Applied Engineering Letters Vol.8, No.

2, 80-90 (2023) e-ISSN: 2466-4847

COMMUNICATION CONVERGENCE FOR IMPROVEMENT OF THE

UNMANNED AERIAL SEARCH AND RESCUE MISSIONS
UDC:316.776:623.746-519
Original scientific paper https://doi.org/10.18485/aeletters.2023.8.2.5

Damir Nozica1 , Damir Blazevic1* , Tomislav Keser1

1
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Croatia

Abstract: ARTICLE HISTORY

Exponential progress in integrated circuits, best described by Moore's Law, Received: 2 February 2023
has enabled tremendous advances in applied computing. Today, more than Revised: 8 May 2023
ever, there are palm-sized embedded devices with computational Accepted: 21 June 2023
Published: 30 June 2023
capabilities millions of times greater than those of what was once the lead
computer on the Apollo 11 mission. Such levels of integration enable the
fusion of functionalities that were once unimaginable, or at least
KEYWORDS
impractical. Furthermore, today's existing technologies rely mostly on one
Communication convergence,
or two communication technologies to organize UAVs for efficient search aerial search and rescue, UAV
and rescue missions that largely do not utilize the communication swarm, UAV formation
convergence principle, thus omitting the potential for better search yield topology
and rescue success. This paper recognizes that niche where communication
convergence lacks its potential and presents a concept for the convergence
of Wi-Fi, Bluetooth, LoRa and/or satellite IoT communication technologies
to serve as an airborne communication infrastructure, a backbone generally,
and enables a swarm of unmanned aerial vehicles (UAV) to communicate
efficiently wherever there is no local terrestrial communication
infrastructure (such as GSM, Wi-Fi, digital radio, etc.). The concept was
elaborated and applied to a use-case localization application scenario (of
Wi-Fi enabled devices) for the purpose of search during rescue operations.

1. INTRODUCTION communication technologies such as GPS,

Bluetooth, and Wi-Fi to maintain their flight and
In related works, numerous authors have formation and perform advanced search and
proposed solutions that include different uses of rescue missions in remote areas to search for
UAVs to support the location of people during people with Wi-Fi enabled devices (smartphones,
search and rescue missions (SAR) [1-3]. The recent laptops, etc.). This innovative approach to search
development of UAV devices has enabled such a and rescue missions has the potential to
concept. The increase in UAV payload has enabled revolutionize the field. By utilizing unmanned
the installation of more and more electronic aerial vehicles equipped with advanced
components on board. This has allowed for greater communication technologies, emergency
capabilities in terms of communication, sensing, responders can quickly and efficiently locate
and data processing. As a result, UAVs have individuals in remote areas. The use of GPS,
become an increasingly valuable tool in SAR Bluetooth, and Wi-Fi allows for precise tracking
operations, allowing for more efficient and and communication between the drones, ensuring
effective searches in difficult terrain or hazardous they maintain their formation throughout the
conditions. In the research paper [4], the authors mission. In addition to locating individuals with Wi-
hypothesized that it would be possible to deploy a Fi-enabled devices, these drones could also be
swarm of unmanned aerial vehicles using used to survey disaster zones and provide real-

*
CONTACT: Damir Blazevic, e-mail: damir.blazevic@ferit.hr © 2023 Published by the Serbian Academic Center
 D. Nozica et al. / Applied Engineering Letters Vol.8, No.2, 80-90 (2023)

