Evaluating Use of ARQ Strategies in Communication Protocols for

Search and Rescue

Antonello Calabr

, Eda Marchetti

and Maria Teresa Paratore

Istituto di Scienza e Tecnologie dell’Informazione “A. Faedo”, CNR, Pisa, Italy

Keywords:

LoRa, SAR, ARQ, Reliability.

Abstract:

Search and Rescue (SAR) operations often occur in remote and challenging environments where conventional

communication infrastructures are unavailable or unreliable. Effective communication is crucial for mission

success. Low-power wide area network (LPWAN) protocols, particularly the LORA (Long Range) protocol,

have gained traction due to their low power consumption and extended range. However, LORA’s low relia-

bility presents signiﬁcant challenges in the time-sensitive context of SAR operations, necessitating effective

communication strategies. This paper examines the reliability of communication protocols in real scenarios,

focusing on Stop & Wait (S&W) and Selective Repeat (SR) Automatic Repeat reQuest (ARQ) protocols. It

evaluates their suitability by addressing operational constraints such as geographic barriers, time sensitivity,

and simplicity of implementation. Key contributions include investigating the current literature, a real imple-

mentation of the ARQ algorithm, and a comparative analysis of these protocols under real-world conditions.

Furthermore, the study presents a real-world implementation of ARQ mechanisms and evaluates their opera-

tional trade-offs in SAR scenarios, considering both computational constraints and deployment feasibility.

1 INTRODUCTION

In Search and Rescue (SAR) operations, effective

communication is the backbone of successful mis-

sions (Alsaeedy and Chong, 2020). These opera-

tions often occur in remote, hostile, or challenging en-

vironments where traditional communication infras-

tructures, such as cellular networks or satellite com-

munications, are either lacking or unreliable (Calabr

and Marchetti, 2024). In these conditions, the avail-

ability of robust and streamlined communication pro-

tocols that can operate under these difﬁcult conditions

is a stringent need.

Among the various communication technologies

available, low-power wide area network (LPWAN)

protocols, particularly the LORA (Long Range) pro-

tocol, have gained signiﬁcant traction due to their

capability to facilitate communication over extensive

distances while consuming minimal power. LORA

protocol is especially valuable in SAR missions,

where team members or equipment may spread across

large areas that are difﬁcult to access. LORA net-

https://orcid.org/0000-0001-5502-303X

https://orcid.org/0000-0003-4223-8036

https://orcid.org/0000-0002-9089-8445

works can support many devices with lower en-

ergy requirements, permitting extended operational

periods without frequent recharging or replacement.

However, despite these advantages, a notable draw-

back of LPWAN protocols like LORA is their inher-

ently low reliability. Given the time-sensitive nature

of SAR operations, failures in communication can

have critical implications. Consequently, maintain-

ing data integrity and ensuring timely transmissions is

extremely critical. Numerous strategies have been ex-

plored in the existing literature to enhance the reliabil-

ity of communications in SAR contexts (Mendelsohn

et al., 2024; Mabulu et al., 2024; Akgun et al., 2023),

such as multi-hop communication (Anuradha et al.,

2022) and Automatic Repeat reQuest (ARQ) (Vasiliev

and Abilov, 2015).

Each method possesses distinct advantages and

disadvantages that are particularly relevant to the

unique contexts in which SAR operations are con-

ducted. Speciﬁcally, multi-hop communication

strategies focus on relaying data through intermediate

nodes to extend the transmission range and enhance

connectivity. This approach can signiﬁcantly bolster

communication in scenarios where geographical fea-

tures, such as mountains or dense forest,s hinder di-

rect transmission. However, this method often relies

Calabrò, A., Marchetti, E. and Paratore, M. T.

Evaluating Use of ARQ Strategies in Communication Protocols for Search and Rescue.

DOI: 10.5220/0013704800003985

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 21st International Conference on Web Information Systems and Technologies (WEBIST 2025), pages 59-70

ISBN: 978-989-758-772-6; ISSN: 2184-3252

on having a certain level of infrastructure in place,

such as additional communication nodes or devices.

In SAR operations, this infrastructure may not always

be available, potentially limiting the effectiveness of

multi-hop communication. The Automatic Repeat re-

Quest mechanisms provide a systematic approach to

retransmitting lost or erroneous packets, thereby im-

proving the reliability of data communication. Differ-

ent ARQ protocols (YOSHIMOTO et al., 1991), such

as Go-Back-N, Selective Repeat, and Stop & Wait,

can be used to identify and recover transmission er-

rors efﬁciently. However, ARQ can introduce addi-

tional latency into the communication process. In a

SAR context, where timely updates are crucial, this

latency may not be the most efﬁcient solution, espe-

cially in simpler, single-core systems where the over-

head of managing ARQ may outweigh its beneﬁts.

This paper aims to investigate these enforcements

on communication, focusing on the peculiarities of

ARQ. In particular, the paper focuses on the hypoth-

esis that the beneﬁts of the Selective Repeat (SR)

ARQ protocol are often limited, resulting in perfor-

mance that may be functionally similar to Stop and

Wait (S&W) in many situations. Therefore, we will

evaluate its cost-effectiveness in SAR environments

through several key analyses, including:

• Investigation of current literature: We provide an

overview of the current methodologies in pro-

tocol reliability to classify their challenges and

limitations. This will assist in identifying opti-

mal guidelines for applying these communication

strategies in SAR operations.

