspects, such as the registration of drone licenses. This will ensure the authorities identify the specific drone that has created an inconvenience in the jurisdiction [85–91]. Moreover, there must be flying permits allotted after necessary training to limit the illegal utilization of drones in any activity.

Another recommendation is to educate the public about the legal and illegal usage of such autonomous unmanned aerial vehicles and the laws related to it so that if they witness anything around, they may easily report it. In any vicinity, there are some restricted zones; thus, the market drones must be operated based on a built-in map [92–95] as per the local regions. In this way, when any drone is forced to enter into any barred jurisdiction, it will automatically revert to the ground using the vertical take-off and landing (VTOL) mode. In terms of security tools, this paper proposes the machine learning-based IDS system [96–98] to improve the security infrastructure of UAVs, and lastly, there should be rigid multi-factor authentication methods that tackle the security threats easily.

There are several future recommendations to increase the standards for the security and privacy aspects of UAVs. These aspects are improved by proposing an approach which is based on a pairing certificate so that other strange entities may not easily connect or communicate with our UAV. Some of them are based on identification/authentication protocols [99–103]. To secure the drone more and make it less vulnerable, researchers have also used a symmetric searchable encryption method (SSE) [104] as well. Some researchers have proposed an internet of things (IoT) feature also for the same purpose [105–108]. In terms of identifying an unknown input observance, one may see an intelligent control algorithm that stabilizes the UAV in the presence of an unknown input [108–110] and devise it in trajectory and altitude levels to identify unknown system dynamics online by utilizing filtering manipulations that possess a concise structure, low calculation consumption, and asymptotic error convergence. In addition to this, the manuscript highlights the major domains and compares them with some of the latest review papers on UAVs for contrast. This is the significance of this article, which is seen in Table 6.


**Table 6.** Summary of all major domains along with reference list.

One can observe from the above chart that the first column lists the issues that have already been covered in the paper, whereas the first row lists the number of review manuscripts that state or debate the same theme. After reading through Table 6, one can find this article to be more thorough in determining the future answer quickly and effectively. A black dot in the table above indicates the articles in which the topics were directly mentioned.

A statement in support of integrating UAVs with contemporary trends of communication and a control system is developed by evaluating various research contributions linked to UAVs and communication aspects to identify the limits [104,111]. To get over the limitations, more research is still needed to examine the subtopics below:

• There is a need to address the area of high-speed mobility, as there are huge chances to hack the communication links through the ground control room or with neighboring UAVs.


Thus, with all the above recommendations and in contrast to these topics, our manuscript provides a detailed direction for future work.

#### **10. Conclusions**

The use of UAVs has increased dramatically, ushering in an era of autonomous systems and vehicles. These drones are quite important because they have many benefits for both civil and military concerns. However, with this rise in usage, severe privacy and security concerns are also evident. The most frequent reason why these drones are chosen in any sneaky assault is because they are readily available and inexpensive to obtain.

There have been numerous scientific contributions that have already addressed the countermeasures to these worries; however, there are still some areas that have not been addressed and can, therefore, still be exploited for any negative purposes, such as privacy and security issues. In this current, technological age, these two challenges cannot be disregarded. As a result, this review paper offers a thorough examination of these two pressing issues by providing a quick summary of the causes of each worry, as well as potential solutions. The existing solutions and a number of recommendations are presented in this study, which claims that the UAV drones can be enhanced if proper data integration, authentication, and accessibility factors are treated seriously.

**Author Contributions:** Conceptualization, G.E.M.A., A.L., R.J.M., V.S.A. and S.A.B.M.Z.; methodology, A.L.; software, G.E.M.A. and A.L.; validation, V.S.A., R.J.M. and S.A.B.M.Z.; formal analysis, A.L.; investigation, V.S.A. and A.L.; resources, S.A.B.M.Z. and G.E.M.A.; data curation, S.A.B.M.Z., V.S.A. and A.L.; writing—original draft preparation, G.E.M.A.; writing—review and editing, G.E.M.A. and S.A.B.M.Z.; visualization, V.S.A. and S.A.B.M.Z.; supervision, A.L., S.A.B.M.Z., R.J.M. and V.S.A.; project administration, G.E.M.A.; funding acquisition, S.A.B.M.Z. and G.E.M.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** The article processing charges were financed by the Research Management Center, Universiti Teknologi PETRONAS, Malaysia, under the Research Fund, which was supported by Yayasan Universiti Teknologi Petronas (YUTP), award number 015LC0-316.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** The authors would like to express their gratitude to the Centre of Graduate Studies (CGS), Universiti Teknologi, PETRONAS, Malaysia, and the Erasmus+ Program for giving them the opportunity to study abroad as exchange students at Telecom SudParis, France and conduct this research using the cutting-edge resources of that country's university. Finally, I would like to express my gratitude to the Department of Electrical and Electronic Engineering for offering me a position to pursue PhD studies through YUTP Funding, grant number 015CL0-316.

**Conflicts of Interest:** The authors declare no conflict of interest.
