Article
citation information:
Mammadov,
A., Sadigov, G. Implementation of complex solutions
for protection against cyber threats of electronic systems of unmanned aerial
vehicles. Scientific Journal of Silesian
University of Technology. Series Transport. 2025, 127, 155-164. ISSN: 0209-3324. DOI: https://doi.org/10.20858/sjsutst.2025.127.9
Aftandil MAMMADOV[1], Gurban SADİGOV[2]
IMPLEMENTATION OF
COMPLEX SOLUTIONS FOR PROTECTION AGAINST CYBER THREATS OF ELECTRONIC SYSTEMS OF
UNMANNED AERIAL VEHICLES
Summary. In the article, the
possibilities of applying complex solutions to protect the electronic systems
of unmanned aerial vehicles (UAVs) and other important processes from cyber
threats were considered. The current application of UAVs in various fields
makes the security of their electronic systems relevant. Aerial vehicles can be
subject to attacks by attackers who exploit physical, network, and software
vulnerabilities in the data exchange process. The article analyzes
such methods as data encryption, encoding methods, publishing fake data, and
protection against Global Positioning System spoofing attacks. The proposed
complex approaches are integrated into the existing systems of UAVs, allowing
to increase the level of their security. Complex solutions involve the
integration of a special encryption and coding block, as well as the
implementation of a module for comparing coordinates and broadcasting false
signals.
Keywords: unmanned aerial vehicles, electronic systems, cyber-attacks, encoding
and encryption, GPS spoofing, sensor, inertial navigation system, communication
system
1. INTRODUCTION
UAVs, which in the early days were used only for
military and observation purposes, have also begun to be applied in
engineering, scientific research and civilian areas due to the reduction in
costs and increased accessibility with the technology that has developed in
recent years [1]. UAVs use various active or passive sensors to obtain the
necessary information from these areas with high accuracy. UAVs with these
capabilities are capable of providing more accurate, sensitive, fast
information and analytical solutions than manned flight equipment or satellite
imagery.
While UAVs provide professionals with numerous
opportunities in the military, commercial, and civilian fields, their
management and information exchange face serious cyber threats. These threats
can allow criminal attackers to easily perform attacks against systems.
Therefore, it is very important to strengthen information exchange channels,
apply encryption and encoding methods.
This ensures widespread use in natural disaster
monitoring, military operations, agriculture, and logistics. UAVs are engaged
in the collection and transmission of sensitive information (for example,
intelligence, military, or strategic information). If this information is
intercepted by the enemy, serious security risks may arise [2].
In recent years, UAVs, especially those used
during wartime, have been exposed to numerous cyber threats. In addition, since
UAVs operate alone rather than in a group form, it becomes conveniently
possible to apply outside effects to them.
2. LITERATURE
REVIEW
Every
year, the types of cyberattacks and the possibility of threats against UAVs are
increasing. This is due to the fact that they become more functional depending
on the requirements that are pursued against them. Currently, as SDR (Software
Definition Radio) is widespread, the number of possible attack channels is
multiplying. As examples of cyber-attacks against UAVs, the following can be
highlighted:
-
GPS spoofing;
-
Interception of communication channels;
-
Malware attacks;
-
Denial-of-Service (DoS) attacks;
-
Using Firmware vulnerabilities;
-
Man-in-the-Middle (MitM) attack;
-
Theft of camera and sensor data;
-
Fake UAV attacks (Clone Attack).
Many
of these attacks took place in real, and some were executed for testing.
In
GPS spoofing attacks, an attacker manipulates the UAV's location information by
transmitting fake GPS signals to it. This can cause the drone to be misdirected
or deviated from the designated area. An example of such interference can be seen in
the 2011 incident, when Iran claimed that the United States had launched an
RQ-170 Sentinel unmanned aerial vehicle by spoofing GPS signals. Iran
manipulated the UAV with fake GPS signals and landed it in what it identified
as a friendly area [8].
The
communication channel between the UAV and the control center
is vulnerable to cyberattacks. By capturing this channel, attackers can change
the control of the UAV or steal sensitive information. During the analysis of
DJI drones in 2018, it was determined that communication could be intercepted
due to poor encryption of control signals [6].
By
embedding malware in the software or operating system of the UAV, its normal
operation can be disrupted. An attacker can disable the drone's sensors,
cameras, or other components through malware. In 2019, researchers in this
field, were able to exploit vulnerabilities in some UAV models
to install malware on the system and take full control of the drone [11, 12].