• Implementation of the algorithms: We will de-

tail the design and application of various algo-

rithms speciﬁcally tailored for simplistic systems.

These algorithms aim to optimize transmission

times without signiﬁcantly increasing complexity

or cost.

• Comparative performance analysis: we will

showcase the comparative analysis of the imple-

mented transmission improvement algorithms un-

der real-world conditions to ascertain their effec-

tiveness.

Therefore, the paper wants to provide a realistic

investigation of the operational constraints inherent

in SAR missions, providing valuable data to support

informed decision-making when selecting communi-

cation strategies. In our exploration, we will provide

essential components to support our ﬁndings, includ-

ing a detailed description of i) the hardware conﬁgura-

tions and technical speciﬁcations, ii) the execution en-

vironment, including actors such as geographic loca-

tion, terrain characteristics, and environmental chal-

lenges that could impact possible SAR operations; iii)

the data packets, including their structures and spe-

ciﬁc transmission requirements.

This will allow for possible experiment replica-

tion with different protocols and conditions. Through

this extensive exploration, we aim to contribute valu-

able knowledge that could enhance the communi-

cation strategies employed in SAR operations. By

advancing our understanding of ARQ protocols and

their alternatives, we ultimately seek to improve the

efﬁcacy and safety of these critical missions, ensur-

ing that personnel can communicate effectively and

respond swiftly to emergencies.

Although Go-Back-N ARQ is widely studied, this

work focuses on Stop & Wait and Selective Repeat

due to their lower computational requirements and

suitability for low-power devices, which are more

aligned with SAR operation constraints.

Roadmap. The remainder of this article is orga-

nized as follows. The main background and related

works are presented in Section 2, while the proposed

architectural design and its reference behavior are de-

scribed in Section 3. The Selective Repeat ARQ im-

plementation and the system deployed for its execu-

tion are presented in Section 4. Furthermore, in Sec-

tion 5, the experimental results are evaluated and dis-

cussed. Finally, in Section 7, conclusions and possi-

ble future works are highlighted.

2 BACKGROUND AND RELATED

WORK

This section reports the current research activity in

optimizing LoRa communications and the required

background knowledge in Section 2.2 and 2.1, respec-

tively.

2.1 State-of-the-Art

Optimization of LoRa communications has stimu-

lated different research ﬁeld solutions (Hilmani et al.,

2022; Kamal et al., 2023). Most proposed solutions

use intermediate nodes (end devices or gateways) to

expand the network coverage (Bor et al., 2016). The

intermediate nodes forward data until they reach a ﬁ-

nal gateway according to speciﬁc routing protocols

(Jouhari et al., 2023; Leonardi et al., 2023; Lundell

et al., 2018; Paredes et al., 2023) that can take into

consideration also quality parameters such as usage,

remaining battery life, and trafﬁc rate (Anedda et al.,

2018; Wong et al., 2024; Islam et al., 2023; Ebi et al.,

2019; Zhao et al., 2023).

WEBIST 2025 - 21st International Conference on Web Information Systems and Technologies

In (Zorbas et al., 2021), the authors tackle the

problem of scheduling the retransmission of buffered

data in LoRa networks, typical of the situations in

which connectivity between gateways is not available,

and propose a time-slotted transmission scheduling

mechanism. It is also highlighted that synchronous

communications positively affect data collection time

and network performance. In (Chen et al., 2019),

the authors present a cost-effective hardware archi-

tecture for a LoRa gateway that increases through-

put by improving bandwidth usage. Other solutions

adopt organization algorithms of the nodes to spe-

ciﬁc topology (Haubro et al., 2020; Leenders et al.,

2023; Gkotsiopoulos et al., 2021) or cluster (Cotrim

and Kleinschmidt, 2020; Sun et al., 2022; Almuhaya

et al., 2022; Mamour and Congduc, 2019) to ensure

reliable data transfer to the target gateway (Dwijak-

sara et al., 2019; Wong et al., 2024; Bomgni et al.,

2023).

A machine learning (ML) approach is used

in (Abubakar et al., 2022), proposing a scalability op-

timization for LoRaWAN networks based on a com-

bination of ML algorithms. The authors identify the

average distance between LoRa end-devices and gate-

ways as a function to be minimized with the constraint

of improving the received signal strength (RSSI) at

each end-device, then solve this optimization problem

by combining K-Means clustering (to optimize gate-

ways’ locations) and Regression Neural Networks (to

maximize RSSIs). Furthermore, they perform a se-

ries of trials to form the foundation for developing an

adaptive algorithm capable of dynamically allocating

bandwidth autonomously.

Finally, hybrid solutions exist, which rely on mes-

sage transmission and synchronization techniques to

reduce energy consumption while improving cover-

age and dependability at the same time (Zhou et al.,

2019; Tanjung et al., 2020).

2.2 LoRa Communication and ARQ

This section details the two main baseline concepts

used in this paper: LoRa communication and Auto-

matic Repeat Request (ARQ).

LoRa Communication. LoRa is a wireless commu-

nication technology designed for long-range, low-

power applications conceived by Cyleo (Greno-

ble) in 2009 and released in January 2015 (Sornin

et al., 2015). It is based on the Chirp Spread

Spectrum (CSS), which allows maintaining com-

munications over several kilometers, making it

an ideal solution for scalable applications such

as the Internet of Things (IoT) (Grunwald et al.,

2019), smart cities, digital agriculture (Codeluppi

et al., 2020), environmental monitoring (Fraga-

Lamas et al., 2019), and asset tracking. One of

the main advantages of LoRa is its ability to pro-

vide deep indoor penetration and connectivity in

challenging environments, such as urban areas or

buildings with thick walls, where traditional wire-

less technologies may struggle to maintain reli-

able communication. In addition, LoRa supports

bi-directional communications, which means that

devices can transmit data and receive commands

or updates from a central server or gateway.