time data to aid in rescue efforts. While there are the problem statement of current approaches and
still challenges to overcome, such as battery life their benefits, or lack thereof. A proposed
and weather conditions, the potential benefits of methodology for the communication convergence
this technology are immense. It could save of currently available communication technologies,
countless lives and improve response times in with emphasis on their uniting (converging), is laid
emergency situations. In addition, authors in out in Section 3, along with the resolutions and
research [5] presented Wi-Fi communication proposals for efficient data-path optimization and
technology that was used to ping and find layer topologies in data exchange and its flow. A
terminals, with great success. By including search and rescue mission using such a formed and
additional communication technologies, such as controlled swarm of UAVs is the subject of Section
Bluetooth or Bluetooth mesh technology [6], it can 4, which applies the findings and solutions and
be used to synchronize UAVs and share data, and then conducts application analysis. Furthermore, a
in synergy with GPS technology that is used to discussion is given as a post-result analysis, where
determine the location of each UAV, a tremendous it is additionally discussed on the proposed use of
localization potential is achievable. Furthermore, technology by the authors in the paper [7], where
the use of this technology can also enhance the it (LoRa) can be employed to utilize other available
safety of UAV operations by providing real-time communication technologies (Satellite IoT network
monitoring and control. It can also improve the using Low Earth Orbit satellites) to reach CC while
efficiency of UAV missions by enabling multiple eliminating limitations when using such technology
UAVs to work together seamlessly. The authors of in terrestrial setups. The research also presented a
those studies have presented some of the comparative financial calculation when using the
challenges as well as solutions to the problem in concept in real cases against the price of an
the current research. The research is largely based average SAR mission.
on the experiment presented by research [5],
which provided a proof of concept using an ALIX 2. RELATED WORK
x86 system board to ping and detect Wi-Fi devices
at an effective range of 240 m and up to 15 m The development of UAVs and their widespread
height (an optimal height). In the study [4], the availability have laid a good foundation for the
authors propose an arbitrary search pattern and development of new products and applications
UAV formation but do not propose any inter-drone that address niches that were previously unknown.
communication or real-time result uploading One such application is the use of UAVs in search
solution, so this search mission cannot be used in a and rescue missions, which has proven to be cost-
time frame of up to one minute since their effective and efficient. Additionally, the use of
proposed approach requires manual data UAVs in SAR missions reduces the risk to human
downloading once all UAVs have landed or life by eliminating the need for human rescuers to
returned to the initial location. The proposed UAV enter dangerous or inaccessible areas. Such UAVs
flight path was predefined off-site, a concept that are often equipped with multispectral imaging,
the authors of this paper will maintain. In this thermal imaging, surveillance, and other
paper, the authors propose an additional extension applications that are of use in a given situation, as
of the concept presented in study [5] and extended described in research [8]. Furthermore, UAVs can
in study [4], in such a way that the UAVs will be cover large areas quickly and provide real-time
able to communicate with each other by applying data, making them ideal for disaster response and
the principle of convergence of communication search and rescue operations. The use of UAVs in
technologies. In addition, they will be able to these missions has also been shown to increase
communicate with a command center (CC) in near the chances of finding survivors and reduce
real-time once a target is detected, regardless of response times. Finally, among these UAV
the UAV's location or possible (or lack of) applications, the search and rescue niche has been
terrestrial communication infrastructure. With the subject of numerous studies and is the main
defined upgrades, the proposed technology could concern of this paper.
greatly enhance search and rescue efforts. Several authors mentioned the use of Wi-Fi
The article is structured as follows: A detailed technology to locate Wi-Fi-enabled devices as a
insight into UAV use, formation, and control for method for finding stranded individuals [9–11].
on-terrain surveillance missions is given in Section Some authors focused on the fact that a targeted
2, along with the state of the art on the survey and device (a device of interest) regularly sends pings
81
 D. Nozica et al. / Applied Engineering Letters Vol.8, No.2, 80-90 (2023)

when searching for access points and wanted to presented without obvious limitations. In research
exploit this technological fact to locate such a [10], a similar consideration is made (based on the
device (e.g., passive method). The work of research passive method), but the author adds an
[4] clearly distinguishes between active and interesting layer of research that focuses on micro-
passive Wi-Fi search technology, a concept that is location detection and triangulation of detected
not widely used among authors of related work. signals using a multi-pass method of UAVs. This
The difference lies in the initiation of approach works well for the micro-location of the
communication between the drone (UAV) and the target device but reduces the flight time of the
target device, making the chosen method active. UAVs and their overall coverage of the search area.
As previously mentioned, the majority of authors In research [9], an innovative approach to search
used a passive approach, assuming that the target pattern and localization was presented by the
device would send an active ping, which a drone- authors. Emphasis was also placed on the
mounted communication array would then receive technology used to build the communications
and process. This passive method was flawed system, as well as additional considerations for
because the UAV's time in the air (and time over a performance versus power requirements and area
given point in space above the targeted Wi-Fi coverage using commercially available drones and
device) is limited, so the target device was devices. This was another example of a passive
expected to be on at the right time. Sending a ping search method, where the communication field
at the right time was a risk factor that could make was divided into two segments (two Wi-Fi maps)
search and rescue impossible or significantly for better communication with the target device.
degrade search results. Instead, as suggested by When searching outside the Wi-Fi spectrum, the
research [5] and accepted by the authors of research presented mainly cases where the search
research [4], the active method should be used. was in the visual or infrared (thermal) spectrum
The active method assumes that an active ping (apart from the agricultural application), adding
signal was sent from the communication network beacon integration and cross-flight search patterns
to the drone (UAV). It could and would be received to better locate the person (the example of
by the targeted device, which would then respond avalanche search and rescue). An example of such
to this ping. This method ensures an immediate integration was presented in the study [16]. The
response from the targeted device once it is within study also highlighted the importance of real-time
range of a pinging device (UAV). Since the Wi-Fi communication between the search team and the
standard provides for 11 (14) available channels in person being rescued. Additionally, the study
the 2.4 GHz spectrum, a given device can use one emphasized the need for proper training and
of the channels at any given time. For this reason, equipment for search and rescue teams to
a method has been described in research [4] to effectively carry out their tasks.
ensure that a ping drone always sends the correct Among the research presented, the authors
ping on the correct channel so that the target could not find any related work dealing with the
device is able to send a response and be processed. active Wi-Fi search method in combination with
As presented in research [7, 12-15], LoRa the geographically secure communication method
technology could be used in some cases to that would ensure the delivery of useful data to
successfully detect a target device or to transmit the CC (Command Center) independent of the
information between different elements of the terrestrial communication network infrastructure.
communication system. Like the previous research, This niche has been used to propose a novel
they also used the passive method for Wi-Fi solution that uses communication convergence to
scanning and relied on LoRa technology for long transfer information from one endpoint to another.
range penetration to provide successful detection
information. However, the mentioned study does 3. COMMUNICATION CONVERGENCE
not take into account a possible complex METHODOLOGY AND UAVS FORMATION
geomorphology that may hinder the LoRa signal TOPOLOGY IN LOCALIZATION AND RESCUE
from being sent to a listening station on the MISSIONS
ground, as UAVs tasked with detection cannot fly
too high although some authors suggest a higher The aim of this research was to investigate
flight altitude). In most of the research papers related work and technological solutions,
mentioned above, a clear case of data processing especially in the last 2-3 years, trying to establish a
and logic for detection and transmission of data is commonality between the discovered research in
82
 D. Nozica et al. / Applied Engineering Letters Vol.8, No.2, 80-90 (2023)