Cameras
and sensors on UAVs collect highly sensitive information. Attackers can steal
this information in real time and use it for their own purposes. In 2020, some
security researchers took advantage of the fact that drones are not encrypted
during the transmission of video data, proving that this information can be
easily stolen [9].
An
attacker can create a fake UAV (clone) and present it as a real drone, or
manipulate the exchange of information by moving to the location of the real
drone. For example, during hostilities, the attackers ensured the misdirection of
intelligence information by using fake UAVs that looked like real UAVs [7].
It
should be noted that more avionics systems are threatened during cyberattacks
on UAVs. Here, in particular, we can highlight the communication and navigation
systems.
3. ISSUE
SOLVING AND DISCUSSION
It is quite important to have a complex approach to ensure the complete
safety of UAVs. The proposed comprehensive approach includes two main methods:
1. The first of these approaches is the use of special
encryption, encoding, and authentication equipment at any stage of the
information exchange process and in the case of transmission through any
channel. This approach can allow preserving the confidentiality and integrity
of information. To accomplish this task, an encoding and encryption module must
be added to the standard structural scheme. As a central unit, the module will
control the exchange of information over all channels. Of course, it is impossible
to fully achieve a solution to the issue with the integration of this module
alone. In order to achieve the desired result, it is important to improve the
modules mentioned in the following
parts of .
2. If we look at the second approach, it is intended to
create and broadcast false signals in order to distract the effects that are a
threat. In order to strengthen security, it is possible to create fake access
points that the other party can detect and allow keeping the real data
transmission confidential. Many models of UAVs are known as Wi-Fi access points
and allow operators to control them by connecting. In the proposed new
approach, our module creates several access points that correspond to real UAV
parameters in different frequency bands or different channels, depending on
their purpose.
For the complex application of both of the above approach, a structural
scheme was developed using the SysML system modelling language (Fig. 1).
During the construction of this scheme, it is planned to implement a
complex method with the integration of as few blocks as possible. In order for
UAVs to perform the tasks, it is necessary to focus on mass issues.
3.1. Determination of element base
The application of special modules is necessary for the implementation of
this complex approach. Taking into account that the UAVs currently in use are
exchanged for more and more information, the elements selected must also meet modern requirements. Of great
importance are the image images taken by UAVs or the video data they reflect in
real time. It is important to take into account such moments and introduce
special processors and other devices. But on the other hand, it is necessary to
pay attention to weight and dimensions so that the UAV does not lose its
agility. Taking into account the stated, the analysis and selection of modules
presented in the structural scheme were carried out.
Fig. 1. Structural scheme of an integrated
approach to ensuring the safety of UAVs
When
selecting components, it is essential to consider their compatibility and
potential for integration with the existing systems used on UAVs.
3.1.1. Data processing module
The data processing module (edge processing) executes many operations. At
its core, this module provides data collection. It involves receiving
information from sensors, cameras, GPS receivers, and other sources. Then the
process of preparing the data for transmission is provided. Here, UART, I2C,
SPI interfaces can be used to connect sensors. Interfaces that can be saved do
not have built-in authentication or encryption mechanisms. These interfaces do
not guarantee that the data comes from the right device. During their use, the
risk of signal interception or interference from outside devices increases.
However, currently mentioned interfaces are considered to be suitable for
ensuring the security of coding and encryption block reflected in the
structural scheme.
Information enters this module from the receiver. The data processing
module consists of a controller and a processor. Depending on the specific
purpose of the UAV, different controllers may be used.
The authors examined the Pixhawk, CubePilot and MicroPilot MP2128
controllers with high tactical and technical characteristics. Pixhawk
controller supports Px4 and ArduPilot. The CubePilot controller is a
high-performance controller with reservation. The MicroPilot MP2128 is a
certified controller for professional and military applications [3].
3.1.2. Data analysis module
The data analysis module is designed for processing large amounts of data
coming from UAV sensors and communication systems. This module plays a key role
in ensuring the autonomy and efficiency of the
hardware, including increasing resistance to cyber threats. The mentioned
module analyzes the information received using the algorithms on the device in
real-time cross-section.
