Received 1 April 2024, accepted 19 April 2024, date of publication 29 April 2024, date of current version 6 May 2024.

Digital Object Identifier 10.1109/ACCESS.2024.3394617

A Systematic Literature Review of the IoT in

Agriculture—Global Adoption, Innovations,
Security, and Privacy Challenges
ASMA NASEER 1 , MUHAMMAD SHMOON 1 , TANZEELA SHAKEEL 1,

SHAFIQ UR REHMAN 2 , AWAIS AHMAD2 , AND VOLKER GRUHN3

1 Department of Computer Science, National University of Computer and Emerging Sciences, Lahore 54770, Pakistan
2 College of Computer and Information Sciences, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh 11432, Saudi Arabia
3 Department of Software Engineering, University of Duisburg-Essen, 45141 Essen, Germany
Corresponding author: Shafiq Ur Rehman ([email protected])
This work was supported by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University (IMSIU) under Grant
IMSIU-RP23042.

ABSTRACT Over the past decade, an abundance of research has been conducted in the area of agricultural
technology and innovations. The Internet of Things (IoT) has demonstrated its ability to connect numerous
agricultural equipment, sensors, and specialists, boosting agricultural procedures in off-the-grid regions.
Agriculture has experienced considerable improvements in production, cost reductions, service accessibility,
and operational efficiency. With an emphasis on security, developments and trends in the sector, and
technological implementation, this research paper offers an up-to-date analysis of existing and projected
IoT applications in agriculture. In this article, enabling technologies, agricultural applications based on
cutting-edge machine learning models, and services are all examined in relation to the development of
IoT deployment in tackling diverse agricultural concerns. In the IoT-based agriculture system, potential
challenges and limitations are also addressed. In its conclusion, this research provides an extensive review
of the various aspects of IoT in agriculture, with the goal of empowering future researchers enthusiastic to
make contributions to and advancement in their quest for a more in-depth comprehension of this field of
study. A total of 96 papers were chosen for the selection from 2018 to 2023, and each was categorized using
predetermined standards. The research’s results have been thoroughly examined, providing an overview of
IoT in agriculture.

INDEX TERMS Systematic literature review (SLR), IoT, smart agriculture, crop disease identifications,
sensors, communication technologies, applications, machine learning, deep learning, security, blockchain.

I. INTRODUCTION technologies, Internet protocols, applications, and sensor

The Internet of Things (IoT) has been implemented in networks, the IoT transforms conventional objects into
a variety of applications, including smart agriculture [1], intelligent ones.
smart factories [2], personal gadgets [3], smart cities and Agriculture serves as the backbone of survival for mankind
homes [4], [5], smart health systems [6], connected trans- since it is the primary source of grains for food and other
portation [7], and smart unmanned aircraft [8]. The Internet basic resources. The development of the nation’s economy is
of Things enables real-world objects to interact with one significantly impacted by it. Additionally, it offers individuals
another, exchange data, and make decisions collaboratively. numerous and plentiful opportunities for employment. The
Using its underlying technologies, such as communication long-term stability of the country’s economy depends on
the expansion of the agriculture sector. However, a lot of
The associate editor coordinating the review of this manuscript and farmers keep cultivating their land using outdated techniques,
approving it for publication was Pedro R. M. Inácio . which causes an insufficient yield of agricultural products.
2024 The Authors. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.
60986 For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/ VOLUME 12, 2024
A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

Even when technology was employed and human labor thefts, among others [21], [22]. Furthermore, IoT enables a
was substituted by automatic machinery, the yield increased well-planned use of limited resources, ensuring that optimal
[9]. Significant developments have been achieved over the utilization of IoT increases productivity.
evolution of humanity to maximize agricultural productivity There are six significant challenges to developing an
with minimal assets and worker demands. Nonetheless, the environmentally friendly IoT-based agriculture system: hard-
rapid increase in population has never enabled both supply ware, analytics of data, repair and maintenance, connectivity,
and demand to be equated during all of these eras.The infrastructure, security of data, and privacy. The most
rapidly expanding global population is projected to reach significant hardware issues are around the selection of sensors
approximately 10 billion by 2060, a 25% increase from and distance for IoT devices. As a result, there are several
today’s figure, according to the united nations survey [10]. sensor categories that may be utilized in IoT applications (for
Nonetheless, due to population growth, consumer demand example, pressure sensors, temperature sensors, chemical
for edible grains has increased dramatically in recent sensors,proximity sensors, humidity sensors, water quality
years. Unluckily, the consumption of grain is indirectly sensors, gas sensors, and so on) [23]. The analytics of data
linked to population expansion. Consequently, the worldwide issue entails applying predictive algorithms and machine
production of food will need to be increased in the learning (e.g., deep learning methods) to IoT data in order
forthcoming years.To satisfy the demands of this larger, more to achieve a nutritional solution for smart agriculture [24].
urban, and wealthier population, agricultural production The repair and maintenance issue requires frequent sensor
needs to quadruple by 2050 [11]. More specifically, the checks for each IoT device, which can be easily harmed in
present annual wheat output of 2.2 billion tons should the agricultural field.The connectivity issues are related to
be increased as needs grow to almost 3.0 billion tons, the variety of wireless technologies (e.g., 4G, 5G, Zigbee,
and yearly livestock production should increase by over WiFi, LoRa, and 6LowPan) that are capable of connecting
225 million tons in order to meet the consumption of sensors spread across a vast region in agriculture [25].
500 million tons [12].The IoT has currently established a The infrastructure issues are related to the setting up
big influence on the agricultural sector, with an enormous and implementation of IoT network architecture employing
variety of sensors employed to accomplish different smart cutting-edge technologies such as fog computing, cloud
agriculture purposes. A significant amount of research has computing, virtualization of networks, and more [26]. The
been conducted in the agricultural field with Internet of key challenge in the development of environmentally friendly
Things technologies for implementing smart agricultural IoT-based agriculture is not physical backing; it’s more about
solutions. Every year, the number of IoT applications rises ensuring both security and privacy. Through the widespread
dramatically. By investigating numerous challenges and implementation of IoT-based agriculture, an attacker might
obstacles in farming, the IoT has contributed to a big shift in notice new ways to breach the system (for example, through
the field of agriculture [13]. With the advent of technological a fake data injection attack), generating significant security
advances, it is now predicted that farmers and technological and privacy concerns and advocating for secure sharing of
experts will leverage IoT to overcome challenges that information in the smart agricultural field [27].The taxonomy
farmers face, such as water shortages, managing costs, and of the IoT based agriculture system is depicted in Fig.1.
productivity challenges. All of these concerns have been The aim of the proposed study is to publish the results
addressed by cutting-edge IoT technologies, which have of a systematic literature review (SLR) in the field of IoT
provided solutions to boost production while decreasing in agriculture. On this subject, numerous research studies
costs [14].The worldwide smart farming sector is predicted have recently been published. The state-of-the-art research is
to grow to $21 billion by 2030, up from $6.5 billion in 2019 gathered, examined, and summarized in this SLR. This SLR
[15]. Smart agriculture is expected to become a significant has been carried out to gather and integrate recent practical
internet of things domain in countries that export agricultural studies with scientific panoramas so that other scientists and
products. Internet of Things applications for smart agriculture practitioners can find directions for implementing IoT in
have recently been introduced employing wireless sensor agriculture. This SLR research is the most comprehensive that
networks (WSNs), including irrigation sensor networks [16], has been carried out based on the potential of IoT applications
soil farming precision [17], frost event prediction [18], smart in agricultural systems. The major contributions of this study
agriculture precision [19].Research carried out on wireless are outlined below:
sensors Networks allow users to gather data from sensors • According to our research investigation, this is the
and transmit it to central servers [20]. Data received by first survey that offers extensive comparisons of the
sensors provides insight into various environmental variables, IoT-based applications for agriculture based on research
allowing the entire system to be effectively monitored. that has been previously published.
Monitoring weather conditions or yields of crops is not just • The foreseeable future of smart agriculture is addressed
one consideration for crop evaluation; instead, it includes in this study, along with its magnificent potential to
numerous additional factors that impact crop productivity, change people’s lifestyles by delivering food with high
such as land and field management, crop and soil monitoring, yields. Analyzing the most recent global state-of-the-
motion of a not-wanted object, attacks by wild creatures, and art developments in IoT-based agriculture and also

VOLUME 12, 2024 60987

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

FIGURE 1. Taxonomy internet of things based agriculture system.

highlighting the requirements for effectively implement- the architecture of an IoT based agriculture system.
ing these innovative technologies. Section IV-B describes the communication technologies
• The paper discusses commercially available agriculture- of an IoT based agriculture system. Section V discusses
based IoT sensors and devices, communication tech- security and privacy challenges and solutions for smart
nologies, and infrastructure for the agriculture-based agriculture. Section VI discusses state of the art machine
Internet of Things. learning models for smart agriculture systems applications.
• The paper highlights risk models, security and privacy Section VII discusses Research open challenges and future
challenges, and blockchain-based Agriculture IoT secu- research directions Finally, Section VIII concludes the
rity solutions in depth, focusing on IoT-based agriculture paper.
applications as well as future research directions for
smart agriculture systems.
Overall, these contributions offer a comprehensive and II. RESEARCH METHODOLOGY
forward-looking analysis of IoT-based applications in agri- A systematic literature review (SLR) is a methodical and
culture, covering technological advancements, implementa- structured way of finding, assessing, and interpreting prior
tion requirements, security challenges, and future research research that is pertinent to a certain research question
directions. This holistic perspective distinguishes it from in software engineering, according to [28] and [29]. The
existing surveys by providing novel insights and addressing literature has suggested a number of strategies for conducting
key aspects that may have been overlooked in previous an SLR. This systematic review was carried out in accordance
literature. This paper is organised as follows: Section II with the Preferred Reporting Items for Systematic Reviews
discusses research methodology with relevant research and Meta-Analyses reporting checklist [30]. The research
questions, search strategy, inclusion and exclusion cri- includes the development of Research Objectives, Research
teria, screening and selection, and quality assessment. Questions with motivation, a detailed Search String strategy,
Section III highlights the recent developments and trends in screening of the searched results, Data extraction, and
IoT-based smart agriculture systems. Section IV describes evaluation, as illustrated in Fig.2

60988 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

FIGURE 2. Key steps for systematical literature review [28].

A. RESEARCH OBJECTIVE
The main objective of this SLR is to examine and summarise
the state of the art in the subject of smart agriculture. In order
to accomplish this, prominent research objectives (ROs) have
been established to direct the review procedure and serve FIGURE 3. Count of articles as per time line.
a well-structured framework for analysis and interpretation.
To address important elements of smart agriculture, the
following research goals have been established that have been for relevant and representative papers, a unified searching
listed in Table 1: methodology was employed. The main databases of scientific
publications, including IEEE, Scopus, and MDPI, were
TABLE 1. Table of research objectives for IoT based agriculture SLR.
utilized for the search. The selected time-frame for inclusion
of publications in this SLR spanned from May 2018 to May
2023. By employing this comprehensive search strategy, the
aim is to capture the most up-to-date and relevant literature
in the field.

D. IDENTIFYING RELEVANT ARTICLES

Once the initial search was complete, it became clear
that not all of the papers returned were directly related
to the research’s objectives. It was required to carefully
evaluate their actual importance as a result. The first round
B. RESEARCH QUESTIONS of screening consisted of choosing articles based on their
Creating research questions is essential to carrying out titles, which allowed us to omit papers that were considered
a systematic literature review. Research questions provide irrelevant to the research topic, as mentioned in Table 4.
the review with a distinct and focused direction, directing After the initial search, screenings were carried out in four
everything from study selection to data analysis. Reviewers stages. First, duplicate research papers are discarded. There
can limit the scope of the review and make sure that only were 1776 papers after that stage. The duplicate and unrelated
relevant articles are included by developing clearly stated papers were eliminated based on their titles. After the process,
research questions. The following Table 2 shows the research 241 papers remained. Thirdly, papers were screened based on
questions (RQs) that have been created for the study. These abstracts that contained an extensive amount of information
research topics allow for the categorization of current IoT relevant to the paper and were retained for further analysis.
research in agriculture and the identification of potential After this stage, there were still 124 articles on the list.
future study fields. Finally, during the full-text search stage, papers that did not
address the research topics were discarded. The research
C. SEARCH STRING was carefully evaluated to establish whether or not it was
The second phase of the SLR focuses on the primary experimentally validated. The 96 papers reached a conclusion
goal of finding relevant research related to the research after this stage was completed, as presented in Table 5. Fig. 3
topics. This is achieved by gathering published papers that shows the results of a particular search string for publications
align with the specific research areas using a carefully published between 2018 and 2023, and Fig.4 shows the ratio
designed search string. During the initial search, keywords of papers published in conferences and journals, while Fig.5
were used to narrow down the focus to IoT applications displays the number of publications published by each search
in agriculture. However, the pilot search also included domain.
IoT communication protocols, agricultural sensors, and IoT
security and privacy measures to ensure comprehensive E. QUALITY ASSESSMENT
coverage. To gather information, a thorough internet research Quality assessment plays a pivotal role in the systematic
approach was adopted, utilizing various search engines and literature review process, and thus, a questionnaire was
digital libraries, as mentioned in Table 3. In the search developed to evaluate the quality of the chosen papers. The

VOLUME 12, 2024 60989

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

TABLE 2. Table of research questions for IoT based agriculture SLR.

TABLE 3. Selection of search strings for different libraries.

FIGURE 4. Ratio of articles: Journals vs. Conferences.

TABLE 4. Result of search strings from different libraries.

TABLE 5. Count of articles after screening results.

FIGURE 5. Count of articles from agriculture-based IoT domain.