Automatic Repeat Request. Automatic Repeat Re-

quest (ARQ) (Lin et al., 1984) is an error-control

strategy used in data communications to ensure

reliable information delivery over unreliable or

error-prone transmission channels. It includes

mechanisms for checking errors on received data,

such as timeouts and feedback signals (acknowl-

edgments), to handle the re-transmission of er-

roneous or missing data segments. There are

several ARQ protocols described in the litera-

ture (Makridis et al., 2022; Kalør et al., 2022;

Choi et al., 2020), including

• Stop-and-Wait ARQ, which is based on the ac-

knowledgment of reception for one segment of

data before sending the next segment

• Go-Back-N ARQ, which allows the transition

of multiple data segments without waiting for

individual acknowledgments and

• Selective Repeat ARQ, which allows the indi-

vidual acknowledgment of each received seg-

ment and provides for re-transmission only in

the event of segments with errors.

Table 1 summarizes the main features of the

three ARQ protocols according to the literature

overview(Makridis et al., 2022; Kalør et al., 2022;

Choi et al., 2020). The table’s purpose is to high-

light each protocol’s strengths and weaknesses so

that easy implementation decisions can be made

and to assist in identifying optimal guidelines for

applying these communication strategies in SAR

operations.

While Go-Back-N represents a moderate trade-off

between complexity and performance, its requirement

for retransmission of multiple frames upon a single er-

ror is suboptimal in SAR contexts with high energy

constraints. Our implementation instead prioritizes

resource efﬁciency and simplicity, aligning with the

limitations of real SAR equipment. Considering the

current state of the art, a contribution of the paper

is a comparative analysis of these algorithms under

real-world conditions to conﬁrm the overall analysis

Evaluating Use of ARQ Strategies in Communication Protocols for Search and Rescue

Table 1: Main features of the three ARQ strategies.

Stop and Wait Go-Back-N Selective Repeat

Channel Utilization Poor Better than SW Best among the 3

Implementation Complexity Simple Moderate Most complex

Memory Requirements Low Moderate High

Bandwidth Efﬁciency Low Moderate High

Error Handling

Retransmit

current packet

Retransmit all packets

from error point

Retransmit only

erroneous packets

Packet Reordering Not needed Not needed Required at receiver

Error Recovery Time Fast but inefﬁcient

Can be slow due to

mass retransmission

Efﬁcient, only retransmits

needed packets

provided in Table1 and ascertain their practical limi-

tation and effectiveness. For this, we ﬁrst focused our

attention on the Selective Repeat and Stop&Wait al-

gorithm since these strategies proved to be better per-

forming than others (such as the single-hop method

used by LoRaWAN (Choi et al., 2020)) in improving

reliability across LoRa networks (Choi et al., 2020;

Abedina et al., 2023). Eventually, we adopted the Se-

lective Repeat ARQ protocol, as it is the most suit-

able for increasing data transmission’s reliability and

robustness, ensuring data integrity while minimizing

retransmissions and energy consumption.

3 ARCHITECTURAL DESIGN

This section describes the architecture for imple-

menting Selective Repeat (SR) and the Stop & Wait

(S&W) ARQ protocols. Figure 1 illustrates the base-

line main components of the instantiated system (i.e.,

the Mobile Station and the Base Station) connected

through a LoRa channel. The Base Station remains

ﬁxed in position, enhancing the signal reception and

the overall spatial coverage. It is responsible for man-

aging the activities done by the receiver of the two

ARQ protocols in practical scenarios when the trans-

mitter device moves in a natural environment.

As shown in Figure 1, both the Base Station and

Mobile Station infrastructures include computational

devices (Computer) for data storage and management

operations. In particular, the Base Station includes

a database, while the Mobile Station relies on a log-

ger for the storage activities. Base Station and Mobile

Station communicates through a LoRa channel. The

Mobile Station and Base Station include a LoRa node

with an ARQ Logic artifact for implementing the ARQ

strategy (SR or S&W). Finally, the Mobile Station in-

cludes a GPS device for gathering the position of the

Mobile Station. More details of Mobile Station and

Base Station implementation are provided in the next

section.

Figure 1: Architectural overview.

4 PROTOTYPE

IMPLEMENTATION

This section presents an instance of the architecture

depicted previously. In particular, details of the two

components, Base Station, and Mobile Station, are

provided in Section 4.1 and 4.2.In comparison, the

details related to the implementation of the ARQ pro-

tocol are provided in Section 4.3.

4.1 Base Station

Figure 2 shows the instantiation of the Base Sta-

tion presented in Figure 1. In realizing the Base

Station devices, the following have been selected:

A barebone computer for the Computer component.