Data analysis allows the identification of cyber-attacks, including
attempts to seize control, interfere with data transmission, or apply malicious
software. Here it is
necessary to pay attention to the high productivity and the choice of elements
for performing complex calculations. Given the rapid development of technology
in the near future, NVIDIA Jetson or Intel Movidius processors will be suitable
for the mentioned operations. After eliminating noise and unnecessary
information in the data analysis algorithms, clustering should be carried out. In the process of
clustering, data is divided into groups to facilitate analysis. For specific tasks, object recognition and
prediction can be applied based on trained models.
Since the reconciliation of elements selected by electronics systems is
an important issue, this point is taken into account when selecting parts of
the data analysis module and other element base. The data analysis module is an important part
of modern UAVs, ensuring their reliable and safe operation even in difficult
conditions. The module transmits information in two directions. One of them is
the encoding and encryption block and the other is the comparison module.
3.1.3. Encoding and encryption block
Encoding and encryption block for Unmanned Aerial Vehicles is a critical
component for ensuring security, increasing data protection and resistance to
intrusion. This block is used both for communication and for data protection in
the flight system.
Various encryption technologies can be applied for UAVs with the proposed
block tool. If we mention the types of symmetric encryption (AES) here AES-256
is the most common standard in UAVs. This standard is high-performance and
optimized for application in small-sized UAVs. Asymmetric encryption (RSA, ECC)
is used for more secure key exchange. Elliptic Curve Cryptography provides the
same level of security as smaller keys and saves resources.
Various encryption technologies can be applied for UAVs with the proposed
block tool. If we mention the types of symmetric encryption (AES) here AES-256
is the most common standard in UAVs. This standard is high-performance and
optimized for application in small-sized UAVs. Asymmetric encryption (RSA, ECC)
is used for more secure key exchange. Elliptic Curve Cryptography provides the
same level of security as smaller keys and saves resources. In addition to those mentioned, the
completeness of the information must be ensured. For this, HMAC (Hash-Based
Message Authentication Code) technology is applied. This technology confirms
that the data sent has not been changed. Hash functions such as SHA-256 or
SHA-512 are used here [4].
Regarding coding protocols, it can be noted that it is intended that the
data coding method will be applied in protocols such as MAVLink for compact and
effective data transmission. Both high performance and flexibility are provided
when the implementation of coding and encryption block realization is carried
out with software and hardware. With software (Software-based), algorithms such
as AES, RSA, or ECC are applied at the previously mentioned software level.
These methods are more flexible, but inferior in performance to hardware-based
encryption. With hardware (Hardware-based), chips such as TPM (Trusted Platform
Module) or HSM (Hardware Security Module) which are used. This method ensures
that data is encrypted with higher performance and more securely. The
implementation of FPGA or ASIC-based encryption blocks is envisaged. It is
necessary that navigation, communication, and control information, as well as
the images obtained-in short, all-important information that affects the
security of the UAV, enter the coding and encryption block.
One of the drawbacks is the existence of resource constraints. Encryption and encoding should use the computing
resources of the device to a minimum. Because it is important to minimize
delays. Delays are unacceptable, as processes are carried out in real time. It
should be borne in mind that strong encryption and encoding algorithms require
more resources. The proposed systems must be resistant to radio-electronic
obstacles.
3.1.4. Comparison and generation block
A comparison and generation signal broadcasting block is intended to be
used as part of a system against attempts to seize control or fraudulent navigation.
This module tracks the true coordinates of the UAV and generates false signals
against external threats. The initial function of the mentioned block is the
comparison of the received, benchmark, and generated false coordinates. As a
benchmark, the most appropriate coordinate of the UAV is considered to perform
the assigned task. Benchmark coordinates tend to match the flight plan;
however, current coordinates may not always match benchmark coordinates. It
receives the current coordinates of the UAV through the built-in inertial
navigation system (INS) and GPS. It also compares this data with the benchmark
route stored in the system's memory. To confuse the enemy side, it creates and
transmits false coordinates, different from the current benchmark and current
coordinates. The signal is not broadcast when the error is detected. For
example, data is not sent to the transmitter if the created coordinate matches
the benchmark or current location.
The module can be executed on the example of general-purpose processors
with comparison algorithms, such as STM32. Because the block implements the
processing logic after obtaining GPS coordinates, INS data, and a benchmark
route (a planned route previously stored in memory). At this step, the
comparison algorithm comes into play. Taking into account the pattern of
movement at the output of this block, false coordinates are created, and
misinformation is sent to the transmitter (translator of false signals) of
false signals through interfaces such as UART, SPI, or CAN.