(b) Evaluating whether the study offers a clear solution to

agricultural challenges using IoT The scoring system
incorporates 1 for ‘‘Yes,’’ 0.5 for ‘‘Partially,’’ and 0 for
‘‘No.’’
questionnaire comprises several questions designed to assign (c) Examining whether the study’s outcomes are empir-
scores for quality assessment. These inquiries encompass: ically validated A score of 1 is assigned to ‘‘Yes’’
(a) Assessing the study’s contribution regarding the explo- responses and 0 to ‘‘No’’ responses.
ration of IoT’s potential in agriculture A score of 1 is (d) Examining whether the study effectively bridge the
allotted for ‘‘Yes’’ responses and 0 for ‘‘No’’ responses. domains of agriculture, IoT technology, and relevant

60990 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

disciplines The scoring system incorporates 1 for ‘‘Yes,’’ systems that modify irrigation based on plant needs [32].
0.5 for ‘‘Partially,’’ and 0 for ‘‘No Farmers may now manage irrigation from anywhere
By scoring these questions, the selected papers’ qual- thanks to standard features like remote monitoring and
ity assessment is classified, ensuring the inclusion of management via mobile application.
high-quality studies in the SLR analysis. • Soil Management: Agriculture depends heavily on
maintaining healthy soil. Real-time soil monitoring
III. RECENT DEVELOPMENTS AND TRENDS IN IOT-BASED using IoT technologies makes measuring variables,
SMART AGRICULTURE SYSTEMS including moisture content, pH levels, and nutrient
By incorporating cutting-edge technology like the IoT, data levels possible [33]. Data is gathered by soil sensors,
analytics, and automation with conventional agricultural which are then analyzed to produce valuable insights.
practices, smart agriculture, sometimes called precision A recent trend is combining IoT-based soil manage-
agriculture, has completely transformed the farming sector. ment systems with machine learning algorithms to
Smart agriculture systems based on the IoT have been provide individualized suggestions for soil adaptation
increasingly popular in recent years because of their potential and fertilization. These systems encourage sustainable
to increase productivity, cut down on resource waste, and farming practices by maximizing nutrient utilization and
increase crop yields [31]. The most recent innovations lowering the chance of over fertilization.
and trends in IoT-based smart agriculture systems will • Weather Management: The weather is crucial to
be examined in this SLR, emphasizing six major ones: agriculture, and IoT has improved the accuracy and
water management/irrigation management, soil management, accessibility of weather monitoring. Real-time informa-
weather management, nutrient management, waste manage- tion on temperature, humidity, wind speed, and rainfall
ment, and crop management,as illustrated in Fig.6. is provided by IoT-based weather stations that are
equipped with a variety of sensors [34]. Through mobile
apps and online platforms, farmers have access to this
data, enabling them to make educated choices regarding
pest management, planting, and harvesting. Predictive
analytics and meteorological data have recently been
combined to assess climate threats and predict weather
trends. This allows farmers to adjust to shifting weather
patterns and reduce crop losses.
• Nutrient Management: Crop health and production
improvement depend on effective nutrient management.
The Internet of things may continuously monitor
nutrient levels in the soil and plants. Farmers may
improve their fertilization plans while minimizing waste
and environmental effects with sensors and data analyt-
ics [35]. Smart drones and autonomous vehicles are two
recent advancements in IoT-based fertilizer management
that allow exact fertilizer application. These technolo-
gies help reduce resource waste and environmental
damage by enabling site-specific fertilizer delivery.
• Waste Management: Keeping agricultural waste to
a minimum benefits both the environment and the
economy. Smart agriculture technologies driven by the
FIGURE 6. Applications of IoT based smart agriculture. Internet of Things aid farmers in maximizing resource
use and minimizing waste. IoT sensors have been used
• Water Management /Irrigation Management: Agri- recently to track waste produced during manufacturing
culture is becoming increasingly concerned about the and keep an eye on post-harvest storage conditions.
limited supply of water. Therefore, effective water Systems for intelligent sorting and recycling have also
management is essential. Smart irrigation systems that been created to effectively handle agricultural trash [36].
are IoT-based have become a significant solution. These developments lessen agriculture’s environmental
These systems optimize irrigation schedules using sen- impact and support sustainable farming methods.
sors, weather predictions, and data analytics, lowering • Crop Management: Smart agriculture is leading the
water use and boosting agricultural yields. Recent way with IoT-based crop management systems. These
innovations include incorporating artificial intelligence systems use sensors to monitor crop health, develop-
(AI) algorithms to forecast soil moisture levels, real- ment, and pest infestations. The use of multi-spectral
time water quality monitoring, and automated control cameras on drones to take high-resolution photos of

VOLUME 12, 2024 60991

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

crops is a recent discovery [37]. Then, preemptive

interventions are made possible by AI-powered picture
analysis algorithms that spot early indications of illness
or stress [38]. IoT-based agricultural management sys-
tems also frequently connect to equipment for automated
harvesting and precise planting. These developments
save labor expenses while improving crop quality and
output overall.
IoT-based smart agriculture systems are constantly chang-
ing due to new trends and advancements in many areas of
agricultural management. With the help of IoT technology,
farmers can now make data-driven decisions, increase
resource efficiency, and implement sustainable farming
methods in water, soil, weather, nutrients, waste, and crop FIGURE 7. Five layer architecture for an IoT-based agriculture system.
management. In addition to boosting agricultural output,
these developments are also helping to preserve natural
resources and lessen their harmful effects on the environ-
ment [39]. The future of smart agriculture appears bright, support, formulations, cloud computing, and middle-ware
with even more breakthroughs on the horizon as IoT technologies. It is able to manage and support numerous
technology develops. lower-layer services, such as alerts, retrieval of data, big
data processing, cloud computing, and data analyses for
IV. DIFFERENT LAYERS OF THE INTERNET OF THINGS
IoT-based agriculture applications via cloud platform ser-
ARCHITECTURE USED IN SMART AGRICULTURE SYSTEMS
vices like Amazon Web Services, Microsoft Azure, Google
To guarantee the highest level of reliability, considering the Cloud, and others. The application layer is responsible for
extensive variety of agricultural systems, it is essential to offering application-specific services to the user. This layer
manage numerous pieces of agricultural equipment and sen- includes devices such as monitoring equipment, geolocation
sors that are connected to the Internet in a heterogeneous way. devices, surveillance systems, crop diagnostic systems, and
An adaptive, multilayered framework is therefore essential. so on [41].The business layer manages all aspects of the
Each framework, which builds on IoT protocol stacks ranging IoT-based agriculture system, which include apps, security,
from three to five layers, provides various advantages. business and profit models, and user privacy. Each feature
Although the three-layer (perception, network, and applica- introduces new security and privacy concerns. As such,
tion) stack has a simplified structure, the four-layer stack security and privacy issues are classified accordingly based
provides additional services,and the five-layer framework on their prevalence in each layer [42].
has the greatest potential for agriculture. Precision farming,
resource optimization, and environmental monitoring depend A. LEADING SENSOR TECHNOLOGIES USED FOR DATA
on higher levels of data interpretation and interoperability, COLLECTION AND MONITORING IN
which makes the five-layer IoT protocol stack the best AGRICULTURAL FIELDS
option for agricultural IoT applications given the complexity In agriculture and IoT infrastructure, the perception layer is
of agricultural operations and the wide range of sensors the physical layer that collects data from the world using
and devices involved. The five-layer IoT-based agriculture sensors, imaging devices, and other technological equipment.
proposal proposes an organized structure with each of These peripherals are in charge of gathering data on moisture
the layers carrying out a specific task [40]. According to in the soil, relative humidity, temperature, quality of the
Fig.7, the five layers include perception, communication air, water levels at ground level, and other parameters
network, platform (processing), application, and business that are essential to monitoring and controlling agricultural
layers. Hardware such as sensors, actuators, and controllers procedures [43]. The perception layer is the fundamental
are included in the device layer, which is also referred component of an agricultural IoT system because it acquires
to as the perception layer. The communication network data from the field in real time. The data is then sent to the
layer is in charge of connecting to servers, other smart IoT higher levels of the system to be processed, evaluated, and
sensors and devices, and network equipment. In addition, the applied to make decisions. Farmers and agricultural experts
processing and sending of sensor data enables effective use may decide alternatives about cultivation, fertilisation, pest
of its strengths. Networks (such as RFID, Bluetooth, NFC, control, and other vital aspects of farming when they have
WiFi, 5G, and LAN) are used to send sensor data from the access to timely and reliable data about the agricultural
layer of perception to the platform (processing) layer.The environment. The following is a list of some devices and
data processing (middle-ware) layer stores, analyzes, and sensors that are frequently used in the perception layer for
processes gigabytes of data while also providing service agricultural purposes:

60992 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

1) Micro controller ESP8266: The ESP8266 is an 11) Wind sensors: Keep track of the wind’s direction and
autonomous system on chip (SOC) connected with speed, which can affect the way crops are pollinated,
TCP and IP that enables the micro-controller to estab- how quickly water evaporates, and how far pesticide
lish a connection to a Wi-Fi network.The ESP8266 may spraying drifts.
enable or block all Wi-Fi network functions from other 12) pH sensors: These devices measure the level of acidity
application processors. or alkalinity of the soil and provide information about
2) Micro controller Arduino Mega: The Arduino Mega how this impacts the amount of nutrients that are
is an AT-mega-based micro-controller board with available to plants.
16 analog pins, 54 digital input and output pins, 13) Nutrient sensors: These devices monitor key elements
a crystal oscillator running at 16 MHZ,a USB port such as nitrogen, phosphorus, and potassium in the soil,
for communication, an ICSP header, a barrel jack for enabling more exact fertiliser techniques.
power, and a reset push button. 14) Leaf wetness sensors:These sensors identify the
3) The nRF24L01 radio transceiver module: The presence of moisture or water retention on plant leaves,
Rf24L01 module is a 2.4 GHz wireless single chip with which is crucial for preventing diseases and estimating
a baseband protocol engine. The wide range of In order the optimal time to spray herbicides on plants.
to implement smart farming, a 1100 m long transceiver 15) Crop health sensors: A variety of sensors or imaging
module that works in the ISM frequency band is technologies are used to measure the health of crops
necessary.It is in charge of transmitting agricultural by examining chlorophyll stages, leaf colour, or other
data to the layer of application. IPv4 and IPv6 versions environmental stress or disease signs.
are available to offer technological features such as 16) Water flow sensors: Track water flow in irrigation sys-
network layer communication. The IPv6 protocol was tems to identify leakage and improve the distribution of
developed in response to the growing number of water.
addressing devices. In this scenario, the WiFi module 17) Evaporotranspiration (ET): Measure the rate of
ESP8266 communicates with the network layer using water loss from the soil due to evaporation and plant
IPv4 at 2.400 - 2.4835 GHz. A serial peripheral transpiration. This helps with irrigation schedules.
interface was used to adapt and control this wireless 18) Tensiometers: Determine the soil moisture tension,
transceiver module. which signifies the force needed by plants to draw
4) Soil moisture sensors: These devices measure the water from the soil and facilitates decisions on
amount of water that is already present in the ground, irrigation.
thereby assisting farmers in determining when and how 19) Water level sensors: Ensure that there is sufficient
much water to use for irrigation. water for irrigation by monitoring the water level in
5) Temperature and humidity sensors:Monitor the reservoirs, tanks, or other sources of water.
environmental conditions to gain an understanding of 20) Soil compaction sensors: Determine the degree to
the relationship between climate and the growth of which the soil is compacted, which can have an effect
crops. on the growth of plant roots and the overall health of
6) Air quality sensors:Detect pollutants or gases that the crop.
may impact the health of plants or the air quality in the 21) Weather radar or satellite sensors:They should be
surrounding area. used to provide more comprehensive weather data and
7) Groundwater level sensors: These devices are used to forecasts in order to better guide agricultural decision-
measure the levels of groundwater, which is essential making.
for the management of water resources and the 22) Carbon dioxide (CO2) sensors: Optimise plant devel-
prevention of over-extraction. opment and photosynthesis by measuring the amounts
8) Weather stations: These stations are responsible for of carbon dioxide in greenhouses and other controlled
collecting a wide variety of data pertaining to the environments.
weather, such as humidity, wind speed, temperature, 23) NPK sensors: The levels of nitrogen, phosphorus, and
and amount of rainfall. potassium in the soil are measured and monitored with
9) Remote sensing devices: These are sensors that can the assistance of NPK sensors, which are utilized in
be based on drones or satellites that take images the agricultural industry. With this insight, farmers
and multidimensional data. These sensors are used to have the ability to make intelligent choices regarding
evaluate the health of crops and find early indicators of fertilization, leading to improved plant growth and
stress or disease. increased agricultural yields.
10) Light sensors: These sensor help farmers optimise The data that these sensors and devices receive is often
planting places by measuring the intensity of sunshine sent wirelessly to a central data processing system, where it
and regulating artificially generated light in artificial is analysed, displayed, and used to start automated actions or
environments such as greenhouses make suggestions for farmers. Farmers can use the data from