As in Figure 2, Barebone is in charge of manag-

ing the storage of the data collected during the ex-

ecution. The Barebone device executes two arti-

facts: MySQL, which is an instance of the popular

open-source DB, where the SerialManager stores the

data gathered by the transmission received through

WEBIST 2025 - 21st International Conference on Web Information Systems and Technologies

Figure 2: Base Station Architecture.

the LoRa Wi-Fi module. The SerialManager arti-

fact has been developed using Java code. The Bare-

bone device is connected to the LoRa Wi-Fi module

through a USB port that emulates a serial port. The

LoRa Wi-Fi module manages the transmission using

the LoRa radio protocol. In particular, the Receiver-

ARQ artifact is the component developed on top of

the LoRa Wi-Fi Module for executing the ARQ pro-

tocol (i.e., Selective Repeat or Stop & Wait). In the

proposed implementation, the Barebone device is re-

alized through a Lenovo ThinkEdge SE10 running

Windows 10 IoT Enterprise; the LoRa Wi-Fi Module

is built by means of a LilyGo LoRa32 V2.1 device

equipped with an ESP32 and an SX1262 LoRa node

chip. The Receiver-ARQ software has been developed

using Arduino IDE

, and LilyGo libraries

. The pro-

totype is shown in Figure 5.

4.2 Mobile Station

Figure 3 details the instantiation of the architecture

of Mobile Station. It is composed of three different

devices:

• a Raspberry PI Model 4B (R-PI 4 Mod.B) that rep-

resents the instantiation of the Computer shown in

Figure 1;

• a LoRa Wi-Fi Module, connected using a USB

port that emulates a serial port to the R-PI 4

Mod.B;

• and a GPS device connected using another USB

port of the R-PI 4 Mod.B.

On the R-PI 4 Mod.B device, three artifacts have

been deployed: the Java-developed SerialManager

https://www.arduino.cc/en/software

https://github.com/Xinyuan-LilyGo/

TTGO-LoRa-Series

Figure 3: Mobile Station Architecture.

Figure 4: Mobile Station.

artifact for sending and receiving data from the LoRa

Wi-Fi module related to communication packets and

the GPS device from which it receives the current

GPS position. These data are stored locally by the

LocalLogger artifact. Also, in this case, the Receiver-

ARQ artifact is the component developed on top of the

LoRa Wi-Fi Module for executing the ARQ protocol

(i.e., Selective Repeat or Stop and Wait).

In the proposed implementation, the R-PI 4 Mod.B

device is realized by a Raspberry Pi 4 Model B de-

vice with 8GB of RAM, on which a 3.5-inch touch-

screen display for debugging and management pur-

poses has been connected. It is powered by a bat-

tery shield PiSugar 5000mah

. The LoRa Wi-Fi Mod-

ule has been instantiated using a Heltec WiFi LoRa

https://www.pisugar.com/

Evaluating Use of ARQ Strategies in Communication Protocols for Search and Rescue

Figure 5: Base Station.

32(V3) device equipped with an ESP32-S3FN8 and

an SX1262 LoRa node chip powered directly by a

USB port of the R-PI 4 Mod.B node. The GPS device

is a generic GPS USB dongle capable of providing

data according to the GPS & GLONASS standard on

a serial port.

The software for running the SerialManager and

the GPSReader artifacts has been developed using

Java, while the ARQ Logic software has been devel-

oped using Arduino IDE. The LocalLogger artifact

has been developed using a Python script for storing

data gathered by the SerialManager. The prototype

of the Mobile Station is shown in Figure 4.

4.3 Implementation of ARQ Methods

As discussed in Section 2.2, the standard LoRa proto-

col suffers from a high packet loss rate, which causes

low transmission reliability. The ARQ methods have

been implemented to address and mitigate this issue,

enhance transmission, and recover any missing pack-

ets. In the current implementation, the ARQ protocols

are implemented both on the Mobile station (i.e., the

transmitter) and on the Base Station (i.e., the receiver)

to enable re-transmission each time a missing packet

is detected. The ARQ logic was implemented from

scratch in Arduino IDE using low-level control struc-

tures, without relying on pre-existing ARQ libraries.

This allowed full customization for constrained envi-

ronments and provided valuable insights into practical

limitations during integration and tuning.

Figure 6 shows the enhanced packet structure used

for communicating the GPS position on which the

Mobile station is located to the Base station. The

packet structure has been developed to incorporate the

ARQ mechanism and include an integrity check of the

data. As depicted in Figure 6, it includes:

• Two bytes to identify the transmitting node;

• Two checksum bytes to control packet integrity;

• Two bytes to determine the role of the transmitter,

i.e., the Base Station or the Mobile Station accord-

ing to the “master” or “slave” protocol.

• Two chars, one at the beginning and one at the

end, to clearly deﬁne the packet’s start and end.

Implementing the ARQ mechanism requires iden-

tifying the sender and receiver because data transfer

can occur simultaneously in two directions, generat-

ing packet loss, errors, and collisions. Additionally,

the implementation requires a mechanism for check-

ing transmission errors to improve the reliability of

the data received. Finally, information regarding the

order of the sent packets is necessary to reconstruct

the transmitted data.

Due to the computational limitation of the em-

ployed LoRa devices, it is possible to execute only a

single process with a main ”loop” cycle on which both

send and receive operations must be executed. To

enhance transmission quality, LoRa devices are pro-

grammed to have their main cycle in listening mode

for incoming messages. Transitioning to transmit

mode happens only when a new message is queued

in the buffer, and the system is not busy receiving a

message. This approach reduces reception errors by

ensuring the receiver is always ready to process data

while minimizing packet collisions by prioritizing in-

coming signals over outgoing transmissions.