When emitting false signals and transmitting false object, there must
necessarily be a block of comparisons so that the coordinates of the fake
targets applied by chance and the real UAV are not the same. At present, the
management of the UAV and other tactical actions cannot be entrusted to
artificial intelligence, since the security of the information subject is not
fully ensured.
However, if the security required by this method is ensured and the
management issues go to a completely new stage, it will already be possible to
develop more flexible and highly maneuverable UAVs through the use of
artificial intelligence [10].
3.1.5. Data storage block
To create a system that is resistant to cyber threats, it is necessary to
adhere to the main triad in the exchange of information. The triad means that
information is confidential, integral and, available (CIA). For this reason,
information must necessarily be stored, and memory devices are available for
this. The memory device allows telemetry and reading of flight records from the
database only after identification. It is because of the memory device that
information remains accessible to the on-site management staff and other
services [5].
The main role of the memory device in the exchange of information is
shown in Fig. 2.
Fig. 2. Data
Flow Diagram (DFD) for the UAV
In the proposed module, local memory is used for temporary storage of
incoming data. The memory should be high-speed and reliable, for example, based
on eMMC or NVMe. The memory device can be used as a kind of redundant system in
the future. It can be used as an additional layer of protection to create
backup copies of key data and limit attackers' access to the system. The distribution block of protected
information manages the transmission of encrypted information. From here,
information is transmitted to control surfaces, autopilot and, if necessary, to
other channels.
4. ALGORITHM DEVELOPMENT
For the exact functionality of the proposed complex method, in addition
to the specific work algorithm of the modules and elements, there must be a
special algorithm of the general UAV. Through the special algorithm,
coordination will be established between the elements and systems, and the
information exchange will occur in accordance with the sequence we have
determined. The algorithm shown in Fig. 3 has been constructed for the
implementation of the proposed method.
To protect the UAV, especially its avionics systems, from cyber threats,
very effective information exchange must be implemented. When developing this
algorithm, it is necessary to consider several main factors. Information must
be distributed and transmitted quickly across all channels. Blocks that are
most likely to fail must be integrated in a special way. More precisely, the
re-entry of these blocks into the cycle must be organized as soon as possible.
Since current UAVs are functional, the data flow can be large. In addition, due
to certain reserves, the number of data transmission buses may be excessive.
Taking all of this into account, the development of the algorithm should be as
simple as possible. The simplicity of the algorithm will prevent the system
from being overloaded.
Fig. 3. Data Flow Diagram (DFD) for the UAV
The process of sequencing information received from transmitters and
other channels is reflected in the algorithm. After the integrated elements and
blocks, we can see the principle of operation of the system in this algorithm.
The received information is first sent to the edge processing block. UAVs are
equipped with various transmitters (GPS, accelerometer, gyroscope, barometer,
cameras, etc.). Altitude, speed, direction and position information is received
from the transmitters. The processing block collects the information received
from the mentioned transmitters and stores it in the correct format. Position
data is necessary for navigation and tracking the path.
To ensure that UAVs fly the correct route, the processing unit monitors
the position and movement of UAVs based on GPS data and other navigation
systems. This ensures that the flight plan is executed correctly. The
comparison and generation block analyzes information about the flight plan,
current coordinates and produces new values for transmitting coordinates. When
comparing and generating coordinates, this block will use the Haversine formula
and spherical trigonometric formulas.
(1)
(2)
(3)
Where:
: latitudes of the two points in radians;
: longitudes of the two points in radians;
: difference in latitude;
difference in
longitude;
R: radius of
the Earth (mean value is approximately 6,371);
d:
great-circle distance between the two points.
Spherical trigonometry deals with the relationships between angles and
arcs on a sphere, often used in navigation. Below are the basic formulas of
spherical trigonometry that are used in the generation of coordinates.
Law of cosines for distance:
(4)
Law of cosines for angles:
(5)
Law of sines:
Using these formulas, we can calculate distances and bearings between two
points on the globe.
The presented algorithm allows providing a high level of protection
against external interference. In addition, the integration of the coding,
encryption block, as well as the implementation of the comparison and false
signal generation module somewhat complicates the task of the system. In such a
case, a special algorithm is necessary to exclude errors.