VOLUME 12, 2024 60993

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

1) PERCEPTION LAYER PROTOCOLS FOR AGRICULTURE IOT

Perception layer protocols typically use the IEEE 802.15.4
standard, which is well-known for its simplicity, low cost,
and low power consumption.This standard enables wireless
communication at a significantly lower data rate and was
designed specifically for Internet of Things devices with little
bandwidth [47]. It specifies the PHY (physical) and MAC
FIGURE 8. Layer-wise communication protocol taxonomy. (media access control) layers for an extensive variety of
devices, such as fixed, portable, and mobile devices with low
power consumption. The IEEE 802.15.4 standard implements
a wide-band physical layer using the Direct Sequence
these sensors to make data-driven choices and use precision Spread Spectrum (DSSS) approach.Three frequency bands
agriculture techniques to boost productivity and sustainabil- are used to govern devices operating on the physical layer:
ity. Selected publications have been summarized in Table 6 (a) the 915 MHz band, which is licensed in the US;
in order to evaluate the essential sensor technologies of IoT (b) the 868 MHz band, which is licensed in Europe; and
in agriculture. (c) the 2.4 GHz ISM band, which is an unlicensed but
frequently used band. There are 27 broadcast channels acces-
B. COMMUNICATION PROTOCOLS AND APPLICATION sible across these three bands. The IEEE 802.15.4 physical
LAYER PROTOCOLS ARE FREQUENTLY USED IN layer covers a variety of low-level functions, such as sending
IOT-BASED SMART AGRICULTURE and receiving data, signal strength, monitoring,channel
Various protocols have been adopted by the IoT network to energy sensing, clear channel analysis, and more. whereas
transmit data, and these protocols are essential in deciding the IEEE 802.15.4 data link MAC layer is in charge of
how things communicate during data transmission. These tasks such as establishing the Personal Area Network (PAN),
protocols are a set of syntax and logical principles that control enabling GTS data transmission,implementing Carrier Sense
how the machine’s network operates when exchanging data. Multiple Access with Collision Avoidance (CSMA-CA) for
The widespread use of several sensors that are challenging channel allocation, establishing reliable links between peer
to transform into the standard addressing format makes entities, managing control repeater transmissions, and ensur-
setting up an IoT network complicated, even though common ing repeater synchronization. IEEE 802.15.4 is compatible
Internet protocols (IPs) fall short in ensuring efficient with and well-suited for several IoT protocols, including
data transmission. This constraint makes it difficult to ZigBee, LoRA,Bluetooth, and LowPan [48]. Agriculture
set up adequately functional nodes [44]. Since IoT nodes systems frequently use perception layer protocols to reliably
significantly depend on a continuous power supply, particular and efficiently acquire data from sensors in the fields.
memory characteristics, and channel throughput capacity, • Bluetooth: Bluetooth is a form of wireless communi-
efficient resource management is crucial to overcoming these cation involving radio signals at ultra-high frequencies
difficulties. The addition of a data sink to the network is (UHF) to assist in communication over short distances.
required in the setting of WSNs. This procedure entails Bluetooth enables wireless communication between
gathering data from numerous nodes and sending it to the numerous agricultural IoT devices [49]. Bluetooth
sink for additional processing. The significance of opting allows connections between devices up to 100 meters
for an adequate data transmission technology is highlighted apart by using the 2.4 GHz radio frequency. It secures
by the effective placement of sensors and sinks, which can data transfer using authentication and encryption tech-
considerably improve the ability to transmit data and the niques, and its key benefits are its low cost and
efficiency of an IoT network [45]. power efficiency. Bluetooth is more suited for local
Security and privacy are two essential characteristics that communication within the farm or a particular location
such a technique ensures by restricting different sensors from because of its constrained range.
exchanging the same data [46]. It also requires preserving • RFID (Radio-Frequency Identification): RFID is a
energy, which is an essential priority in IoT applications. Pro- technology that grants every object in the data it records
tocols for communicating with the agriculture-based IoT can a special identification number. RFID tags can be used in
be classified into three main classifications that are related to agriculture IoT to track and monitor a variety of assets,
the perception, network, and application levels of the IoAT including livestock, products, and agriculture machin-
architecture, as presented in Fig.8. Table IV-D summarizes ery. RFID tags exist in a variety of sizes and shapes, with
the properties of the protocols used in IoAT at the perception passive tags being less expensive than active tags. These
and network layers. This research further illustrates the tags may provide crucial information about the tagged
adaptability and potential advantages of these protocols objects, like moisture content, temperature, humidity,
across various sectors by providing agriculture-based IoT and other pertinent details. Real-time asset tracking and
indications of how each protocol might be implemented in environmental monitoring are two areas where RFID
the agricultural sector. technology is extremely beneficial [50].

60994 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

• Infrared: Infrared communication is an inexpensive and • LoRaWAN: LoRaWAN is a low-power, long-range

low-cost technique for wirelessly sending small amounts wireless network based on long-range chirp spread
of data. It appears in agriculture and IoT, such as in spectrum (CSS) modulation and is optimal for IoT
thermometers and cameras that monitor temperature and applications. It is implemented in a variety of sectors,
ambient conditions [51]. Infrared technology is widely including agriculture, smart cities, metering, logistics,
utilized in remote controls and simple electronics, with and others. LoRaWAN enables efficient as well as
data transmitted using protocols such as NEC and reliable communication, dealing with significant chal-
RC5. It is appropriate for short-distance communication lenges including energy management, the preservation
within a narrow area. of natural resources, and disaster prevention [55].
• Ultra-Wideband (UWB):The wireless protocol known • Zigbee: Zigbee is a well-known standard for transmit-
as ultra-wideband is intended for close-proximity com- ting data between agricultural devices. Zigbee, which
munications between devices. UWB, which operates operates on the 2.4 GHz frequency band, has a wider
at a greater frequency range of 3.1 to 10.6 GHz, communication range than Bluetooth. It employs a mesh
allows for accurate measurements of distance between network topology, comprising components such as end
transmitters. This high degree of precision is obtained devices, network routers, and processing gateways. The
by transmitting billions of radio pulses over a wide numerous advantages of Zigbee include low power
frequency band. In the IoT for agriculture, UWB can consumption, rapid transfer rates, and high network
be employed in applications such as asset tracking and throughput [56].
precision farming that require precise positioning and • 6LoWPAN: 6LoWPAN, also known as IPv6 over
location tracking [52]. Low-Power Wireless Personal Area Networks, is a low-
power wireless mesh network that gives each node
In conclusion, the perception layer protocols for the a distinct IPv6 address. This makes it possible to
agriculture IoT include Bluetooth for short-range com- connect directly to the internet using open, standard-
munication, RFID for asset tracking and environmental ized protocols. With the help of 6LoWPAN, the IoT
monitoring, Infrared for simple data transmission, and ecosystem is accessible to even the smallest devices
Ultra-Wideband for applications requiring accurate location with minimal computational power [57]. Standard IPv6
and positioning. Each protocol serves a distinct function, packets are sent through IEEE 802.15.4-based networks
enhancing the effectiveness and automation of agricultural using encapsulation and header compression techniques.
activities. 6LoWPAN enables seamless sensor communication
with middleware platforms and network routers for
2) NETWORK LAYER PROTOCOLS FOR AGRICULTURE IOT agricultural sensors, enabling them to connect to IP
In agriculture IoT, the network layer encompasses essential networks.
technologies such as gateways, networks of access points, • NB-IoT(Narrowband IoT): NB-IoT is a wireless
and routing devices that are accountable for internet protocol communication technology intended primarily for low-
(IP) addresses and other networking operations. A variety power, wide-area IoT applications [58]. It works on
of protocols are implemented at this level, including WiFi authorized wireless bands, which provide greater con-
and Zigbee, which are becoming the most ubiquitous for nectivity and penetration, making it appropriate for
agricultural IoT. Furthermore, LoRaWAN and 6LoWPAN are agriculture internet of things implementations in distant
employed to simplify communication with sensor networks and rural locations. NB-IoT is employed in applications
that are wireless [53]. Remote data transfer and connectivity such as soil moisture monitoring, precision farming, and
can be accomplished using conventional cellular wireless smart drip irrigation systems.
technologies such as GPRS or 3/4/5G. • LTE-M (Long-Term Evolution for Machines): LTE-
M is a wireless connection technology intended exclu-
• WiFi: The IEEE 802.11-based WiFi standard spans sively for IoT devices. It has faster data rates and lower
a broad spectrum of frequencies, from 5 GHz to latency than NB-IoT, making it suited for real-time
2.2 GHz. WiFi can communicate over a distance of data transfer applications [59]. In agriculture IoT, LTE-
20 to 100 meters with data transfer speeds ranging M is used to track livestock, monitor equipment, and
from 1 Mb/s to 7 Gb/s. Since WiFi connects remote manage storage on enormous farms and plantations for
monitoring and control devices efficiently and securely, cultivation [60].
it has become commonly used in IoT-based agriculture • Sigfox: Sigfox is a low-power, wide-area network tech-
systems in order to connect and monitor various nology that allows IoT devices to communicate across
assets such as irrigation systems, weather stations, vast distances. It is appropriate for IoT applications
and agricultural machines [54]. This enables farmers in agriculture that need cheap and energy-efficient
to access real-time data, manage equipment from connectivity. Sigfox technology is capable of helping
a distance, and streamline agricultural practices for track assets, monitor conditions in the environment, and
increased effectiveness and productivity. deploy smart drip irrigation systems.
VOLUME 12, 2024 60995
 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

• LTE Cat-M1: LTE Cat-M1 is a low-power, wide-area that corresponds to the IETF (Internet Engineering Task
wireless communication technology that improves IoT Force) standards for cross-messaging, telepresence, and
device coverage and battery life [115]. It is well-suited video and audio calling, regardless of the existence of
for agriculture IoT applications that require dependable, modern protocols such as MQTT. The dependability and
long-range communication, such as asset tracking, soil flexibility of XMPP enable the development of further
monitoring, and remote equipment management. IoT applications [119].
These network communication technologies provide a • Advanced Message Queuing Protocol (AMQP):
diverse set of possibilities for agricultural IoT infrastructure, AMQP has been developed with the demands of the
adhering to a variety of connectivity requirements, power industry in mind, offering message routing, queueing,
consumption, and geographical regions. Each technology has switching, privacy, security, and trustworthiness. The
its own distinct capabilities, allowing farmers and agricultural publish-subscribe mechanism employed by AMQP,
enterprises to develop efficient and integrated IoT solutions which is like XMPP, ensures effective message delivery
for increased production and efficiency. with at-least-once, at-most-once, or just-once defini-
tions. The store-and-forward aspect of AMQP, which
3) APPLICATION LAYER PROTOCOLS FOR AGRICULTURE IOT ensures accuracy and reliability, is one of its most
Table 8 comprises fundamental technical characteristics to significant advantages [120]. However, it may cause
accompany the application layer protocols that are frequently network disruptions. Transmission Control Protocol
used in IoT-based agriculture applications. These protocols (TCP) is the backbone feature of AMQP for secure data
are essential for ensuring connectivity among agricultural transfer and reception.
equipment and enabling effective data exchange and control. • Hypertext Transfer Protocol (HTTP): HTTP is a
A variety of application communication protocols are being commonly employed application protocol for browsing
designed in order to meet the diverse requirements of IoT the web, but it is also utilized in IoT for agriculture.
applications for agriculture as the IoT keeps evolving [116]. It allows web-based apps and IoT devices to commu-
nicate, enabling farmers to acquire data and operate
• Constrained Application Protocol (CoAP): CoAP machinery for agriculture through internet interfaces.
is intended for IoT communication demands that HTTP is extremely advantageous for handling IoT
are responsive to traffic congestion-induced efficiency gateways, cloud-based data storage, and agricultural
decline. It runs as a low-bit-rate web transport protocol, remote monitoring systems [121].
making it ideal for devices that have limited compu-
tational capability and memory [117]. User Datagram Various usage scenarios, device capacity, and network
Protocol (UDP) is the framework upon which CoAP constraints all play an essential part in determining the
has been established for transport, resulting in minimal application layer protocol that is used in the agriculture IoT.
overhead and utilization of resources. It leverages a Farmers and agricultural enterprises may develop complex
request-response paradigm, rendering it suitable for a and effective IoT systems to increase production, optimize
variety of IoT applications in agriculture. The benefits utilization of resources, and boost sustainability overall by
of the star topology of this protocol include low power utilizing these communication protocols.
consumption, low cost, easy deployment, and security.
• Message Queue Telemetry Transport (MQTT): The C. DATA PROCESSING LAYER
publish-subscribe mechanism has been implemented by The Data Processing Layer plays an essential role in
the asynchronous MQTT protocol to enable seamless IoT-based agriculture by transforming contemporary farming
communication between machines. It integrates embed- techniques. This layer enables the smooth gathering of
ded network infrastructure into middleware systems and significant agricultural data by integrating numerous IoT
applications, which makes it appropriate for IoT imple- devices, such as sensors, drones, and weather stations. The
mentations with limited resources and low-bandwidth data gathered includes, among other things, data on soil
networked devices. When limited memory use, low moisture, temperature, humidity, crop health, and machinery
cost, and low-power devices are required, MQTT condition. The data is preprocessed after gathering to
performs exceptionally well [118]. It is adequate for guarantee correctness and consistency, and then it is stored
telemetry-style messaging from IoT devices to the server in databases [122]. The rapid analysis enabled by real-time
because it has a high latency or constrained bandwidth data processing enables farmers to make decisions quickly in
for the transmission of data. response to rapidly altering environmental circumstances.In
• Extensible Messaging and Presence Protocol order to extract useful insights from the gathered data, data
(XMPP): In an IoT context, XMPP is mainly employed analytics is a crucial part of this layer. By using statistical
for the exchange of messages. XMPP utilizes the analysis and machine learning algorithms, patterns and trends
publish-subscribe technique, which is more suited for may be identified. These patterns and trends can then be uti-
IoT applications than the request-response technique of lized to build prediction models for better agricultural yields,
CoAP. XMPP has been an endorsed internet protocol disease detection, and irrigation plan optimization [123].

60996 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

TABLE 6. IoT-based agriculture technologies classification and quality assessment.

VOLUME 12, 2024 60997

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

TABLE 6. (Continued.) IoT-based agriculture technologies classification and quality assessment.

60998 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

TABLE 6. (Continued.) IoT-based agriculture technologies classification and quality assessment.