terms of communication technologies used and proposed concept in terms of communications is

finally proposing new communication technologies shown in Fig. 1.
for maintaining end-to-end communication. A
focus was placed on the active vs. passive search
method in the edge layer and on the use of LoRa
technology in the fog and cloud layers, with a
critical overview of the approaches used in the
referenced research. An innovative setup was
designed and tested by selecting current products
(UAV) and introducing new communication
technologies (Satellite IoT for the cloud layer, LoRa
for the obstacle-free fog layer) to transform the
previously known setups from related research
into an innovative setup that enables an
uninterrupted communication flow without the
existing terrestrial communication network. Finally,
the research results were presented using known
factors (cost of an average SAR mission, number of
missions, etc.) to prove that the presented
innovative solution is sound. The proposed
solution was based on the best practice of active
Wi-Fi scanning and the experimental solution
presented by the authors in research [5] and
swarm upgrades presented by the authors in
research [4], adding new communication layers to Fig. 1. High-level conceptual convergence design -
communication overview with defined layers
ensure up-to-the-minute propagation of success
detection information in a layered approach using
The edge layer, as presented in the research of
communication convergence methods regardless
[5] and [4], was used as it is, and this research did
of geomorphology (terrain). This paper does not
address the actual flight speeds of UAVs, as this not contribute specifically to this layer. Since the
publication of the cited work, new technology has
topic has already been discussed in detail. Usable
become available that allows for some
flight speeds were defined as 5 m/s in the
detection portion of the flight and as fast as improvements in this segment of the
possible in the non-detection portions of the flight. communication field. As defined in [5], the target
The proposed solution is based on the use of device was or could be tuned to one of the 11 (or
14) Wi-Fi channels in the spectrum. For this reason
multiple swarms of UAVs (horizontal) and divided
and to maintain adequate UAV movement speed, a
into three communication layers (vertical), which
multimodal ping device should be placed on a UAV
together form a communication convergence:
so that it sends a ping on all 11 (14) Wi-Fi channels
• Edge Layer; consisting of a ping and a
at once or cycles through them in the manner
detection UAV that flies a predefined search
described in research [5]. This behavior greatly
pattern for the targeted Wi-Fi device.
increases the chances of positive identification of a
• Fog Layer; consisting of the detection UAVs
Wi-Fi target device in a shorter time. The target
capable of transmitting status messages via
device would respond to the ping signal and reply
LoRa technology to a master UAV at the site
on the same channel. At this point, the
and via Bluetooth or Bluetooth mesh
communication array on the board of the
technology to other UAVs in the pattern.
detection group drone (UAV) could detect the
• Cloud Layer; consisting of master UAVs
signal and relay the confirmation to the main
capable of sending status messages to the
drone. Communication within the swarm could be
Internet Control Center via a low-earth orbit
maintained via a Bluetooth communication
IoT satellite network.
protocol if needed, especially if faster coordination
between UAVs is required (e.g., to maintain
Other UAV functionalities, control methods, or
formation or to reduce or extend a separation
communication via proprietary protocols were not
distance since each UAV follows a pre-planned
considered in this paper. An overview of the
83
 D. Nozica et al. / Applied Engineering Letters Vol.8, No.2, 80-90 (2023)