An optimal interleave time of approximately 1000

ms has been calculated, balancing message schedul-

ing to improve throughput and reduce interference,

particularly in dense IoT networks. This mechanism

forms the foundation for implementing Stop & Wait

and Selective Repeat ARQ protocols, which enhance

reliability by enabling acknowledgment and retrans-

mission of lost or corrupted packets. The selective

repeat mechanism beneﬁts from the receiver’s contin-

uous mode by ensuring efﬁcient reordering and error

recovery without compromising transmission quality.

5 SHOWCASE SCENARIOS

The experiments have been executed in a speciﬁc

showcase scenario to provide a comparative perfor-

mance analysis of the ARQ mechanism in an actual

situation. In particular, experimental data have been

collected around the omissis research campus in omis-

sis (see Figure 7). Even if the location is not the typi-

cal environment of a SAR operation, the topology in-

cludes green areas, buildings, areas with a density of

WEBIST 2025 - 21st International Conference on Web Information Systems and Technologies

Figure 6: Structure of the enhanced packet with ARQ features.

Figure 7: Scenario’s path to execute.

different signals, and rooms and buildings with signal

shielding for research purposes.

In Figure 8, it is possible to identify the LoRa an-

tenna placed at the bottom left of the ﬁgure. The path

selected for the experiments has a radius of 150m

(light yellow line in Figure 7), and the antenna has

been placed in the building’s roof corner to increase

transmission difﬁculties (red spot in Figure 7).

The experiment has been executed in favorable

weather conditions without rain or fog to prevent sig-

nal attenuation. The device shown in Figure 4 has

been placed on a bike that travels across the path at an

average speed of around 10 Km/h.

The scope of the experiments was to determine

the quality of the received signals and the number

of packets correctly delivered as the communication

distance between two nodes varies. The transmitter

sends the detected GPS position to the receiver at a

constant frequency ( 1000ms).

Three instances of the experiments have been ex-

ecuted:

1. The path has been crossed without any retransmis-

sion mechanism: see Figure 8 (NO-ARQ);

2. the path has been crossed enabling Stop & Wait

ARQ mechanism: see Figure 9 (S&W-ARQ);

3. the path has been crossed, enabling the Selec-

tive Repeat ARQ mechanism: see Figure 10 (SR-

ARQ).

In all three experiments, the path has been crossed,

moving in the direction of the arrows in the ﬁgures.

Additionally, in all the ﬁgures, the parts of the path

colored in green represent the messages correctly sent

from the transmitter and received by the LoRa An-

tenna without relying on requiring a retransmission

mechanism (if enabled). Notably, consistent message

loss in speciﬁc areas corresponds to known shadow

zones caused by building structures and vegetation

density, which impede line-of-sight (LOS) transmis-

sion and are typical challenges in SAR deployments.

In all three experiments, the data (transmitted and

received) were stored locally on the database instance

of the Base Station and Mobile Station. For evaluating

the packet loss rate, the following formula has been

considered:

PacketLossRate = 1 −

(1)

where N

and N

are the total number of received

and transmitted packets, respectively.

6 EXPERIMENT EXECUTION

In this section, details about the execution of the three

experiments are provided. Each experiment lasted ap-

proximately 15 minutes, with the mobile station con-

tinuously transmitting GPS packets every 1 second.

All trials were repeated under similar conditions to

ensure consistency in results.

Evaluating Use of ARQ Strategies in Communication Protocols for Search and Rescue

Figure 8: TX-RX data without ARQ.

Figure 9: TX-RX data with Stop & Wait ARQ.

Figure 10: TX-RX data with Selective Repeat ARQ.

WEBIST 2025 - 21st International Conference on Web Information Systems and Technologies

In this regard, the use of color-coded geographic

maps was intentionally preferred over numeric tables

to visually highlight the spatial impact of packet loss

concerning the surrounding environment—an essen-

tial factor in SAR contexts. Rather than proposing a

theoretical model, we focused on experimental val-

idation, emphasizing computational limitations, im-

plementation simplicity, and the realistic effective-

ness of the protocols. Therefore, metrics such as en-

ergy consumption or theoretical delay were deliber-

ately not explored in depth, in favor of producing re-

sults that are easily replicable in practical scenarios.

Execution with NO-ARQ Enabled. The results of

the execution of the ﬁrst experiment are depicted in

Figure 8. As shown, the part of the path colored in red

identiﬁes the areas where the antenna has received no

messages.

A post-analysis of the data collected during the

ﬁrst experiment by the Base Station and the Mobile

Station allows the computation of the packet loss rate

in the executed path. In particular, the derived data

were as follows:

• #packets generated by the Mobile Station = 3148;

• #packets received by the Base Station = 1287.

Consequently, the packet loss rate is:

PacketLossRate = 1 −

1287

3148

= 59, 11% (2)

Execution with S&W-ARQ Enabled. The execu-

tion of the second experiment is depicted in Figure 9.

As in the ﬁgure, the part of the path colored in orange

identiﬁes the areas where the antenna has received

messages after retransmission, while the Mobile Sta-

tion comes back in LOS (green part of the path). The

part colored in green represents the situation that did

not require a retransmission mechanism.

The packet delivery has been guaranteed using the

ARQ mechanism, and the results of the second exper-

iment provide a 100% delivery rate.

• #packets generated by the Mobile Station = 3291;

• #packets received by the Base Station = 3291.