VOLUME 12, 2024 60999

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

TABLE 6. (Continued.) IoT-based agriculture technologies classification and quality assessment.

61000 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

TABLE 6. (Continued.) IoT-based agriculture technologies classification and quality assessment.

VOLUME 12, 2024 61001

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

TABLE 6. (Continued.) IoT-based agriculture technologies classification and quality assessment.

The transformed data is then utilized to create decision Monitoring and managing livestock is a different group
support systems that provide farmers with actionable advice, of applications that is quite common. Farmers can monitor
enabling more effective resource management and crop man- the health and behavior of their animals, spot sickness early,
agement.Automation and control of agricultural activities and improve feeding and breeding procedures thanks to IoT-
also become possible through the data processing layer [41]. based solutions. These programs boost livestock agricultural
Farmers are able to manage irrigation, track the behavior of production overall, lower veterinary expenses, and improve
livestock, and improve the efficiency of machinery by feeding animal wellbeing.
the findings back into IoT devices and control systems.
However, data security and privacy have become crucial
because of the rise in the processing of sensitive agricultural E. BUSINESS LAYER
data. In order to ensure that the potential advantages The effective integration and application of the internet of
of IoT-based agriculture are achieved while preserving things in the farming and agricultural sectors is greatly
farmers’ sensitive data, the implementation of strong security impacted by the Business layer of IoT in agriculture. It entails
measures provides safety from unauthorized access and data developing strong business models, generating distinct value
breaches. propositions, and carrying out in-depth market research to
comprehend the requirements of farmers and stakeholders.
D. APPLICATION LAYER The Business Layer promotes the adoption of IoT solutions in
The data and insights from the Data Processing Layer agriculture by creating the appropriate business models, such
are converted into useful and approachable applications in as pay-per-use or subscription-based pricing, and delivering
IoT-based agriculture at the Application Layer. This layer attractive value propositions that highlight enhanced crop
entails developing of software solutions that provide farmers yields, resource efficiency, and cost savings [126]. The
and agricultural stakeholders with useful information, real- development of a strong IoT ecosystem for agriculture is
time monitoring capabilities, and decision-making tools to further enabled by creating strong connections and estab-
help them maximize farm operations. These programs are lishing partnerships with technology providers, data analytics
accessible to farmers on a variety of devices since they may companies, and equipment manufacturers. To make sure that
be accessed through web interfaces, mobile apps, or specific IoT solutions adhere to standard practices and data protection
platforms. Renowned IoT-based farm apps encompass an rules, the Business Layer also handles significant regulatory
enormous range of features [124]. Several applications are and compliance aspects [127]. Farmers may successfully
focused on precision farming, offering farmers specific install IoT technology on their farms by engaging individuals
information about soil conditions, weather forecasts, and crop and providing them with the technical support and continuous
health. These programs frequently provide functions like crop help they need. IoT solutions could adapt to changing
disease monitoring, automatic irrigation scheduling, and cus- market needs and evolving technology by developing a
tomized fertilization advice depending on the requirements of scalable and adaptable business strategy, permitting the
each crop. Applications for precision farming seek to increase agricultural sector to take advantage of IoT’s opportunities for
yields while reducing resource waste, helping to promote expansion, sustainability, and increased efficiency in farming
environmentally friendly and sustainable agriculture [125]. methods.

61002 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

TABLE 8. Summary of characteristics of IoT application layer protocols.

V. MAJOR SECURITY & PRIVACY CONCERNS IN IOT

ARCHITECTURE BASED ON SMART AGRICULTURE
According to modern agricultural studies, more than 90%
of the total IoT sensor communications throughout the field
of agriculture are carried out without adequate encryption.
According to this concerning estimate, around 57% of smart
devices deployed in agriculture are affected by vulnerabilities
in security that might expose confidential data [128]. The
disruption of agricultural systems caused by these breaches
TABLE 7. Summary of agriculture-based IoT perception and network layer protocols.

of security goes beyond that; there is also an imminent threat

to the security of individuals.In the agricultural environment,
cyberattacks may have broad implications, especially for
the lives of individuals. A cybersecurity breach has a
tendency to become devastating, particularly in instances
endangering individuals. The swift development and broader
implementation of the Internet of Things in agriculture,
especially in emergency situations such as infectious dis-
eases, has heightened security and ethical issues. Ensuring
the confidentiality of crucial and sensitive data related to
agriculture transforms into a harder challenge. Several differ-
ent kinds of attacks, hazards, and threats might be targeting
multi-layers of the IoT framework, demanding tight security
and privacy standards in the field of agriculture [129].
It is essential to apply techniques like cryptographic and
non-cryptographic algorithms to efficiently identify and
prevent invasions. There are also plenty of malware attacks
that can threaten the security of data, authenticity, reliability,
and accessibility that have been identified as consequences
for IoT systems. In accordance with the aforementioned
security-related issues, the agricultural operation emphasizes
key management, detection of intrusions, authenticating
measures, and controlling access in its existing security

VOLUME 12, 2024 61003

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

strategy. This preventive approach is critical for guaranteeing enables the seamless synchronization of many opera-
the confidentiality of agricultural data and the continuous tions, eliminating devices from waiting endlessly since
trustworthiness of IoT devices in the agricultural sector. The data is processed [132]. It allows tractors, for example,
Table 9 highlights the latest research on IoT-based agriculture to keep cultivating or harvesting when data concerning
security challenges and effective solutions. the soil is investigated and adjusted in the meantime.
Spooling is essential for increasing the efficiency of
A. MULTI-LAYERED SECURITY THREATS / ATTACKS contemporary agricultural practices since it helps to
An IoT-based agriculture system has plenty of security reduce downtime and makes sure that equipment is used
challenges, including data collection, storage, processing, effectively in accordance with real-time data inputs.
and transmission via Internet access, as well as malicious • Node Capturing/ Node Outage: Node capture or
application and business data activity [130]. In a multi- node outage is a significant threat in agricultural
layered framework, Fig.9 depicts vulnerabilities in security networks of sensors and IoT implementations [133].
in the IoT-based agricultural system. Security flaws are Numerous sensors or data-gathering sensors scattered
often inadvertently or unexpectedly detected. Agriculture over farming land to track different indicators such as
surroundings may be accessed by animals, farm workers, moisture in the soil, temperature, and the condition of
and machine equipment, all of which might generate crops are referred to as nodes. When these nodes are
issues. A significant number of security risks are extremely deliberately manipulated or taken away, data gathering
widespread; however, several are specific to individuals is disrupted, and the reliability of agricultural tasks and
who work in challenging environments such as IoT-based decision-making processes may be threatened. Node
agriculture. outage, on the other hand, refers to the temporary or
permanent failure of these nodes owing to technical
faults or environmental variables, resulting in gaps in
1) PERCEPTION LAYER THREATS
data collection and reducing the efficacy of precision
This is mostly around tangible devices, including sensors and
agricultural operations. Both node capture and node
actuators. Physical equipment may malfunction as a result
outages can result in inefficient resource management,
of malicious or unintentional actions by people, computer
lower agricultural yields, and financial losses.
viruses, malware, or cyberattacks. IoT-based agriculture
• Routing Attacks: An attacker generates network pack-
applications employ a diverse set of sensors and technology,
ets that are used to deceive different devices by altering
which introduces a number of security threats, such as the
or concealing the source address.Agriculture routing
ones listed below:
hacks include the malicious modification of networked
• Signal Jamming/Radio Jamming: Signal jamming, data channels. The security of agricultural operations
mainly radio jamming, causes an enormous risk to may be jeopardized as such attacks, which might result
agriculture by halting critical systems that include in inaccurate data routing, disrupt vital activities [134].
remote imagery for monitoring crops,GPS for precision It is critical to put adequate safety precautions in place
agriculture, and communication devices for cooperation. to prevent such attacks.
As a result of this disruption, location may not be • Threat to NDP Protocol: Numerous attacks may be
precise, data gathering may be interrupted, communi- attempted against the Network Dynamic Data Exchange
cation may be disrupted, and there may even be the (NDP) protocol, which is frequently employed in
possibility of damage or loss [131]. Signal jamming agricultural monitoring networks. Unauthorized parties’
may result in financial consequences, inefficiency in monitoring may threaten the security of essential agri-
operation, and security risks in the agricultural sector, cultural data. Implementing robust security methods like
which is more dependent on cutting-edge technology authentication, encryption, and detection of breaches,
such as unmanned agricultural machinery. To pro- as well as frequent monitoring and modifications,
tect agricultural operations and sustain production, is crucial to guaranteeing the authenticity of the NDP
mitigation solutions include reliability in navigation protocol [135].
and communication systems, signal monitoring, adher- • Data Transit Attacks: In the agricultural sector, data
ence to regulations, and precautions for physical transit attacks encompass a wide range of fraudulent
security. activities aimed at disrupting the reliability, privacy,
• Spooling: Spooling, which stands for ‘‘Simultaneous or accessibility of data once it is transmitted over
Peripheral Operations On-Line,’’ is a crucial procedure the internet [136]. Monitoring confidential data related
in the field of agriculture that facilitates the seamless to agriculture during transit is a prime instance of
control of information exchange and jobs. Spooling a breach that could lead to stolen information or
is the temporary storage of data within a queue or spying. Additionally, interference with data in transit
buffer until it is processed by agricultural equipment might cause data to be inaccurate, which could lead to
or computer systems, including planting guidelines, agricultural alternatives that are inadequately informed.
irrigation timetables, or harvest data. This approach To ensure the safe and reliable transfer of crucial

61004 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

FIGURE 9. Taxonomy of multi-layered security and privacy concerns in IoT based agriculture.

data in agriculture, powerful encryption technologies, discrepancies, and the possibility of data theft. A few of these
secure communication protocols, and extensive network are listed here.
security measures are required.
• Cross-Site Request Forger: On IoT equipment that
• Denial-of-Service (DoS) Attacks: A DoS attack pre-
implements RESTful APIs, that sort of attack appears
vents consumers from incorporating devices or other
to be more prevalent. The end-user is deceived by
resources on the network. Through overwhelming
replying to an insecure application without recognizing
intended devices or resources on the network and
it because of the CSRF method [138]. The web-based
numerous excessive requests, this activity is carried out
user interface of the IoT layer is at risk of CSRF attacks
in order to make it challenging or unattainable for a
if its configurations are not adequate.
variety of consumers to communicate [137]. An agricul-
• Cross-Site Scripting: An XSS attack potentially attacks
tural operation frequently includes a significant number
RESTful applications for the Internet of Things through
of interconnected nodes and categories; consequently,
inserting side scripting into websites to bypass accessi-
identical kinds of attacks are feasible in the environment
bility restrictions [139]. These kinds of attacks are made
of smart agriculture. Such attacks can merely affect
possible through the websites of cloud-connected IoT
the routine functioning of various units within a single
apps.
farming operation; however, they may also serve to
• Session Hijacking: This form of attack is widespread
interfere with legal cyber services across various areas.
among RESTful-based Internet of Things technologies.
Since several IoT gadgets retain session connectivity
2) COMMUNICATION LAYER THREATS at the website user interface sessions can be hacked
The communication layer security flaws in smart agriculture enabling a hacker to hack into session data [140].
are comparable to those discovered in various IoT applica-
tions when taking into account the architecture and usage of 4) APPLICATION LAYER THREATS
the technology. For instance, compared to other protocols, According to the increased scalability and adaptation of cloud
WiFi encryption is more susceptible to password cracking. computing, application companies are getting increasingly
There are some basic security issues at the communication interested in hosting applications there. This layer serves
layer mentioned below: as the intuitive interface for IoT-based agricultural devices
connected to users with the middle-ware. The kinds of
potential attacks on this outermost layer include:
3) MIDDLE-WARE LAYER THREATS
The security risks and threats related to the computing and • SQL Injection: A SQL injection vulnerability has been
software solutions that facilitate data transfer and processing identified in an agricultural management system. When
among agricultural devices and sensors and higher-level a hacker injects an erroneous SQL query into a web
applications are referred to as the middle layer of threats in server database. A resilient SQL injection attack might
agriculture. The security and integrity of agricultural oper- compromise or alter agriculture information, providing
ations may be threatened by these risks, which additionally a substantial threat to IoT devices, notably those used in
comprise hacking of data, control of access flaws, gateway agriculture [141].