route but actual field conditions, such as wind or group in formation flight and send ping signals to
gusts, may be different for different UAVs). the detection area, preferably from a higher
Upstream communication between the detection altitude to enable better signal reception.
group UAV and the master UAV was maintained
In the experiment conducted by the authors in
using LoRa communication technology. LoRa takes
research [5], an empirical limit of about 15 m
advantage of long-distance transmission and a
above the ground was set for the detection drones
relatively low speed and content capacity. In
to achieve successful detection in the range of up
addition, LoRa is limited to 1% transmission time
to 240 m using the system board ALIX with a 500
by a legal standard.
MHz CPU. However, this limitation did not apply to
The communication layer between the UAVs in
the ping group, since its main purpose was to
the detection group and the master UAV was
transmit a wake-up or ping signal, which can be
considered the fog communication layer. This
done from a higher altitude without the 15 m
means that additional computational and
restriction. In the research presented by research
information processing has been moved to this
[5], the authors also addressed the different Wi-Fi
layer. In addition to positive detection information,
channels. These channels were limited and had to
the fog layer devices can also exchange location
be managed.
data or mission completion data that can be
Theoretically, the targeted Wi-Fi-enabled
uploaded to CC at any time or upon request via a
device (at the time of the drone overflight) could
master UAV. All communication in this layer occurs
be set to any Wi-Fi channel. At this point, it was
between UAVs and does not reach any external
important to explain that the experiment and
receivers. For this reason, no external or terrestrial
research were based on the 2.4 GHz spectrum,
infrastructure was required.
since most (if not all) devices still had a (basic)
Upstream of the master drone was a cloud
frequency range enabled, and signal propagation
communication layer. A master UAV transmits
on 2.4 GHz is much less affected by obstacles than
data to a control center via a satellite uplink. This
on 5 GHz. The ping group drone must either be
method used as a use case (UC) for search and
equipped with a multi-channel communications
rescue missions, is often conducted in remote
array to ping all channels simultaneously (for a
areas without any communications infrastructure.
higher success rate), or be able to cycle through
Since the success of the mission depends on the
the channels relatively quickly to ensure that a
timely and accurate location of those in distress
ping signal was sent to all channels. As described in
(targeted Wi-Fi-enabled devices), the information
research [4], the flight path of the ping group (as
about successful identification should be
explained in research [17] in a different but
transmitted without delay and over a channel that
relevant area) could be pre-determined in the
ensures connectivity regardless of location. Due to
control center when planning the site flight.
the cost, mass, and complexity of electronic
Optimizations were possible and are not part of
devices, only one main drone was equipped with a
this research. A detection group is where useful
satellite uplink communication array. Once
information is received. As illustrated in research
information is successfully sent to the IoT satellite
[5], a device receiving a ping signal would respond
network, it is relayed to CC within minutes.
and attempt to make contact.
This event enables detection by the UAV
3.1 Edge layer structure and topology detection group. For the same reason that the ping
group UAV must be able to transmit on any Wi-Fi
In the entire formation and communication channel, a detection group UAV must be able to
topology, there were two groups of UAVs that receive such a signal. Ideally, the UAV should be
represented the edge layer of communication. equipped with multiple Wi-Fi modules. Since the
Edge layer UAVs are arranged into: payload of a drone of this size is limited, cycling
• Ping UAV (group); through channels with a single module was an
• Detection UAV (group). acceptable method, but one that carries risks and
These groups, as presented by the authors in leaves room for future upgrades and
research [4], had different functions within the enhancements to this concept. This means that in
presented solution, which included active search future experiments, the detection group drone
for Wi-Fi enabled devices. The ping drone or a could be equipped with a different type of
group of drones was to fly ahead of the detection communication array for the edge communication

84
 D. Nozica et al. / Applied Engineering Letters Vol.8, No.2, 80-90 (2023)

layer and may have different payload system (date, time, exact location, UAV ID) and
requirements on the drone itself. A visual transmit this data packet to the master UAV using
representation of the edge layer can be seen in Fig. LoRa communication technology in the fog
2. communication layer. This information could also
be shared between UAVs using Bluetooth
technology in the same layer to keep it safe or
upload redundant information when needed. A
LoRa transmitter could be placed on the top of the
UAV to provide a clear line of sight to the master
UAV. This communication method followed the
logic presented in the study [7]; however, it
eliminated the challenges associated with available
terrestrial receiving towers or problems reaching
those receiving towers when the UAV is flying low
Fig. 2. Edge layer structure and topology and is hindered by local terrain, which was an
obvious drawback of the solution proposed in the
The basic organization and utilization of the study [7]. Based on the conducted research, it was
messaging/pinging and acting in edge proposed that the LoRa signal with positive
communication layer can be summarized as: detection data be sent to the master UAV to be
• Ping/pinging; conducted by a Ping UAV; positioned high above all detection groups of UAVs.
• Reply; data management given/received In this way, communication transmission to the top
by the target device(s); of the fog communication layer is guaranteed since
• Detection; actions for sensing and there are no obstacles between the detection
detection of objects, managed by UAVs and the master UAVs. A message reception
Detection UAV. acknowledgement method can be used to
complete a positive information transfer between
3.2 Fog layer structure and topology units. This is native LoRa functionality that can be
used. Aside from positive detection data, the UAVs
A fog layer consisted of detection group UAVs in the detection group can also send position
and master UAVs. The detection group of UAVs reports or percentages of path completion reports
was discussed in more detail in the previous across the fog layer to the master UAV, which can
chapter. A master UAV was a dedicated UAV then forward this data to CC. A visual
responsible for collecting data from a group of representation of the fog layer is shown in Fig. 3.
detection groups. This (master) UAV was equipped
with a LoRa receive communications array at the
bottom of the UAV and a satellite uplink
communications array at the top. For this
communication layer, only the LoRa
communication array on the master UAV is
relevant. The UAVs in the detection group were
equipped with a LoRa transmit communications
array on the top of the UAV. As a communication
technology, LoRa can provide an extended range
that is more than sufficient for the application
described in this research (up to 15 km at line-of- Fig. 3. Fog layer structure and topology
sight, as described in [18]). The previously
mentioned LoRa-related research used various The following structural and functional
devices (such as Arduino or Raspberry boards) to topology represents the basic application and
incorporate LoRa communication technology. The functional role of the fog communication layer:
authors of this paper proposed to use the same or • Transmission of information; for successful
more industrial products as suggested in research data detection, management and routing;
[18]. In the event of a positive detection, the UAV • Status update; for data management and
would combine the data received from the validation;
receiving array with the data from the internal