The computed packet loss rate is:

PacketLossRate = 1 −

3291

= 0% (3)

Execution with SR-ARQ Enabled. The execution

of the third experiment is depicted in Figure 10. As in

the previous experiment, in the ﬁgure, the part of the

path colored in orange identiﬁes the areas where the

antenna has received messages after retransmission.

Figure 11: Segment of the SR-ARQ execution with packet

receiving order.

The parts colored in purple identify the areas where

Selective Repeat has an effect, causing the messages

in the purple set to be received before those of the

previous orange segment.

To clarify, Figure 11 shows a segment of the ex-

ecuted path. The order of sent messages was recon-

structed after the data analysis. As depicted in the

ﬁgure, when the Mobile Station reached the marked

position in the green segment, the Base Station re-

ceived in order ﬁrst the messages of the purple seg-

ment numbered 1, then the messages of the orange

segment numbered 2 and ﬁnally the messages of the

orange segment numbered 3. The number of packets

generated by the Mobile Station was 3364, with all

successfully received by the Base Station, resulting in

a packet loss rate of 0%. However, transmission order

required post-processing due to reordering inherent to

SR-ARQ.

Experiment Evaluation. Analyzing the data gath-

ered during the experiment, several key insights can

be drawn regarding selecting an appropriate ARQ

protocol, as summarized in Table 1.

1. Channel Utilization and Memory

Requirements: Selective Repeat necessi-

tates buffering out-of-order packets at both the

sender and receiver, alongside additional mech-

anisms for managing cumulative ACKs. The

simpler Stop & Wait protocol may suit systems

with limited computational resources or memory.

2. Bandwidth Efficiency and Energy

Consumption: In scenarios where minimiz-

ing energy consumption is crucial, such as LoRa

transmissions, the overhead associated with

managing the sliding window in Selective Repeat

may outweigh its beneﬁts, making Stop & Wait a

more efﬁcient choice.

3. Packet Size and Buffer Limitations:

Given the constrained buffer resources available

on LoRa low-power devices, Selective Repeat

demands substantial computational power to

Evaluating Use of ARQ Strategies in Communication Protocols for Search and Rescue

manage retransmissions and maintain packet

order, particularly in environments with frequent

packet errors. In such contexts, the straight-

forward Stop-and-Wait protocol offers a more

feasible solution due to its simplicity.

7 DISCUSSION AND

CONCLUSION

The paper discussed the crucial role of effective com-

munication in Search and Rescue (SAR) operations,

particularly in challenging environments where tradi-

tional communication methods are often unreliable.

It highlighted the advantages of low-power wide area

network (LPWAN) protocols, speciﬁcally the LoRa

protocol, which allows for long-range communication

with minimal power consumption, making it suitable

for SAR missions where resources are limited.

The paper aimed to investigate the effectiveness

and cost-efﬁciency of these communication strategies

in SAR environments, particularly focusing on the Se-

lective Repeat ARQ protocol. It included an overview

of the literature about current methodologies, the im-

plementation of tailored algorithms for simpler sys-

tems, and a comparative performance analysis of

these algorithms in real-world scenarios. In partic-

ular, the paper evaluated the performance and cost-

effectiveness of ARQ protocols that can be applied

in the context of Search and Rescue (SAR), focusing

mainly on Stop and Wait (S&W) and Selective Re-

peat (SR). The ultimate goal was to provide valuable

insights and guidelines for selecting optimal commu-

nication strategies in SAR missions, addressing the

operational constraints these missions face.

The analysis revealed that while SR offers theoret-

ical advantages in managing lost packets and reducing

latency, these beneﬁts are largely negated in scenar-

ios involving small packets, low latency requirements,

and frequent messages, as demonstrated through the

showcased scenario. In such cases, the higher imple-

mentation costs and complexity of SR are not justi-

ﬁed, as S&W achieves comparable performance with

signiﬁcantly lower resource demands. Implementing

tailored algorithms and conducting comparative per-

formance analyses under real-world conditions, we

provided actionable insights into the practical appli-

cation of LoRa for small packets and frequent mes-

sage transmission. These ﬁndings highlight the im-

portance of context-speciﬁc communication strate-

gies that strike a balance between reliability and sim-

plicity, particularly in resource-constrained and time-

sensitive operations.

Furthermore, the system architecture was inten-

tionally kept minimal (point-to-point) to reﬂect the

real operational limits typical of SAR missions, where

complex infrastructures or multi-hop networks are of-

ten unavailable. However, we acknowledge that ex-

ploring scalability in broader contexts would be valu-

able and is considered a direction for future work.

While the article does not include a direct compar-

ison with alternative technologies (e.g., Wi-Fi mesh,

LTE, BLE), the choice of LoRa is motivated by its fa-

vorable characteristics for SAR missions: low energy

consumption, long-range coverage, and suitability for

remote environments.

This study is a foundation for decision-making in

choosing communication protocols for SAR scenar-

ios, enhancing effective and dependable communica-

tion while reducing operational costs and complex-

ity. While the study’s analysis is mostly qualitative, it

highlights practical deployment trade-offs that are of-

ten overlooked in more theoretical evaluations. These

insights are valuable for practitioners facing real SAR

mission constraints. Future work may explore hybrid

approaches or further optimization of S&W by tun-

ing parameters such as the Spreading Factor or other

transmission settings to enhance performance across

more diverse SAR environments.

ACKNOWLEDGEMENTS

This work is supported by RESTART (PE00000001)

under the PNRR of the Italian MUR program

NextGenerationEU.