VOLUME 12, 2024 61005

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

• Account Hijacking: Various IoT products employ especially susceptible access points for attackers, and agri-
inadequate security or transmit data in plain text over the cultural data might be stolen. A recent analysis of IoT-based
Internet. When a packet gets captured once a consumer agricultural devices found that 83% of them are using
has been verified authorized, a hacker is capable of insecure operating systems [147]. Farmers who use these
hijacking an account [142]. The most significant cause devices might be more vulnerable to attackers who leak
of this attack’s starting point, as reported in various critical agricultural data. As demonstrated in Fig.10, 98% of
cases, is outdated operating systems with insecure all communications between IoT devices are not encrypted,
patches. making 57% of these devices vulnerable to attackers and
• Malicious Hijacking:Installing malicious software that revealing sensitive data on the network. With the proliferation
conducts unlawful operations with the intent to trigger of connected devices, cyberattacks are expected to become
network disruption [142]. more frequent and severe [148].
• Ransomware:Encrypts significant data and intends to
make a significant payment to retrieve it. This threat
has the potential to initially appear on one machine and B. BLOCKCHAIN-BASED AGRICULTURE
spread across the network [143]. Hackers can encrypt IOT SECURITY SOLUTIONS
sensitive data and hold the code for decryption as a token In the IoT-based agriculture system, security threats depend
for cash. on the vulnerabilities that are available at several levels,
including applications and interfaces, network components,
5) BUSINESS LAYER THREATS software, firmware, and physical equipment. In order to
This layer fosters the agricultural services supplier’s enter- achieve a certain security level, mitigation strategies for secu-
prise logic and facilitates every aspect of the organization’s rity threats address the vulnerabilities of this interaction at
process, comprising surveillance, supervisors, and growth. several layers. These countermeasures are more complicated
It is also in charge of processing data related to agriculture as a result of the various deployment protocols.Blockchain
for findings. Threats on this layer are currently being technology is anticipated to have a significant impact on how
investigated, but the damage they cause is more significant agriculture-based IoT systems are managed, controlled, and,
as they include critical agricultural information. Potential most critically, secured. The expectation has been made by
violations include disclosure of data, deception, disruption, both the business and scientific communities. This section
and usurpation. explains how blockchain can be a crucial enabling technology
for offering realistic security solutions to the complex IoT
• Information Disclosure: Unauthorized manipulation of security issues that exist today.
confidential information, such as production records, A blockchain is a distributed, decentralized, and unchange-
threatens the security of the IoT-based agriculture able ledger that keeps track of assets and transactions over
system. A hacker might use previously described a network of peer-to-peer devices. It is composed of a
techniques, for example, session hijacking and CSRF, series of time-stamped and authorized data blocks that are
to obtain illegal access to confidential data [144]. encrypted by elliptic curve encryption (ECC) and SHA-256
• Deception: Disrupted data threatens the integrity of hashing. This innovation improves trust and transparency
data, which may result in serious consequences. Sink- by cutting out the requirement for centralized authorities
holes and attackers in the middle can both issue and empowering users to directly share and verify data.
deceptive data. Nearly fifty-eight percent of companies The block data basically consists of a hash of the previous
do not have a system in place to rectify erroneous block and a list of all transactions. The blockchain offers
information [145]. a cross-border decentralized trust around the world and
• Disruption: When accurate functions or accessibility to maintains an extensive record of all transactions. Centralized
agriculture information are disrupted, the reliability of authority and services, as well as trusted third parties (TTP),
the system endures, which can have fatal repercussions are susceptible to disruption, compromise, and hacking [149].
[145]. A denial-of-service (DoS) attack is a kind of Even if they are trustworthy right now, they might act
cyberattack that attempts to destroy data. inappropriately or have corrupt tendencies in the future. Each
• Usurpation: Unauthorized entrance to particular sec- transaction in the shared public ledger of the blockchain is
tions of the system via attacks such as sinkhole, replay, confirmed by an overall consensus of mining nodes that play
and code injection threatens the integrity of agricultural a role in verifying and validating transactions. Miners on the
IoT equipment [146]. bitcoin network [150] verify the block by computing a hash
with leading zeros to reach the desired level of difficulty.
6) CYBER ATTACKS Block data is immutable, which means that it can never be
Considering the increasing popularity of next-generation deleted or changed once transactions have been authenticated
smart agriculture, attackers may attack equipment that and approved by consensus. Blockchain networks can be built
interconnects both physical and digital environments to in two different ways: (1) permission-ed (or private) networks
access confidential data online. IoT-connected devices offer that are only accessible to a select number of users, or (2)

61006 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

permission-less (or public) networks that allow anybody to and IPv6 addresses. In addition, blockchain provides
join. Block chains with permissions offer greater privacy and 4.3 billion more addresses than IPv6, resulting in a better
improved access control. option for the Internet of Things, especially for devices
A generic blockchain architecture framework is shown in with restricted resources that cannot support an IPv6
the Fig.11. The architectural framework is mostly made up of stack [170].
the header of the block and the block body, which comprises a • Data Integrity and Authentication: Data commu-
collection of transactions. One feature in the block header that nicated by IoT devices connected to the blockchain
can be utilized for tracking software and protocol updates is network will always be cryptographically proofed and
the version code. Time stamps, block sizes, and transaction signed by the authentic sender, who has a unique public
counts are also included in the header. The hash value of key and GUID, ensuring the security and authenticity of
the most recent block appears in the Merkle root field. For the data that is transmitted. The blockchain distributed
effective data verification, Merkle tree hashing is frequently ledger also records all transactions performed on or by
employed in distributed systems and P2P networks. The first an IoT device and enables their secure tracking [171].
counter value generates the hash with leading zeros, and • Authorization, Authentication, and Privacy:
the nonce field is utilized for the proof-of-work procedure. Blockchain smart contracts provide decentralized
In order to maintain a block time of around 10 minutes authentication logic and rules for IoT device single-party
for Bitcoin [151] and 17.5 seconds for Ethereum [152], the and multi-party authentication. Additionally, blockchain
difficulty target, which defines the number of leading zeros, offers a more efficient way of authorizing access
is employed. The level of difficulty is modified frequently rules for connected IoT devices than more traditional
and improves (with additional leading zeros) as computer authorization protocols like OAuth 2.0, Role-Based
processing power advances with time. The block time is Access Management (RBAC), OpenID, LWM2M, and
predetermined by design to take into consideration how long OMA DM, which are frequently used in IoT device
it takes for blocks to propagate to all miners and for each authorization, authentication, and management. Smart
to agree on a block.Bitcoin is one of the first programs contracts also guarantee data privacy by defining access
created using blockchain technology, which has subsequently guidelines, constraints, and time limits, enabling certain
evolved into the basis for several modern cryptocurrencies. people, organizations, and even devices to own, control,
With the unveiling of smart contracts in July 2015, Ethereum or access data both while it is in use and while it is in
broadened the possibilities for blockchain technology. These transfer [172]. Additionally, these smart contracts can
programmable agreements, first proposed by Nick Szabo in specify who has the power to reset IoT devices, create
1994, comprise self-executing contracts that can be written in new key pairs, initiate updates, upgrade IoT software or
Solidity, a programming language identical to JavaScript, and hardware, and patch IoT software or hardware.
carried out on the blockchain of Ethereum using Ethereum • Secure Communications: Conventional IoT network-
Virtual Machines (EVM) [153]. In addition to having its ing and communication protocols, including HTTP,
own cryptocurrency, Ether, Ethereum also uses a blockchain CoAP, MQTT,RPL, and 6LoWPAN, have security
state to facilitate the execution of smart contracts. Accounts, features that are integrated and demand the use of
addresses, codes, and electronic balances are all attributes of additional, complicated protocols like DTLS, TLS,
smart contracts. and IPSec. These resource-intensive security tech-
Ethereum’s EVM storage can be expensive; however, for niques depend on centralized authority over keys.
large-scale storage requirements, decentralized data stores While omitting centralized authentication processes
like IPFS, BitTorrent, or Swarm may be employed for main- and streamlining security strategies, blockchain offers
taining pertinent data hashes. The management, control, and distinct GUIDs and pairs of keys for IoT devices. This
security of IoT-based agriculture devices are all anticipated makes it possible to apply security methods that are less
to benefit significantly from blockchain technology that is intrusive and better suited for IoT devices [173].
based on smart contracts. The core features of blockchain
technology that are listed below can be very advantageous • Administration and Identification of Things: The
for IoT primarily and IoT security specifically. implementation of identity and access management
• Address Space: Blockchain uses a 160-bit address (IAM) for IoT offers a variety of technical issues
space, as compared to IPv6, which uses a 128-bit address that require effective, secure solutions. The fluctuating
space [169]. A public key obtained using the ECDSA ownership and identification relationships between IoT
(Elliptic Curve Digital Signature Algorithm) is hashed devices are a major concern. Ownership of the device
into a 20-byte (160-bit) address on the blockchain. With changes over its lifespan, from the company that
a very low probability of address collision and a 160-bit makes it to the vendor, reseller, and ultimately the
address space, blockchain can generate addresses for end user [174]. End-user ownership may alter due
roughly 1.46 ∗ 1048 IoT devices, ensuring secure to selling, retirement, or negotiation. Maintaining the
and globally unique identifiers without the need for characteristics and interconnections of IoT devices adds
centralized oversight like IANA’s management of IPv4 a further layer of complexity. These characteristics

VOLUME 12, 2024 61007

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

TABLE 9. Security challenges and solutions in IoT-based agriculture.

61008 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

TABLE 9. (Continued.) Security challenges and solutions in IoT-based agriculture.

FIGURE 10. Cyber attacks in IoT based agriculture system.

Blockchain technology provides a strong, secure, and

cost-effective solution to these problems. It is frequently
used to establish reliable and authorized identification
enrollment, keep track of possessions, and retrieve valu-
ables. Blockchain-based technologies like TrustChain
have been put out to enable trustworthy transactions
while maintaining transaction security inside a decen-
tralized system [175]. The capabilities of blockchain
are also advantageous for IoT devices. Blockchain
technology can be utilized to authorize and give
FIGURE 11. Chained blockchain architectural framework with header and identification to interconnected IoT devices, featuring
body fields. a wide range of qualities and complex interconnections
that may be securely recorded and maintained on the
include information on the maker, make, classification, decentralized ledger of the blockchain.
model number, location, and location GPS coordinates.
In addition, IoT devices have complex linkages that
range from links between devices and people to connec- VI. THE STATE-OF-THE-ART ML MODELS AND
tions between devices and services. These connections TECHNIQUES IN SMART AGRICULTURE APPLICATIONS
involve a variety of activities, including installation, use, The world is presently dealing with an unprecedented
logistics, distributing, upgrading, repairing, and more. converge of problems related to food items and agricultural

VOLUME 12, 2024 61009

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

sustainability. Given the continual rise in global population and weather stations, and turn this raw data into valuable
and the introduction of unknown factors due to climate insights.
change into conventional farming methods, the demand for CNNs are a specific kind of deep learning model that is
innovative and effective agricultural solutions has never been particularly effective for processing grid-like data, such as
greater. The fusion of cutting-edge technology, notably the picture data. They are made to automatically and dynamically
IoT and deep learning, has ushered in a new era of agriculture learn spatial relationships of features from massive datasets.
named as ‘‘smart agriculture’’. They are, therefore, beneficial for image analysis jobs, which
It is a difficult challenge to turn raw data into knowledge are frequent in smart agriculture.
that can help farmers and agricultural professionals in real- For instance, CNNs may be applied to crop image analysis
time decision-making. The development of the IoT has to quickly identify diseases. This enables farmers to take
completely changed how data is collected in the agricultural immediate measures to stop the disease’s spread and reduce
sector. A plethora of data on soil moisture levels, temperature, crop loss [181]. Similarly, CNNs can distinguish between
weather patterns, and crop health is continually gathered by crops and weeds based on photos collected in the field. This
IoT sensors implanted in soil, rainfall stations, drones, and can aid in creating efficient weeding systems that focus on
cutting-edge agricultural equipment [176]. This constant flow weeds while ignoring crops [182]. These models may also
of data has the potential to reveal invaluable information be used to count the quantity of fruits based on photos
about how the agricultural environment functions. captured in fields. This can offer an accurately calculated
Smart agriculture, which represents a paradigm shift from yield estimate, which is essential for supply chain and
conventional agricultural practices, is built around precision logistics planning [183]. The categorization of land cover
agriculture. The key to optimising resource use, improving based on satellite photos may also be done using CNNs [184].
agricultural productivity, and reducing environmental impact Various agricultural issues, from disease diagnosis to
is automation and data-driven decision-making. For analysis plant recognition, may be addressed using CNN’s ability to
and decision-making, deep learning models are utilised in precisely recognize and localize objects in pictures. However,
smart agriculture, and the Internet of Things is used for To fully realize the promise of CNNs in smart agriculture,
real-time data collection. Deep learning techniques, for issues with data collecting and computing resources must be
instance, may be used to estimate productivity, manage resolved.
disease and insects, and perform accurate irrigation [177].
The importance of deep learning models in smart agri-
culture will only increase as we continue to tackle major B. REGION-BASED CONVOLUTIONAL
issues like climate change and food security [178]. Deep NEURAL NETWORKS
learning models provide incomparable levels of accuracy and In smart agriculture, region-based convolutional neural
efficiency, revolutionising the way we address agricultural networks (RCNNs) have been used to locate and identify
processes. CNNs,RCNN, ResNet-18 and many more are the leaf diseases. These models can pinpoint the precise areas
few examples of deep learning models that have shown to be in images where a disease is present, which is important
particularly effective in tasks like fruit counting, weed/crop in determining the severity of the condition and selecting
discrimination, and land cover classification. These models the best course of treatment. By utilizing RCNNs in smart
can analyse enormous volumes of data gathered from agriculture, manually monitoring big fields is no longer
several sources, such as IoT devices, drones, and weather necessary. This technology also makes it possible to identify
stations, and turn this unprocessed data into insights that illnesses early, reducing the degradation of plant growth.
can be put to use. As a result, crop yields have increased, A typical RCNN starts by producing a list of region
resource waste has decreased, and agricultural sustainability proposals, which could include an object of interest. These
has increased [179]. This can offer useful information areas are suggestions based on the features that CNN has
for managing fields and planning crops.As part of our learned. The model then assigns each region suggestion to
examination of the many dimensions of smart agriculture, a backdrop or one of the object classes [185]. In the case of
we pinpointed particular uses for DL models. The Table 10 smart agriculture, the backdrop may be healthy plant tissue,
provides a concise summary of our findings while the object classes could be various kinds of plant
diseases. The flexibility of RCNNs to accommodate different
object sizes is one of its main features.
A. CONVOLUTIONAL NEURAL NETWORKS This is significant in smart agriculture, where the extent
Convolutional Neural Networks (CNNs) have become an of the infected region might fluctuate dramatically. RCNNs
effective tool for smart agriculture, offering answers to a do this by employing area suggestions with various aspect
range of problems. CNNs are proficient at identifying and ratios and sizes, which enables them to recognize objects of
classifying plant diseases, differentiating weeds and crops, various sizes and forms. Plant disease detection is one of
counting fruits, and classifying land cover [180]. the main areas where RCNNs are used in smart agriculture.
These models can analyze enormous volumes of data Early disease detection is essential for successful disease
gathered from several sources, such as IoT devices, drones, management since plant diseases can have a significant

61010 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

TABLE 10. Deep learning model’s application in smart agriculture.