85
 D. Nozica et al. / Applied Engineering Letters Vol.8, No.2, 80-90 (2023)

• Synchronization between the units of the and future revisions may address a broadband
recognition group or the units of the ping communications channel transmitted on a satellite
group(s). network capable of carrying more data.
Swarm IoT technology [19] provides an option
3.3 Cloud layer structure and topology to use low-power and lightweight aerial vehicles to
send data to the LEO-based IoT satellite network. A
Elements of the cloud layer were master drones visual representation of the cloud layer can be
(UAVs) and satellite communications infrastructure seen in Fig. 4.
(and further via the Internet to CC). In the paper
presented by research [7], a terrestrial LoRa
receiving station (or person) was expected and
required to maintain communication between the
UAV and the CC. Although the authors did not
refer to the communication layer used as the cloud
layer, they proposed to use terrestrial LoRa
reception to receive information from the UAV.
The hypothesis previously stated was that there is
no terrestrial GSM network or other means of Fig. 4. Cloud layer structure and topology
digital communication over long distances. Since
search and rescue missions are conducted in the Briefly, the cloud communication layer manages
most remote areas where there is no following tasks and actions in UAVs group/swarm:
communications infrastructure and (perhaps more • Transmission of information; for successful
importantly) where terrain configuration could
data detection, management and routing;
pose a challenge to propagating an
• Status update; for data management and
electromagnetic signal from the UAV to a
validation;
terrestrial station or person, another method of
• Receiving commands from CC via the
communication must be secured.
satellite network.
This research proposes a cloud communication
layer and an upstream satellite communication
4. COMMUNICATION CONVERGENCE IN UAVS
method. Master UAVs are potentially larger
FOR SEARCH AND RESCUE MISSIONS –
aircraft that can perform multiple tasks. For the
APPLICATION AND ANALYSIS
purpose of this research, these UAVs would collect
positive identification data sent by the UAVs in the
In search and rescue missions it is of great
detection group and transmit it to the Internet via
importance to do the sweep search of targeted
IoT satellite communication methods such as the
terrain as fast as is possible. Right choice of
Starlink SWARM IoT Network or a similar low earth
searching formation is crucial, and that is applied
orbit satellite communication network (LEO). A
also in case of UAVs use for search and rescue. The
master drone was planned for each of several
intended range for conducting search missions
search areas and detection groups. Individual
depends on the selection of UAVs and their
experimentation or a use case (or budget) can
particular characteristics (flight time, speed, range,
define the exact need and setup. LEO satellite
total flight distance, service ceiling, flight
networks provide better connectivity
conditions, etc.). In all the mentioned research
characteristics than traditional geostationary
papers, the authors proposed or conducted
satellite networks. The higher speed and lower
experiments with different commercially available
latency provide better conditions for the current
UAVs. The common features are a flight time
use case but also for future enhancements, such as
between 30 and 45 minutes, a usable speed at
streaming thermal imagery, etc.
detection of about 5 m/s, and a maximum speed in
Although the master drones could be larger and
the range of 18-20 m/s for the non-detectable
have a longer flight time, it was imperative to
parts of the flight. These numbers are used in this
maintain the lowest possible load to ensure
paper to create as comparable a scenario as
maximum flight autonomy. To this end, the basic
possible.
IoT facility was selected from the range of
As larger UAVs become available at the prices
narrowband IoT products within the LEO satellite
mentioned in this paper, these numbers may be
services. This was the limitation of the research,

86
 D. Nozica et al. / Applied Engineering Letters Vol.8, No.2, 80-90 (2023)

revised in the future. For UAVs performing ping drone, another drone was planned. The UAVs of
and detection functions in the boundary layer, the the detection group have a maximum lateral
maximum flight time of 45 minutes proposed in distance of 4.25 km from the master UAV and a
research [20] was used. The flight speed is 5 m/s height difference of up to 2 km, resulting in a
for detection and 20 m/sec for a flight without maximum total distance of 4.7 km, which is within
detection. Commercially available UAVs for ping all LoRa line-of-sight parameters of the cited work,
and detection functions were available for under even for lower-spec devices, ensuring connectivity
1000 EUR/piece at the time of writing [20]. in the fog layer.
Table 1 lists the approximate area coverage
when no crosswinds or turbulence are expected in
the interest of a search mission of 4 square areas
of about 3x3km each side by side.