REFERENCES

Abedina, T., Yawa, C. T., Koha, S. P., Hannanc, M., and

Kiong, S. (2023). The energy-efﬁcient control solu-

tions of smart street lighting systems: A review, is-

sues, and recommendations. Engineering and Tech-

nology Journal, 41(8):1–24.

Abubakar, A. K., Shore, T., and Sastry, N. R. (2022). Con-

strained machine learning for lora gateway location

optimisation. Proceedings of the 17th Asian Internet

Engineering Conference.

Akgun, S. A., Ghafurian, M., Crowley, M., and Dauten-

hahn, K. (2023). Using affect as a communication

modality to improve human-robot communication in

robot-assisted search and rescue scenarios. IEEE

Trans. Affect. Comput., 14(4):3013–3030.

Almuhaya, M. A., Jabbar, W. A., Sulaiman, N., and Abdul-

malek, S. (2022). A survey on lorawan technology:

Recent trends, opportunities, simulation tools and fu-

ture directions. Electronics, 11(1):164.

Alsaeedy, A. A. R. and Chong, E. K. P. (2020). 5g and

uavs for mission-critical communications: Swift net-

WEBIST 2025 - 21st International Conference on Web Information Systems and Technologies

work recovery for search-and-rescue operations. Mob.

Networks Appl., 25(5):2063–2081.

Anedda, M., Desogus, C., Murroni, M., Giusto, D. D., and

Muntean, G.-M. (2018). An energy-efﬁcient solution

for multi-hop communications in low power wide area

networks. In 2018 IEEE International Symposium

on Broadband Multimedia Systems and Broadcasting

(BMSB), pages 1–5. IEEE.

Anuradha, D., Subramani, N., Khalaf, O. I., Alotaibi, Y.,

Alghamdi, S., and Rajagopal, M. (2022). Chaotic

search-and-rescue-optimization-based multi-hop data

transmission protocol for underwater wireless sensor

networks. Sensors, 22(8).

Bomgni, A. B., Ali, H. M., Shuaib, M., Mtopi Chebu, Y.,

et al. (2023). Multihop uplink communication ap-

proach based on layer clustering in lora networks for

emerging iot applications. Mobile Information Sys-

tems, 2023.

Bor, M. C., Vidler, J., and Roedig, U. (2016). Lora for the

internet of things. In Ewsn, volume 16, pages 361–

366.

Calabr

o, A. and Marchetti, E. (2024). Transponder: Support

for localizing distressed people through a ﬂying drone

network. Drones, 8(9):465.

Chen, J. J., Liu, V., and Caelli, W. J. (2019). An adaptive

and autonomous lora gateway for throughput optimi-

sation. Proceedings of the Australasian Computer Sci-

ence Week Multiconference.

Choi, R., Lee, S., and Lee, S. (2020). Reliability improve-

ment of lora with arq and relay node. Symmetry,

12(4):552.

Codeluppi, G., Cilfone, A., Davoli, L., and Ferrari, G.

(2020). Lorafarm: A lorawan-based smart farming

modular iot architecture. Sensors, 20(7):2028.

Cotrim, J. R. and Kleinschmidt, J. H. (2020). Lorawan

mesh networks: A review and classiﬁcation of mul-

tihop communication. Sensors, 20(15):4273.

Dwijaksara, M. H., Jeon, W. S., and Jeong, D. G. (2019).

Multihop gateway-to-gateway communication proto-

col for lora networks. In 2019 IEEE International

Conference on Industrial Technology (ICIT), pages

949–954. IEEE.

Ebi, C., Schaltegger, F., R

ust, A., and Blumensaat, F.

(2019). Synchronous lora mesh network to monitor

processes in underground infrastructure. IEEE access,

7:57663–57677.

Fraga-Lamas, P., Celaya-Echarri, M., L

opez-Iturri, P.,

Castedo, L., Azpilicueta, L., Aguirre, E., Su

arez-

Albela, M., Falcone, F., and Fern

andez-Caram

es,

T. M. (2019). Design and experimental validation

of a lorawan fog computing based architecture for

iot enabled smart campus applications. Sensors,

19(15):3287.

Gkotsiopoulos, P., Zorbas, D., and Douligeris, C. (2021).

Performance determinants in lora networks: A litera-

ture review. IEEE Communications Surveys & Tutori-

als, 23(3):1721–1758.

Grunwald, A., Schaarschmidt, M., and Westerkamp, C.

(2019). Lorawan in a rural context: Use cases and

opportunities for agricultural businesses. In Mobile

Communication - Technologies and Applications; 24.

ITG-Symposium, pages 1–6.

Haubro, M., Orfanidis, C., Oikonomou, G., and Fafoutis, X.

(2020). Tsch-over-lora: long range and reliable ipv6

multi-hop networks for the internet of things. Internet

Technology Letters, 3(4):e165.

Hilmani, A., Siham, A., and Maizate, A. (2022). An ad-

vanced comparative study of routing protocols in lo-

rawan. In 2022 5th International Conference on Ad-

vanced Communication Technologies and Networking

(CommNet), pages 1–6.

Islam, M. R., Bokhtiar-Al-Zami, M., Paul, B., Palit, R.,

egoire, J.-C., and Islam, S. (2023). Performance

evaluation of multi-hop lorawan. IEEE Access.

Jouhari, M., Saeed, N., Alouini, M.-S., and Amhoud, E. M.