VOLUME 12, 2024 61011

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

TABLE 10. (Continued.) Deep learning model’s application in smart agriculture.

impact on crop production and quality. Farmers now have a focused application of herbicides, may be guided by precise
powerful tool for early disease diagnosis thanks to RCNNs, identification and differentiation of crops and weeds. RCNNs
which can be trained to recognize disease signs from photos may be applied to field image analysis to locate and identify
of plant leaves or stems [186].Identification of crops and crops and weeds precisely [187]. Another research [203]
weeds is another crucial field for RCNN applications in proposed ML and DL methods for weed detection and
smart agriculture. Precision farming techniques, such as the classification in crops, which achieved 90% accuracy with

61012 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

Random Forest and ConvNeXt on the CottonWeedID15 pest identification, and better environmental condition mon-
dataset and 99.5% accuracy with SVM on the early crop weed itoring. Issues with data collecting and computing resources
dataset. must be resolved to fully grasp the potential of LSTMs.
However, RCNNs demand training data and are computa-
tionally expensive. This can be difficult in smart agriculture D. RESIDUAL NETWORKS
since data collecting can be costly and time-consuming. A type of CNN called residual networks (ResNets) invented
Despite these difficulties, RCNNs have demonstrated excel- the idea of ‘‘skip connections,’’ which enables the gradient to
lent potential in smart agriculture. In the upcoming years, be directly back propagated to older layers. The fundamental
it is anticipated that RCNNs will play a more significant part advantage of a ResNet is its capacity to train 100+ layer
in smart agriculture due to continued developments in deep networks with impressive accuracy over a wide range
learning and the expansion of high-quality agricultural data. of datasets (such as ImageNet). ResNets are particularly
They can assist farmers in taking prompt action to reduce crop helpful in addressing challenging agricultural tasks that need
loss and maintain food security by making it possible to detect high-level feature extraction and abstraction because they can
illnesses early and accurately. Techniques like Fast R-CNN efficiently train deep networks.
and Faster RCNN have enhanced the original RCNN. The A single residual block in a ResNet comprises a ReLU
importance of RCNNs in smart agriculture will only increase activation function, batch normalization, and numerous
as we continue to face major issues like climate change and layers of convolutions. ResNets apply to smart agriculture in
food security. a variety of contexts. Precision agriculture can also benefit
from the usage of ResNets. Utilizing cutting-edge technology
C. LONG SHORT-TERM MEMORY
to handle agricultural tasks more precisely and effectively
Long Short-Term Memory (LSTM) networks, a recurrent is known as precision agriculture. This involves accurate
neural network, have shown promise in smart agriculture. fertilizer and pesticide application, variable rate irrigation,
Because LSTMs can characterize temporal sequences and and the best planting and harvesting techniques. To generate
their long-range interactions, they are particularly well insights that help direct precision agricultural practices,
suited for tasks requiring time-series data, typically seen in ResNets can analyze large amounts of data on weather, soil
agricultural applications. LSTM networks may be used To quality, crop trends, and satellite images.
make wise judgments about irrigation, fertilization, and pest Plant disease and pest detection is one of the most popular
management, for example, to anticipate the environmental uses of this technology [192]. Plant diseases and pests
conditions of plants based on past data. Additionally, because significantly influence the productivity and quality of crops.
LSTM networks can process lengthy data sequences, they By prompting response and minimizing crop loss, early
are appropriate for examining the constant stream of data and precise identification of various diseases and pests can
produced by IoT sensors in smart agriculture [188] increase production. With its deep architectures and skip
Environmental condition monitoring is one of the main connections, ResNets can accurately identify numerous plant
uses of LSTMs in smart agriculture. Data on the envi- diseases and pests by extracting detailed information from
ronmental circumstances of plants are continually tracked photos of plant leaves and stems.
in a smart agriculture system. This time-series data may Predicting crop yield is another area in which ResNets
be analyzed using LSTMs to forecast future environmental is used in smart agriculture. An essential component of
conditions [189]. For instance, they may be used to forecast agricultural planning and management is crop production
temperature, humidity, and other environmental variables, prediction [193]. For precise yield estimates, ResNets may be
which are essential for plant development. used to analyze various data sources, such as meteorological
In smart agriculture, the domain of pest detection is a data, soil quality data, and crop patterns.
crucial application of LSTMs combined with CNNs [190]. ResNets have a lot of potential for the subject of smart
The frequency of pests can change over time and is agriculture. Their capacity to simulate highly complex
frequently influenced by environmental factors. LSTMs may patterns and correlations in data can result in more accurate
be utilized to analyze environmental and historical data on production predictions, better disease and pest detection, and
pest occurrences to forecast future pest occurrences. This effective agricultural management techniques.
can make it possible for farmers to minimize crop loss and
take preventative measures. Crop yield prediction may also
VII. RESEARCH OPEN CHALLENGES AND
be done with LSTMs. Environmental conditions, agricultural
FUTURE RESEARCH DIRECTIONS
methods, and crop variety are only a few variables that might
affect crop yield. When combined with other relevant data, There are a number of open research problems and potential
LSTMs may be used to analyze past yield data to produce areas for further investigation in the field of IoT-based
precise predictions [191]. Farmers may be able to organize agricultural applications; some of these open challenges and
their marketing and harvesting operations better, increasing potential areas are mentioned in this section.
their profitability. Their capacity to create complex temporal 1) Implementation and maintenance: A notable chal-
correlations in data can help with yield prediction, precise lenge can be caused by the deployment and continuous

VOLUME 12, 2024 61013

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

maintenance of IoT-based agricultural systems span- are essential for addressing the aforementioned chal-
ning significant agricultural regions. To make sure lenges in order to deliver highly effective smart
that everything runs smoothly and effectively, this IoT-based agriculture solutions.
challenge requires tackling a variety of technological, 5) Big data: The efficient processing of an extensive
logistical, and resource-related challenges. To fully amount of agricultural data is critical in IoT-based
utilize IoT in agriculture, solutions to these challenges agriculture systems. This data is collected by IoT
must be identified that are realistic. sensors placed in the agricultural environment, where
2) Cost: Numerous monetary factors, including pre- scenarios and factors fluctuate constantly. Furthermore,
liminary setup costs and continuing operating costs, as the demand for accurate and reliable assessment
must be taken into account when deploying IoT in of crop conditions advances, sensors and connection
agriculture. Hardware expenses such as IoT devices, points produce an extensive amount of agricultural
sensors, base station infrastructure, and gateways are data in a variety of formats [122]. Processing various
all included in the setup costs. On the other hand, and large data quantities can be difficult, potentially
continuing payments for subscriptions for services like impeding timely access, even though continuous access
centralized data collection, data interchange, and IoT is critical. Any disruptions in accessing data might
device management are included in the running costs. result in serious effects in major instances, including
3) Security: There is presently no unified benchmark to those impacting vital crops or livestock. Substantial
determine the security resilience of suggested solutions drifts in the capacity of data are expected to decline
in the environment of IoT-based agriculture systems. as algorithms for prediction become more capable of
In order to examine the level of security in their predicting such occurrences. Organizations are likely
individual research initiatives, several research groups to participate in both inside and outside education to
separately undertake security evaluations, frequently train young scientists with the ability to overcome
utilizing adversarial analyses. Because of varying these difficulties, according to the Global Institute for
underlying assumptions and guiding concepts, these Analytics. Businesses will prioritize data updates and
evaluations frequently lack uniformity. The devel- shift their focus from managing system procedures to
opment of a system for estimating the level of acquiring computations. Furthermore, more firms are
security in research activities has become crucial for likely to investigate the possibility of exploiting their
the IoT-based agriculture security community [128]. agricultural data resources. There will certainly be sig-
For example, a web-based application that delivers nificant growth for well-known services and suppliers
findings according to user input may be implemented like Kaggle, Algorithmia, and DataXu, leading to the
to establish an agricultural IoT security evaluation emergence of booming algorithm economies in the
system. However, it’s crucial to highlight that neither agriculture sector.
the security analysis of recent agricultural studies nor 6) Artificial intelligence: Network security is one of
the use of cryptography techniques in cryptanalysis are the areas where machine learning and deep learning
included in this work. To accurately gauge the security have become research hotspots. For instance, machine
level of IoT-based agricultural systems, more research learning-based networks, especially detection of intru-
must be conducted. sion techniques, are increasing in prevalence and
4) Advancement of communication & network tech- have conceivable applications in IoT-based agriculture
nologies:Considering the advantages provided by systems. In the area of agriculture IoT, deep learning
developing cutting-edge communication technologies, networks, as discussed in the previous section, have
potential consumers have agreed to endorse smart been used to evaluate the provision of data, such as
agriculture systems [44]. The requirement for rapid and crop disease identification, throughout multiple layers
efficient network connections in order to ensure system of IoT systems for quick identification of potential
coordination has been demonstrated by contemporary problems [187]. Considering the growing popularity
advances in technology that have led to immense of deep learning techniques on agricultural servers for
advances in the processing of data. The advancement tasks such as identifying diseases, it is essential to
of 6G networks has been driven by this demand. investigate their effectiveness in guaranteeing system
It is projected that 6G and IoT technologies will privacy and the security of data.
combine to establish a panorama of smart agriculture 7) Blockchain: The primary focus of blockchain
networks in the future, improving device integration, researchers is the optimization of blockchain-based
increasing network efficiency, and addressing security solutions for IoT devices with limited resources.
issues. Key 6G technology advancements, such as These advancements aim to ensure data integrity
Software-Defined Networking (SDN), Massive Multi- and transparency throughout the agricultural process.
ple Input, Multiple Output (MIMO), Network Function The study specifically explores the implementation of
Virtualization (NFV), Machine-to-Machine (M2M) smart contract technology to streamline agricultural
communication, and millimeter-wave communication, transactions, address interoperability issues, increase

61014 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

data privacy and ownership rights for landowners 10) Eco-sustainable Technologies: The primary focus of
and stakeholders, and develop adaptable blockchain research in eco-sustainable technologies is assessing
frameworks. These strategies are designed to capi- the mineral balance of agricultural treatments and
talize on the potential advantages of blockchain by examining any potential effects on cutting-edge IoT-
addressing the distinctive challenges offered in the based agricultural systems. Furthermore, nanotechnol-
agricultural environment, thus improving the security, ogy can be crucial in enhancing the intelligence and
efficiency, and dependability of IoT applications in adaptability of smart agricultural systems, particularly
agriculture [172]. for IoT devices. To build intelligent and environmen-
8) Agriculture robotic system: The growing importance tally conscious agricultural systems that rely on IoT
of agricultural productivity and effectiveness has drawn sensors for monitoring and improvement purposes, this
attention to the importance of robotics in agriculture, area presents an immense research gap that has to be
notably in activities such as robot crop cutting,drone- addressed.
based field monitoring, spray quadcopters, and robot-
detecting diseases, where innovations in robotics VIII. CONCLUSION
technology have transformed conventional agricultural This review article provides an extensive and systematic
practices. Fundamental, flexible, simple, and afford- analysis of IoT-based agriculture systems, spanning technical
able technological solutions must be prioritized in advancements, sector developments, device sustainability,
the forthcoming generations of agricultural automated applications, communication standards, gaps in research,
machines and systems engineering. switching from significant privacy and security concerns, and solutions that
outdated electrical control devices, sensors, and net- are distinctive to IoT in agriculture, as well as state-of-the-art
working systems to their digital versions may assist ML models and techniques in smart agriculture applications.
these innovations become more prevalent. In contrast It presents an in-depth explanation of IoT technologies and
to traditional electric motors, which can be complex, how they might be used in different sectors of agriculture.
large, and expensive, the agriculture robotics market The report evaluates key findings from previous research
requires simplified and logic-driven technology solu- and emphasizes the particular security concerns of IoT in
tions. The emerging phenomenon of the IoT has the agriculture, considering its broad spectrum of applications.
potential to effortlessly incorporate with agricultural Additionally, it emphasizes significant application fields and
technology and equipment in the future years, boosting research opportunities, offering valuable data to researchers
the efficiency and precision of numerous fields of and industry experts. The importance of IoT security,
agriculture. including technological developments like blockchain for
9) Unmanned Aerial Vehicles (UAVs) and AI Tech- risk mitigation, is also emphasized in the article.In order
nologies in Agriculture: With the expansion of IoT to improve the effectiveness and accuracy of automation,
and the advent of communication technologies, the use modeling, and forecasting systems in agriculture, the debate
of drones in agriculture has accelerated significantly. predicts the rise of hybrid technologies that combine big data
In the future, AI offers a viable way to improve the analytics, data mining, artificial intelligence, and the Internet
potential and effectiveness of UAVs in agricultural of Things.
operations. AI has the potential to transform drone
capabilities by empowering them to perform a wide DECLARATIONS
range of essential functions that enhance agricultural DATA AVAILABILITY
practices. A few examples are soil analysis, automated N/A
planting, targeted crop spraying, real-time crop health
monitoring, precision irrigation, and extensive crop
CONFLICT OF INTEREST
inspection. Furthermore, drones equipped with a vari-
ety of sensors, including 3D cameras, thermal imaging, It is declared by the authors that there is no conflict of interest.
multi-spectral imaging, and optical cameras, make
it easier to monitor various crop conditions, disease CODE AVAILABILITY
outbreaks, vegetation density, and soil parameters. Code can be provided on request.
AI has the potential to analyze the data collected from
these sensors and provide valuable insights such as the AUTHORS CONTRIBUTIONS
need for pesticides, fertilizer distribution, canopy cover Tanzeela Shakeel and Muhammad Shmoon worked on the
mapping, yield forecasts, plant counting, and plant literature review and wrote the initial draft of the article.
height measurement. It demonstrates the various uses Asma Naseer floated the idea, supervised the systematic lit-
of UAVs in agriculture, including mapping, spraying, erature review, and contributed to the article write-up. Awais
harvesting, and sensing, and opens up opportunities Ahmad verified and analyzed the findings and contributed to
for potential advances in this field with the use of AI the article write-up. Shafiq Ur Rehman improved the article
technology. write-up. Volker Gruhn reviewed and supervised the project.