Table 1. Flight times and distances for ping and

detection UAVs according to UAVs with the declared
maximum flight time of 46 minutes
Category Are Area Area Area Master
a1 2 3 4 UAV
Distance from
the control
Approx.
center to the Approx. 1 km
4 km
start of the
search area
Flight time
and distance 1 1 3.5 3.5 Appx 4
to a specific min min min min min Fig. 5. Formation width and flight speed
search area
Flight time in
30 30 30 30 Appx 30 The master UAV maintains a central position
a specific 2000 m above the ground. This altitude ensures a
min min min min min
search area
Search area 9 9 9
clear line of sight between the acquisition group
9 km2 - and the master UAV, as the acquisition group has a
coverage km2 km2 km2
Total search lateral coverage of 1000 m with a lateral distance
Approx. 36 km2 -
area coverage between the units less than 200 m, ensuring
Total search Bluetooth 5.3 [21] connectivity for redundant
area Approx. 6x6 km -
dimensions
transmission of acquisition events. Fig. 6 illustrates
Flight time the search area.
4.5 4.5 8
back to the 8 min 42 min
min min min
control center
Total flight 35.5 35.5 41.5 41.5
42 min
time min min min min
Spare time 10.5 10.5 4.5 4.5 4

The search pattern is taken from research [4].

The control center is the launch position for the
UAVs and is located (arbitrarily) 1 km from the
beginning of the search area.
Fig. 5 shows a top view of the search area and a
UAV swarm formation. According to research [4],
this formation was sufficient to cover ping and
detection at the indicated latitude. Since ping
drones can operate from higher altitudes, one ping
drone was sufficient.
A search area is represented in the form of 2x2
boxes, each 3x3 km in size. A theoretical evaluation
was performed according to the specifications of Fig. 6. Example of the search area
the UAVs used. For the function of the master

87
 D. Nozica et al. / Applied Engineering Letters Vol.8, No.2, 80-90 (2023)

Table 2 lists the required investments for UAVs, capacity allows (the simple calculation already
spare parts, IT parts, services, battery cycles, and takes into account spare parts for maintenance, as
the average cost per square kilometer. shown in Table 2). The proposed search pattern
Table 2. Total cost per km2 searched considering that
theoretically allows a swarm of UAVs to be
the initial investment is less than 30.000 USD including launched from the CC site and reach the beginning
all needed IT equipment of the search zone, which is about 1 km away and
has a size of 6 x 6 km. 16 ping and detection UAVs
Category Title Values net or average would cover this area in about a 30-minute flight
Cost of UAVs* Appx 20.000 USD and return to the launch area within 45 minutes of
Cost of spare parts* Appx 5.000 USD launch. After changing batteries and loading new
Cost of IT parts, [18] Appx 4.200 USD
Cost of satellite Appx 1000 USD /year
trajectories, the same group of drones can
subscription, [19] continue to the next segment, covering at least 36
< 30.000 USD for products km² per hour. This means that an entire urban area
Total Appx 1000 USD for of Vienna (414 km²) can be covered in less than 12
services/year hours with just 16 drones. Since each search and
Investment per km2 Appx 861 USD for a single use
Average battery cycle Advertised 250 cycles; real
rescue mission is divided into two logical segments
appx 150 cycles (1: search and 2: rescue), in order to start the main
Minimum cost per km2 861 USD/150 cycles = 5.74 USD activity (rescue), it is essential to perform and
complete the search function as quickly as
* Obtained from [20] – Although this paper does not propose a
specific brand and model, an estimate was taken from the pool of the
possible. The presented solution enables much
most well know manufacturers with model DJI Mini 3 Pro for UAVs faster detection of target devices and can lead to a
and DJI Mavic 3 Classic for the master UAV. faster indication of the location of a missing
person. The information about the positive
5. DISCUSSION detection event must be transmitted to the CC as
soon as possible and in such a way that the
As the authors point out in a study [22], in communication infrastructure used reduces the
1912, search and rescue missions were conducted possibility of communication loss to the lowest
in Yosemite National Park over a 10-year period, possible value. For this reason, it is of utmost
helping over 2300 people. The average mission importance to ensure a suitable communication
lasted 5 hours and involved 12 people, at a cost of infrastructure, as presented in this paper. A novel
$4400, totalling $8.4 million. Since this research methodology for communication convergence has
focused on events during the period from 1990 to been presented. It ranges from Wi-Fi for pinging
1999, this figure, adjusted for inflation, is and detecting target devices, to Bluetooth for
approximately 17.35 million USD (calculated for possible synchronization between UAVs, LoRa
the year 2023), giving an average value of almost communication technology for transmitting
9100 USD. The above figures are an average, and important information and communication nodes
some of the missions of SAR were marginal, e.g., (master UAV), and a low earth orbit IoT satellite
the search for a missing person lasted several days communication network for extracting information
and more than 200 people were involved. In from remote areas to the Internet and back to the
situations such as this (uncertain whereabouts of a control center.
missing person), UAV-assisted SAR missions can be
of great benefit. Since today's UAVs for the above 6. CONCLUSION
functions and purposes each cost approx. 1000
USD (adjusted for additional equipment on board Technological advances have recently opened
IT on average no more than 250 USD each), a new niches in embedded computing and
simple calculation proves that the use of UAVs is communications integration, while the rapid
an extremely profitable investment, this is development of various types of unmanned aerial
especially true for searching for people in remote vehicles (UAVs) has provided a mobile platform for
areas and areas without communication innovation. A combination of the above advances
infrastructure, where any attempt at local search led to excessive production of remotely piloted
with human personnel would be a much slower quadcopters and similar flying devices, mainly for
and more expensive solution, as the cost per business and recreational purposes, with most
square kilometer can drop to less than 10 USD concepts involving video recording and
after being deployed for as long as their battery transmission of recorded data from the UAVs to
88
 D. Nozica et al. / Applied Engineering Letters Vol.8, No.2, 80-90 (2023)