(2023). A survey on scalable lorawan for massive iot:

Recent advances, potentials, and challenges. IEEE

Communications Surveys & Tutorials.

Kalør, A. E., Kotaba, R., and Popovski, P. (2022). Common

message acknowledgments: Massive arq protocols for

wireless access. IEEE Transactions on Communica-

tions, 70:5258–5270.

Kamal, M. A., Alam, M. M., Sajak, A. A. B., and Su’ud,

M. M. (2023). Requirements, deployments, and chal-

lenges of lora technology: A survey. Computational

Intelligence and Neuroscience, 2023.

Leenders, G., Callebaut, G., Ottoy, G., Van der Perre, L.,

and De Strycker, L. (2023). An energy-efﬁcient lora

multi-hop protocol through preamble sampling for re-

mote sensing. Sensors, 23(11):4994.

Leonardi, L., Bello, L. L., and Patti, G. (2023). Mrt-lora:

A multi-hop real-time communication protocol for in-

dustrial iot applications over lora networks. Computer

Communications, 199:72–86.

Lin, S., Costello, D. J., and Miller, M. J. (1984). Automatic-

repeat-request error-control schemes. IEEE Commu-

nications Magazine, 22:5–17.

Lundell, D., Hedberg, A., Nyberg, C., and Fitzgerald,

E. (2018). A routing protocol for lora mesh net-

works. In 2018 IEEE 19th International Symposium

on” A World of Wireless, Mobile and Multimedia Net-

works”(WoWMoM), pages 14–19. IEEE.

Mabulu, K., Vainqueur, B., and Padir, T. (2024). A gesture-

based communication system for ﬁreﬁghters during

search and rescue missions. In IEEE International

Symposium on Safety Security Rescue Robotics, SSRR

2024, New York, NY, USA, November 12-14, 2024,

pages 72–77. IEEE.

Makridis, E., Charalambous, T., and Hadjicostis, C. (2022).

Arq-based average consensus over unreliable directed

network topologies. Systems and Control.

Mamour, D. and Congduc, P. (2019). Increased ﬂexibility in

long-range iot deployments with transparent and light-

weight 2-hop lora approach. In 2019 Wireless Days

(WD), pages 1–6. IEEE.

Mendelsohn, A., Sofge, D., and Otte, M. W. (2024). En-

hancing search and rescue capabilities in hazardous

communication-denied environments through path-

based sensors with backtracking. In Dastani, M., Sich-

man, J. S., Alechina, N., and Dignum, V., editors,

Evaluating Use of ARQ Strategies in Communication Protocols for Search and Rescue

International Conference on Autonomous Agents and

Multiagent Systems, AAMAS 2024, Auckland, New

Zealand, May 6-10, 2024, pages 2387–2389. ACM.

Paredes, W. D., Kaushal, H., Vakilinia, I., and Prodanoff,

Z. (2023). Lora technology in ﬂying ad hoc net-

works: a survey of challenges and open issues. Sen-

sors, 23(5):2403.

Sornin, N., Luis, M., Eirich, T., Kramp, T., and Hersent,

O. (2015). LoRa Speciﬁcation 1.0, Lora Alliance

Standard speciﬁcation. Available at: www.lora-

alliance.org.

Sun, Z., Yang, H., Liu, K., Yin, Z., Li, Z., and Xu, W.

(2022). Recent advances in lora: A comprehen-

sive survey. ACM Transactions on Sensor Networks,

18(4):1–44.

Tanjung, D., Byeon, S., Kim, D. H., and Kim, J. D.

(2020). Oodc: An opportunistic and on-demand for-

warding mechanism for lpwa networks. In 2020 In-

ternational Conference on Information Networking

(ICOIN), pages 301–306. IEEE.

Vasiliev, D. S. and Abilov, A. (2015). Relaying algorithms

with arq in ﬂying ad hoc networks. In 2015 Interna-

tional Siberian Conference on Control and Communi-

cations (SIBCON), pages 1–5.

Wong, A. W.-L., Goh, S. L., Hasan, M. K., and Fattah, S.

(2024). Multi-hop and mesh for lora networks: Re-

cent advancements, issues, and recommended appli-

cations. ACM Computing Surveys, 56(6):1–43.

YOSHIMOTO, M., TAKINE, T., TAKAHASHI, Y., and

HASEGAWA, T. (1991). Waiting time and queue

length distributions for go-back-n and selective-repeat

arq protocols. In PUJOLLE, G. and PUIGJANER,

R., editors, Data Communication Systems and their

Performance, pages 247–260. North-Holland, Ams-

terdam.

Zhao, G., Lin, K., Chapman, D., Metje, N., and Hao, T.

(2023). Optimizing energy efﬁciency of lorawan-

based wireless underground sensor networks: A

multi-agent reinforcement learning approach. Inter-

net of Things, 22:100776.

Zhou, W., Tong, Z., Dong, Z. Y., and Wang, Y. (2019).

Lora-hybrid: A lorawan based multihop solution for

regional microgrid. In 2019 IEEE 4th International

Conference on Computer and Communication Sys-

tems (ICCCS), pages 650–654. IEEE.

Zorbas, D., Caillouet, C., Hassan, K. A., and Pesch, D.

(2021). Optimal data collection time in lora net-

works—a time-slotted approach. Sensors (Basel,

Switzerland), 21.

WEBIST 2025 - 21st International Conference on Web Information Systems and Technologies