VOLUME 12, 2024 61015

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

ACKNOWLEDGMENT [20] O. Friha, M. A. Ferrag, L. Shu, L. Maglaras, and X. Wang, ‘‘Internet

This work was supported by the Deanship of Scientific of Things for the future of smart agriculture: A comprehensive survey
of emerging technologies,’’ IEEE/CAA J. Autom. Sinica, vol. 8, no. 4,
Research at Imam Mohammad Ibn Saud Islamic University pp. 718–752, Apr. 2021.
(IMSIU) under Grant IMSIU-RP23042. [21] C. Maraveas and T. Bartzanas, ‘‘Application of Internet of Things (IoT)
for optimized greenhouse environments,’’ AgriEngineering, vol. 3, no. 4,
pp. 954–970, Nov. 2021.
REFERENCES [22] X. Zhang, J. Zhang, L. Li, Y. Zhang, and G. Yang, ‘‘Monitoring citrus
[1] N. Ahmed, D. De, and I. Hussain, ‘‘Internet of Things (IoT) for smart soil moisture and nutrients using an IoT based system,’’ Sensors, vol. 17,
precision agriculture and farming in rural areas,’’ IEEE Internet Things no. 3, p. 447, Feb. 2017.
J., vol. 5, no. 6, pp. 4890–4899, Dec. 2018. [23] S. Ratnaparkhi, S. Khan, C. Arya, S. Khapre, P. Singh, M. Diwakar, and
[2] A. H. Bagdadee, L. Zhang, and M. S. H. Remus, ‘‘A brief review of the A. Shankar, ‘‘Smart agriculture sensors in IoT: A review,’’ Mater. Today,
iot-based energy management system in the smart industry,’’ in Artificial Proc., vol. 2, pp. 1–12, 2020, doi: 10.1016/j.matpr.2020.11.138.
Intelligence and Evolutionary Computations in Engineering Systems. [24] R. Akhter and S. A. Sofi, ‘‘Precision agriculture using IoT data analytics
Singapore: Springer, 2020, pp. 443–459. and machine learning,’’ J. King Saud Univ. - Comput. Inf. Sci., vol. 34,
[3] H. Nagamma, ‘‘IoT based smart security gadget for women’s safety,’’ in no. 8, pp. 5602–5618, Sep. 2022.
Proc. 1st Int. Conf. Adv. Inf. Technol. (ICAIT), Jul. 2019, pp. 348–352. [25] W. Tao, L. Zhao, G. Wang, and R. Liang, ‘‘Review of the Internet of
[4] R. Kaur, P. Vats, M. Mandot, S. S. Biswas, and R. Garg, ‘‘Literature Things communication technologies in smart agriculture and challenges,’’
survey for IoT-based smart home automation: A comparative analysis,’’ Comput. Electron. Agricult., vol. 189, Oct. 2021, Art. no. 106352.
in Proc. 9th Int. Conf. Rel., INFOCOM Technol. Optim. (Trends Future [26] T.-C. Hsu, H. Yang, Y.-C. Chung, and C.-H. Hsu, ‘‘A creative IoT
Directions) (ICRITO), Sep. 2021, pp. 1–6. agriculture platform for cloud fog computing,’’ Sustain. Comput.,
[5] Z. Chen, C. B. Sivaparthipan, and B. Muthu, ‘‘IoT based smart and Informat. Syst., vol. 28, Dec. 2020, Art. no. 100285.
intelligent smart city energy optimization,’’ Sustain. Energy Technol. [27] X. Yang, L. Shu, J. Chen, M. A. Ferrag, J. Wu, E. Nurellari, and K. Huang,
Assessments, vol. 49, Feb. 2022, Art. no. 101724. ‘‘A survey on smart agriculture: Development modes, technologies, and
[6] M. N. Mohammed, S. F. Desyansah, S. Al-Zubaidi, and E. Yusuf, security and privacy challenges,’’ IEEE/CAA J. Autom. Sinica, vol. 8,
‘‘An Internet of Things-based smart homes and healthcare monitoring no. 2, pp. 273–302, Feb. 2021.
and management system: Review,’’ J. Phys., Conf. Ser., vol. 1450, no. 1, [28] S. Keele et al., ‘‘Guidelines for performing systematic literature reviews
Feb. 2020, Art. no. 012079. in software engineering,’’ Keele Univ., Durham Univ., Joint Rep., Tech.
[7] E. F. I. Raj and M. Appadurai, ‘‘Internet of Things-based smart Rep. EBSE 2007-001, 2007.
transportation system for smart cities,’’ in Advanced Technologies
[29] T. Shakeel, S. Gul, S. Habib, and A. Naseer, ‘‘A systematic literature
and Societal Change. Cham, Switzerland: Springer, 2022,
review on cognitive radio networks,’’ in Proc. Int. Conf. Innov. Comput.
pp. 39–50.
(ICIC), Nov. 2021, pp. 1–11.
[8] D. A. Sungheetha and D. R. Sharma R, ‘‘Real time monitor-
[30] T. Shakeel, S. Habib, W. Boulila, A. Koubaa, A. R. Javed, M. Rizwan,
ing and fire detection using Internet of Things and cloud based
T. R. Gadekallu, and M. Sufiyan, ‘‘A survey on COVID-19 impact in
drones,’’ J. Soft Comput. Paradigm, vol. 2, no. 3, pp. 168–174,
the healthcare domain: Worldwide market implementation, applications,
Jul. 2020.
security and privacy issues, challenges and future prospects,’’ Complex
[9] J. Xu, B. Gu, and G. Tian, ‘‘Review of agricultural IoT technology,’’ Artif.
Intell. Syst., vol. 9, no. 1, pp. 1027–1058, Feb. 2023.
Intell. Agricult., vol. 6, pp. 10–22, Jan. 2022.
[10] (2022). World Economic Situation and Prospects As of Mid- [31] U. Shafi, R. Mumtaz, J. García-Nieto, S. A. Hassan, S. A. R. Zaidi,
2022. Accessed: Jun. 14, 2023. [Online]. Available: www.un.org/ and N. Iqbal, ‘‘Precision agriculture techniques and practices: From
development/desa/dpad/publication/world- economic-situation-and- considerations to applications,’’ Sensors, vol. 19, no. 17, p. 3796,
prospects-as-of-mid-2022/ Sep. 2019.
[11] (Sep. 2019). How to Feed the World in 2050 By Fao. [32] S. Premkumar and A. Sigappi, ‘‘IoT-enabled edge computing model for
Accessed: Jun. 14, 2023. [Online]. Available: https://www. smart irrigation system,’’ J. Intell. Syst., vol. 31, no. 1, pp. 632–650,
fao.org/wsfs/forum2050/wsfs-forum/en/ May 2022.
[12] A. D. Tripathi, R. Mishra, K. K. Maurya, R. B. Singh, and [33] X. Bai, Y. Huang, W. Ren, M. Coyne, P. Jacinthe, B. Tao, D. Hui, J. Yang,
D. W. Wilson, ‘‘Estimates for world population and global food and C. Matocha, ‘‘Responses of soil carbon sequestration to climate-smart
availability for global health,’’ in The Role of Functional Food agriculture practices: A meta-analysis,’’ Global Change Biol., vol. 25,
Security in Global Health. Amsterdam, The Netherlands: Elsevier, 2019, no. 8, pp. 2591–2606, Aug. 2019.
pp. 3–24. [34] C. Marwa, S. B. Othman, and H. Sakli, ‘‘IoT based low-cost weather
[13] M. S. Farooq, S. Riaz, A. Abid, K. Abid, and M. A. Naeem, ‘‘A survey on station and monitoring system for smart agriculture,’’ in Proc. 20th
the role of IoT in agriculture for the implementation of smart farming,’’ Int. Conf. Sci. Techn. Autom. Control Comput. Eng. (STA), Dec. 2020,
IEEE Access, vol. 7, pp. 156237–156271, 2019. pp. 349–354.
[14] M. Raj, S. Gupta, V. Chamola, A. Elhence, T. Garg, M. Atiquzzaman, and [35] H. Jariwala, R. M. Santos, J. D. Lauzon, A. Dutta, and Y. Wai Chiang,
D. Niyato, ‘‘A survey on the role of Internet of Things for adopting and ‘‘Controlled release fertilizers (CRFs) for climate-smart agriculture
promoting agriculture 4.0,’’ J. Netw. Comput. Appl., vol. 187, Aug. 2021, practices: A comprehensive review on release mechanism, materials,
Art. no. 103107. methods of preparation, and effect on environmental parameters,’’
[15] (Sep. 2019). Global Smart Agriculture Marke. Accessed: Jun, 14, 2023. Environ. Sci. Pollut. Res., vol. 29, no. 36, pp. 53967–53995, Aug. 2022.
[Online]. Available: https://www.zionmarketresearch.com/report/smart- [36] S. Nižetić, N. Djilali, A. Papadopoulos, and J. J. P. C. Rodrigues,
agriculture-market ‘‘Smart technologies for promotion of energy efficiency, utilization of
[16] L. García, L. Parra, J. M. Jimenez, J. Lloret, and P. Lorenz, ‘‘IoT-based sustainable resources and waste management,’’ J. Cleaner Prod., vol. 231,
smart irrigation systems: An overview on the recent trends on sensors pp. 565–591, Sep. 2019.
and IoT systems for irrigation in precision agriculture,’’ Sensors, vol. 20, [37] A. D. Boursianis, M. S. Papadopoulou, P. Diamantoulakis,
no. 4, p. 1042, Feb. 2020. A. Liopa-Tsakalidi, P. Barouchas, G. Salahas, G. Karagiannidis,
[17] W.-L. Chen, Y.-B. Lin, Y.-W. Lin, R. Chen, J.-K. Liao, F.-L. Ng, S. Wan, and S. K. Goudos, ‘‘Internet of Things (IoT) and agricultural
Y.-Y. Chan, Y.-C. Liu, C.-C. Wang, C.-H. Chiu, and T.-H. Yen, ‘‘AgriTalk: unmanned aerial vehicles (UAVs) in smart farming: A comprehensive
IoT for precision soil farming of turmeric cultivation,’’ IEEE Internet review,’’ Internet Things, vol. 18, May 2022, Art. no. 100187.
Things J., vol. 6, no. 3, pp. 5209–5223, Jun. 2019. [38] V. Pallagani, V. Khandelwal, B. Chandra, V. Udutalapally, D. Das, and
[18] J. Mendez and D. Dasig, ‘‘Frost prediction in highland crops management S. P. Mohanty, ‘‘DCrop: A deep-learning based framework for accurate
using IoT-enabled system and multiple regression,’’ in Internet of Things prediction of diseases of crops in smart agriculture,’’ in Proc. IEEE
and Analytics for Agriculture, vol. 2. Cham, Switzerland: Springer, 2020, Int. Symp. Smart Electron. Syst. (iSES) (Formerly iNiS), Dec. 2019,
pp. 261–288. pp. 29–33.
[19] S. V. Gaikwad, A. D. Vibhute, K. V. Kale, and S. C. Mehrotra, ‘‘An [39] I. Bhakta, S. Phadikar, and K. Majumder, ‘‘State-of-the-art technologies
innovative IoT based system for precision farming,’’ Comput. Electron. in precision agriculture: A systematic review,’’ J. Sci. Food Agricult.,
Agricult., vol. 187, Aug. 2021, Art. no. 106291. vol. 99, no. 11, pp. 4878–4888, Aug. 2019.