the operator's location. Innovations followed https://doi.org/10.48550/arXiv.1811.05291

shortly, and new application concepts were [2] R.D. Arnold, H. Yamaguchi, T. Tanaka, Search
introduced. The use of UAVs was mostly associated and rescue with autonomous flying robots
with remote assistance or search and rescue through behavior-based cooperative
missions. Methods ranged from simple video intelligence. Journal of International
surveillance during short-range missions while Humanitarian Action, 3, 2018: 18.
maintaining line-of-sight, to long-range missions or https://doi.org/10.1186/s41018-018-0045-4
missions under difficult geographic conditions. In
[3] A.L. Alfeo, M.G.C.A. Cimino, N. De Francesco,
such scenarios, various communication
A. Lazzeri, M. Lega, G. Vaglini, Swarm
technologies were used, all with certain
coordination of mini-UAVs for target search
advantages and limitations. The authors presented
using imperfect sensors. Intelligent Decision
a conceptual improvement to an existing proposed
Technologies, 18(2), 2018: 149-162.
solution based on a swarm of UAVs with a pre-
programmed flight path actively searching for a https://doi.org/10.3233/IDT-170317
Wi-Fi enabled device that divides the [4] D. Nožica, D. Blažević, T. Keser, Unmanned
communication layers into three layers: the edge Aerial Vehicle Swarm Uses Wi-Fi to Search for
layer, where ground-based UAVs actively search Stranded People in Remote Areas Embedded
for target devices; the nebulous layer, which is devices as active scanners in search for Wi-Fi-
used by detection UAVs to synchronize and enabled mobile phones. 2021 10th
transmit status updates and positive detection Mediterranean Conference on Embedded
information to higher-positioned aircraft; and the Computing (MECO), 7-10 June 2021, Budva,
cloud layer, which is used by main UAVs to Montenegro, pp.1-4.
transmit information via a narrowband IoT https://doi.org/10.1109/MECO52532.2021.94
communications channel to low-earth orbit 60147
satellites and back to the control center via the [5] W. Wang, R. Joshi, A. Kulkarni, W.K. Leong, B.
Internet. The proposed approach addresses flawed Leong, Feasibility study of mobile phone WiFi
attempts where terrestrial infrastructure is detection in aerial search and rescue
required to facilitate information reception or operations. Proceedings of the 4th Asia-Pacific
where the passive search method is used to detect Workshop on Systems, 29 - 30 July 2013,
Wi-Fi enabled devices. The limitations and Singapore, Singapore, pp.1-6.
potential improvements are explained. Future https://doi.org/10.1145/2500727.2500729
advances in aircraft technology and [6] M.A. Khan, I.M. Qureshi, F. Khanzada, A Hybrid
communications modules will enable different Communication Scheme for Efficient and Low-
uses of communications layers and may even Cost Deployment of Future Flying Ad-Hoc
enable live image streaming over satellite Network (FANET). Drones, 3(1), 2019: 16.
communications channels in the same price range
https://doi.org/10.3390/drones3010016
as the current solution within this decade.
[7] A. Calabro, R. Giuliano, Integrated Wi-Fi and
ACKNOWLEDGMENT LoRa network on UAVs for localizing people
during SAR operations. 2021 AEIT International
Conference on Electrical and Electronic
This work results from implementing research
Technologies for Automotive (AEIT
activities on the project “Investing in research
AUTOMOTIVE), 17-19 November 2021, Torino,
activities with the aim of developing a new product
Italy, pp.1-6.
line” (KK.01.2.1.02.0029). The stated project is
funded by the European Regional Development https://doi.org/10.23919/AEITAUTOMOTIVE52
Fund. 815.2021.9662760
[8] F. Ahmed, J.C. Mohanta, A. Keshari, P.S. Yadav,
REFERENCES Recent Advances in Unmanned Aerial Vehicles:
A Review. Arabian Journal for Science and
[1] V. Lomonaco, A. Trotta, M. Ziosi, J.de.D.Y. Engineering, 47, 2022: 7963-7984.
Ávila, N. Díaz-Rodríguez, Intelligent Drone https://doi.org/10.1007/s13369-022-06738-0
Swarm for Search and Rescue Operations at [9] Y.-H. Ho, Y.-R. Chen, L.-J. Chen, Krypto:
Sea. arXiv, 2018. Assisting Search and Rescue Operations using
Wi-Fi Signal with UAV. Proceedings of the First
89
 D. Nozica et al. / Applied Engineering Letters Vol.8, No.2, 80-90 (2023)