5. CONCLUSION
Referring to the
conducted research, it can noted that the article touches on very important
nuances. The proposed complex approach combines encryption, coding operations,
and the dissemination of fake information during data exchange and can be
easily implemented. In this process, no serious changes are made to the main
systems of UAVs. Only the required modules are connected to existing equipment,
and interfaces are adapted. The implementation of this approach can bring UAVs
to a new stage of development. Many currently available digital solutions are
not used due to security problems. Thanks to the integration of the proposed
modules, it is possible to ensure security and remove existing restrictions. If
we minimize the risk of cyber-attack threats with the proposed methods, UAVs
can move to a new level of development in the future. In this case, we can
achieve more functionality of UAVs by applying artificial intelligence and
other new technologies.
References
1.
Chen C.L., Y.Y. Deng, W. Wang, C.H. Chen, Y.J. Chiu, C.M. Wu.
2020. “A traceable and privacy-preserving authentication for UAV communication
control system”. Electronics 9(1). DOI: 10.3390/electronics9010062.
2.
Hassanalian
M., A. Abdelkefi. 2017.
“Classifications, applications, and design challenges of drones: A review”. Progress
in Aerospace Sciences 91(4). DOI: 10.1016/j.paerosci.2017.04.003.
3.
Hong
Young-Woo, Dong-Young Yoo. 2024. “Multiple Intrusion Detection Using Shapley
Additive Explanations and a Heterogeneous Ensemble Model in an Unmanned Aerial
Vehicle’s Controller Area Network”. Applied Sciences 14(13): 5487. DOI: 10.3390/app14135487
4.
Cui J.,
Y. Chen, H. Zhong, D. He, L. Wei, I. Bolodurina, L. Liu. 2023. “Lightweight
encryption and authentication for controller area network of autonomous
vehicles”. IEEE Trans. Veh. Technol. 72(11): 14756-14770.
5.
Sion L., K. Yskout, D. Van Landuyt, A. van den Berghe, W.
Joosen. 2020. “Security Threat Modeling: Are Data Flow Diagrams Enough?”. 2020.
In: Proceedings of the IEEE/ACM 42nd International Conference
on Software Engineering Workshops. New
York, NY, USA, p. 254-257. DOI: 10.1145/3387940.3392221.
6.
Oded
Vanino, Dikla Barda, Roman Zaikin. November 8, 2018. “DJI Drone
Vulnerability”. Available at: https://research.checkpoint.com/2018/dji-drone-vulnerability/?utm_source.
7.
Sihag
Vikas, Gaurav Choudhary, Pankaj Choudhary, Nicola Dragoni. 2023. “Cyber4Drone:
A Systematic Review of Cyber Security and Forensics in Next Generation Drones”. Drones 7(7): 430. DOI: 10.3390/drones7070430.
8.
Simon
Niyonsaba, Karim Konate, Moussa Moindze Soidridine. 2023. “A Survey on
Cybersecurity in Unmanned Aerial Vehicles: Cyberattacks, Defense Techniques and
Future Research Directions”. International Journal of Computer Networks and
Applications (IJCNA) 10(5): 688-705, DOI: 10.22247/ijcna/2023/223417.
9.
Sofiane
Zaidi, Mohammed Atiquzzaman, Carlos T.
Calafate. 2021. “Internet of Flying Things (IoFT): A Survey”. Computer Communications 165: 53-74. DOI: 10.1016/j.comcom.2020.10.023.
10.
Tlili F.,
S. Ayed, L.C. Fourati. 2024.
“Advancing UAV security with artificial intelligence: a comprehensive survey of
techniques and future directions”. Internet of Things 27: 101281. Doi: 10.1016/j.iot.2024.101281.
11.
Zeng Y., R. Zhang, T.J.
Lim. 2016. “Wireless communications with unmanned aerial vehicles:
opportunities and challenges”. In: IEEE Communications Magazine 54(5): 36-42. DOI: 10.1109/MCOM.2016.7470933.
12.
Yu Z., Z. Wang, J. Yu, D. Liu, H.H. Song, Z. Li. 2024. “Cybersecurity
of Unmanned Aerial Vehicles: A Survey”. IEEE Aerospace and Electronic Systems Magazine 39(9): 182-215. DOI: 10.1109/MAES.2023.3318226.
Received 28.01.2025; accepted in revised form 10.04.2025
Scientific Journal of Silesian
University of Technology. Series Transport is licensed under a Creative
Commons Attribution 4.0 International License