Workshop on Micro Aerial Vehicle Networks, Unmanned Vehicles. Sensors, 21(21), 2021:
Systems, and Applications for Civilian Use, 18 6984. https://doi.org/10.3390/s21216984
May 2015, Florence, Italy, pp.3-8. [16] M. Silvagni, A. Tonoli, E. Zenerino, M.
https://doi.org/10.1145/2750675.2750684 Chiaberge, Multipurpose UAV for search and
[10] T.D. Dinh, Pirmagomedov R, Pham VD, et al. rescue operations in mountain avalanche
Unmanned aerial system–assisted wilderness events. Geomatics, Natural Hazards and Risk,
search and rescue mission. International 8(1), 2017: 18-33.
Journal of Distributed Sensor Networks, 15(6), https://doi.org/10.1080/19475705.2016.1238
2019: 155014771985071. 852
https://doi.org/10.1177/1550147719850719 [17] M. Campion, P. Ranganathan, S. Faruque, UAV
[11] S. Kashihara, A. Yamamoto, K. Matsuzaki, K. swarm communication and control
Miyazaki, T. Seki, G. Urakawa, M. Fukumoto, C. architectures: a review. Journal of Unmanned
Ohta, Wi-SF: Aerial Wi-Fi Sensing Function for Vehicle Systems, 7(2), 2019: 93-106.
Enhancing Search and Rescue Operation. 2019 https://doi.org/10.1139/juvs-2018-0009
IEEE Global Humanitarian Technology [18] Advantech Provides LRPv Solution Using
Conference (GHTC), 17-20 October 2019, Semtech LoRa Technology for Intelligent
Seattle, USA, pp.1-4. Monitoring Applications. Advantech Co., Ltd.
https://doi.org/10.1109/GHTC46095.2019.903 https://www.advantech.com/en/resources/ne
3073 ws/96d1606b-4429-4cb2-a9cc-aa3b1ee379c9
[12] O.A. Saraereh, A. Alsaraira, I. Khan, P. (Accessed 2 February 2023).
Uthansakul, Performance Evaluation of UAV- [19] Swarm - Low cost, global satellite connectivity
Enabled LoRa Networks for Disaster for IoT. SWARM. https://swarm.space/
Management Applications. Sensors, 20(8), (Accessed 2 February 2023).
2020: 2396.
[20] DJI Store - Official Store for DJI Drones,
https://doi.org/10.3390/s20082396 Gimbals and Accessories (Croatia). DJI Sky City,
[13] M. Behjati, A.B. Mohd Noh, H.A.H. Alobaidy, Shenzhen, China. https://store.dji.com/hr
M.A. Zulkifley, R. Nordin, N.F. Abdullah, LoRa (Accessed 2 February 2023).
Communications as an Enabler for Internet of [21] Core Specification. Bluetooth SIG, Inc.,
Drones towards Large-Scale Livestock Kirkland, USA.
Monitoring in Rural Farms. Sensors, 21(15),
https://www.bluetooth.com/specifications/sp
2021: 5044.
ecs/core-specification-5-3/
https://doi.org/10.3390/s21155044
(Accessed 2 February 2023).
[14] L. Davoli, E. Pagliari, G. Ferrari, Hybrid LoRa-
[22] E.K. Hung, D.A. Townes, Search and Rescue in
IEEE 802.11s Opportunistic Mesh Networking
Yosemite National Park: A 10-Year Review.
for Flexible UAV Swarming. Drones, 5(2), 2021:
Wilderness & Environmental Medicine, 18(2),
26. https://doi.org/10.3390/drones5020026
2007: 111-116.
[15] A.-M. Dragulinescu, S. Halunga, C. Zamfirescu,
https://doi.org/10.1580/06-WEME-OR-
Unmanned Vehicles’ Placement Optimisation
022R1.1
for Internet of Things and Internet of

90

This work is licensed under a Creative Commons Attribution-Non Commercial 4.0 International License (CC BY-NC 4.0)