61016 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

[40] R. K. Singh, R. Berkvens, and M. Weyn, ‘‘AgriFusion: An architecture for [60] S.-Y. Wang, J.-E. Chang, H. Fan, and Y.-H. Sun, ‘‘Comparing the
IoT and emerging technologies based on a precision agriculture survey,’’ performance of NB-IoT, LTE cat-m1, Sigfox, and LoRa for IoT end
IEEE Access, vol. 9, pp. 136253–136283, 2021. devices moving at high speeds in the air,’’ J. Signal Process. Syst., vol. 94,
[41] H. A. Alharbi and M. Aldossary, ‘‘Energy-efficient edge-fog-cloud no. 1, pp. 81–99, Jan. 2022.
architecture for IoT-based smart agriculture environment,’’ IEEE Access, [61] R. Elashmawy and I. Uysal, ‘‘Precision agriculture using soil sensor
vol. 9, pp. 110480–110492, 2021. driven machine learning for smart strawberry production,’’ Sensors,
[42] M. A. Ferrag, L. Shu, X. Yang, A. Derhab, and L. Maglaras, ‘‘Secu- vol. 23, no. 4, p. 2247, Feb. 2023.
rity and privacy for green IoT-based agriculture: Review, blockchain [62] M. Hosseinzadeh, J. Tanveer, A. M. Rahmani, E. Yousefpoor,
solutions, and challenges,’’ IEEE Access, vol. 8, pp. 32031–32053, M. S. Yousefpoor, F. Khan, and A. Haider, ‘‘A Cluster-Tree-Based
2020. secure routing protocol using dragonfly algorithm (DA) in the Internet of
[43] A. A. Araby, M. M. A. Elhameed, N. M. Magdy, L. A. Said, N. Abdelaal, Things (IoT) for smart agriculture,’’ Mathematics, vol. 11, no. 1, p. 80,
Y. T. Abd Allah, M. S. Darweesh, M. A. Fahim, and H. Mostafa, ‘‘Smart Dec. 2022.
IoT monitoring system for agriculture with predictive analysis,’’ in Proc. [63] D.-M. Chang, T.-C. Hsu, C.-T. Yang, and J. Yang, ‘‘A data factor study for
8th Int. Conf. Modern Circuits Syst. Technol. (MOCAST), May 2019, machine learning on heterogenous edge computing,’’ Appl. Sci., vol. 13,
pp. 1–4. no. 6, p. 3405, Mar. 2023.
[44] M. Pachayappan, C. Ganeshkumar, and N. Sugundan, ‘‘Technological [64] F. P. Correia, S. R. D. Silva, F. B. S. D. Carvalho, M. S. D. Alencar,
implication and its impact in agricultural sector: An IoT based K. D. R. Assis, and R. M. Bacurau, ‘‘LoRaWAN gateway placement in
collaboration framework,’’ Proc. Comput. Sci., vol. 171, pp. 1166–1173, smart agriculture: An analysis of clustering algorithms and performance
Jan. 2020. metrics,’’ Energies, vol. 16, no. 5, p. 2356, Mar. 2023.
[45] A. U. Mentsiev, A. U. Mentsiev, and E. F. Amirova, ‘‘IoT and [65] R. Sanchez-Iborra, A. Zoubir, A. Hamdouchi, A. Idri, and A. Skarmeta,
mechanization in agriculture: Problems, solutions, and prospects,’’ ‘‘Intelligent and efficient IoT through the cooperation of TinyML and
IOP Conf. Ser., Earth Environ. Sci., vol. 548, no. 3, Aug. 2020, edge computing,’’ Informatica, vol. 34, no. 1, pp. 147–168, 2023.
Art. no. 032035. [66] M. Vyas, G. Anand, R. N. Yadav, and S. K. Nayak, ‘‘DASA: An
[46] M. Abbasi, M. Plaza-Hernández, and Y. Mezquita, ‘‘Security of efficient data aggregation algorithm for LoRa enabled fog layer in smart
IoT application layer: Requirements, threats, and solutions,’’ in agriculture,’’ in Proc. Int. Conf. Adv. Inf. Netw. Appl., 2023, pp. 40–52.
Proc. Int. Symp. Ambient Intell. Cham, Switzerland: Springer, 2022, [67] Y. Chang, J. Xu, and K. Z. Ghafoor, ‘‘An IoT and blockchain approach for
pp. 86–100. the smart water management system in agriculture,’’ Scalable Comput.,
[47] S. Verma, R. Gala, S. Madhavan, S. Burkule, S. Chauhan, and C. Prakash, Pract. Exper., vol. 22, no. 2, Oct. 2021, Art. no. e12892.
‘‘An Internet of Things (IoT) architecture for smart agriculture,’’ in Proc. [68] S. K. Apat, J. Mishra, K. S. Raju, and N. Padhy, ‘‘A study on smart
4th Int. Conf. Comput. Commun. Control Autom. (ICCUBEA), Aug. 2018, agriculture using various sensors and agrobot: A case study,’’ in Smart
pp. 1–4. Intelligent Computing and Applications. Cham, Switzerland: Springer,
[48] T. Hidayat, R. Mahardiko, and F. D. Sianturi Tigor, ‘‘Method of 2022, pp. 531–540.
systematic literature review for Internet of Things in ZigBee smart [69] K. Pavelka, P. Raeva, and K. Pavelka, ‘‘Evaluating the performance
agriculture,’’ in Proc. 8th Int. Conf. Inf. Commun. Technol. (ICoICT), of airborne and ground sensors for applications in precision agricul-
Jun. 2020, pp. 1–4. ture: Enhancing the postprocessing state-of-the-art algorithm,’’ Sensors,
[49] D. K. Singh and R. Sobti, ‘‘Wireless communication technologies vol. 22, no. 19, p. 7693, Oct. 2022.
for Internet of Things and precision agriculture: A review,’’ in Proc. [70] M. Pyingkodi, K. Thenmozhi, K. Nanthini, M. Karthikeyan,
6th Int. Conf. Signal Process., Comput. Control (ISPCC), Oct. 2021, S. Palarimath, V. Erajavignesh, and G. B. A. Kumar, ‘‘Sensor based
pp. 765–769. smart agriculture with IoT technologies: A review,’’ in Proc. Int. Conf.
[50] B. Li, Y. Ling, M. Tian, and S. Zheng, ‘‘Design and imple- Comput. Commun. Informat. (ICCCI), Jan. 2022, pp. 1–7.
mentation of sugarcane growth monitoring system based on RFID [71] A. Simo, S. Dzitac, G. E. Badea, and D. Meianu, ‘‘Smart agriculture:
and ZigBee,’’ Int. J. Online Eng. (iJOE), vol. 14, no. 3, p. 96, IoT-based greenhouse monitoring system,’’ Int. J. Comput. Commun.
Mar. 2018. CONTROL, vol. 17, no. 6, pp. 1–17, Dec. 2022.
[51] C. H. Quach, M. T. Pham, T. S. Nguyen, and M. D. Phung, ‘‘Real- [72] M. Cruz, S. Mafra, E. Teixeira, and F. Figueiredo, ‘‘Smart strawberry
time agriculture field monitoring using IoT-based sensors and unmanned farming using edge computing and IoT,’’ Sensors, vol. 22, no. 15, p. 5866,
aerial vehicles,’’ in Proc. 8th NAFOSTED Conf. Inf. Comput. Sci. (NICS), Aug. 2022.
Dec. 2021, pp. 492–497. [73] C. Huang, Y. Ke, X. Hua, J. Yang, M. Sun, and W. Yang, ‘‘Application
[52] D. Khosla and K. S. Malhi, ‘‘An ultra-wideband dielectric resonator status and prospect of edge computing in smart agriculture,’’ Trans. Chin.
antenna for WSN based IoT applications in agriculture,’’ Int. J. Soc. Agric. Eng., vol. 38, pp. 224–234, Jan. 2022.
Sensors, Wireless Commun. Control, vol. 12, no. 4, pp. 281–291, [74] S. S. Baghel, P. Rawat, R. Singh, S. V. Akram, S. Pandey, and
May 2022. A. S. Baghel, ‘‘AI, IoT and cloud computing based smart agriculture,’’
[53] W.-S. Kim, W.-S. Lee, and Y.-J. Kim, ‘‘A review of the applications of in Proc. 5th Int. Conf. Contemp. Comput. Informat. (IC3I), Dec. 2022,
the Internet of Things (IoT) for agricultural automation,’’ J. Biosyst. Eng., pp. 1658–1661.
vol. 45, no. 4, pp. 385–400, Dec. 2020. [75] K. A. Kumar and S. Nandhini, ‘‘Pricing on use of cloud computing and
[54] V.-D. Gavra and O. A. Pop, ‘‘Usage of ZigBee and LoRa Internet of Things for optimized agricultural practices,’’ in Proc. Int.
wireless technologies in IoT systems,’’ in Proc. IEEE 26th Int. Mobile Embedded Technol. Conf. (MECON), Mar. 2022, pp. 395–400.
Symp. Design Technol. Electron. Packag. (SIITME), Oct. 2020, [76] M. A. Ahmed, J. L. Gallardo, M. D. Zuniga, M. A. Pedraza, G. Carvajal,
pp. 221–224. N. Jara, and R. Carvajal, ‘‘LoRa based IoT platform for remote
[55] J. Arshad, M. Aziz, A. A. Al-Huqail, M. H. U. Zaman, M. Husnain, monitoring of large-scale agriculture farms in Chile,’’ Sensors, vol. 22,
A. U. Rehman, and M. Shafiq, ‘‘Implementation of a LoRaWAN based no. 8, p. 2824, Apr. 2022.
smart agriculture decision support system for optimum crop yield,’’ [77] M. E. Amr, A. A. Al-Awamry, M. A. Elmenyawi, and A. S. T. Eldien,
Sustainability, vol. 14, no. 2, p. 827, Jan. 2022. ‘‘Design and implementation of a low-cost IoT node for data processing,
[56] R. Morais, J. Mendes, R. Silva, N. Silva, J. J. Sousa, and E. Peres, case study: Smart agriculture,’’ J. Commun., vol. 17, no. 2, pp. 99–109,
‘‘A versatile, low-power and low-cost IoT device for field data gathering 2022.
in precision agriculture practices,’’ Agriculture, vol. 11, no. 7, p. 619, [78] K. Rajesh, K. Riyaz, B. Pallavi, C. S. S. Reddy, A. Raamkumar, and
Jun. 2021. B. Baranitharan, ‘‘A smart agriculture storaging system with IoT based,’’
[57] S. Premkumar and A. Sigappi, ‘‘ASIS edge computing model to determine in Proc. Int. Conf. Intell. Controller Comput. Smart Power (ICICCSP),
the communication protocols for IoT based irrigation,’’ J. Mobile Jul. 2022, pp. 1–6.
Multimedia, vol. 18, no. 3, p. 13052, 2022. [79] R. L. Rosa, C. Dehollain, M. Costanza, A. Speciale, F. Viola, and
[58] D. K. Fisher, R. S. Fletcher, and S. S. Anapalli, ‘‘Evolving open- P. Livreri, ‘‘A battery-free wireless smart sensor platform with Bluetooth
source technologies offer options for remote sensing and monitor- low energy connectivity for smart agriculture,’’ in Proc. IEEE 21st Medit.
ing in agriculture,’’ Adv. Internet Things, vol. 10, no. 1, pp. 1–10, Electrotechnical Conf. (MELECON), Jun. 2022, pp. 554–558.
2020. [80] S. Anulekshmi and D. R. Durga, ‘‘Performance improvement of the wire-
[59] B. Shabandri, S. R. Madara, and P. Maheshwari, ‘‘IoT-based smart tree less sensor network with proficient power management with supervised
management solution for green cities,’’ in Internet of Things and Analytics multimodel data regression algorithm in precision agriculture,’’ in Proc.
for Agriculture, vol. 2. Springer, 2020, pp. 181–199. Int. Conf. Electron. Renew. Syst. (ICEARS), Mar. 2022, pp. 754–761.

VOLUME 12, 2024 61017

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

[81] A. Valente, C. Costa, L. Pereira, B. Soares, J. Lima, and S. Soares, [100] A. Ali and S. Alshmrany, ‘‘Internet of Things (IoT) embedded smart
‘‘A LoRaWAN IoT system for smart agriculture for vine water status sensors system for agriculture and farm management,’’ Int. J. Adv. Appl.
determination,’’ Agriculture, vol. 12, no. 10, p. 1695, Oct. 2022. Sci., vol. 7, no. 10, pp. 38–45, Oct. 2020.
[82] H. N. Saha, S. Chakraborty, and R. Roy, ‘‘Integration of RFID and sensors [101] M. R. M. Kassim, ‘‘IoT applications in smart agriculture: Issues and
in agriculture using IoT,’’ in AI, Edge and IoT-based Smart Agriculture. challenges,’’ in Proc. IEEE Conf. Open Syst. (ICOS), Nov. 2020,
Amsterdam, The Netherlands: Elsevier, 2022, pp. 361–372. pp. 19–24.
[83] E.-T. Bouali, M. R. Abid, E.-M. Boufounas, T. A. Hamed, and [102] N. Omar, H. Zen, N. Nicole, and W. Waluyo, ‘‘Accuracy and reliability
D. Benhaddou, ‘‘Renewable energy integration into cloud & IoT-based of data in IoT system for smart agriculture,’’ Int. J. Integr. Eng., vol. 12,
smart agriculture,’’ IEEE Access, vol. 10, pp. 1175–1191, 2022. no. 6, pp. 105–116, 2020.
[84] K. Fizza, P. P. Jayaraman, A. Banerjee, D. Georgakopoulos, and [103] L. García, L. Parra, J. M. Jimenez, J. Lloret, P. V. Mauri, and P. Lorenz,
R. Ranjan, ‘‘Evaluating sensor data quality in Internet of Things smart ‘‘DronAway: A proposal on the use of remote sensing drones as mobile
agriculture applications,’’ IEEE Micro, vol. 42, no. 1, pp. 51–60, gateway for WSN in precision agriculture,’’ Appl. Sci., vol. 10, no. 19,
Jan. 2022. p. 6668, Sep. 2020.
[85] G. Gagliardi, M. Lupia, G. Cario, F. Cicchello Gaccio, V. D’Angelo, [104] P. Boonyopakorn and T. Thongna, ‘‘Environment monitoring system
A. I. M. Cosma, and A. Casavola, ‘‘An Internet of Things solution for through LoRaWAN for smart agriculture,’’ in Proc. 5th Int. Conf. Inf.
smart agriculture,’’ Agronomy, vol. 11, no. 11, p. 2140, Oct. 2021. Technol. (InCIT), Oct. 2020, pp. 12–16.
[105] A. Gotsinas, K. Kalovrektis, A. Xenakis, and G. Stamoulis, ‘‘A ZigBee—
[86] M. N. Akhtar, A. J. Shaikh, A. Khan, H. Awais, E. A. Bakar,
Based lightweight wireless sensor system for measuring action potential
and A. R. Othman, ‘‘Smart sensing with edge computing in precision
bio signals in agriculture IoT applications,’’ in Proc. 11th Int. Conf. Inf.,
agriculture for soil assessment and heavy metal monitoring: A review,’’
Intell., Syst. Appl. (IISA, Jul. 2020, pp. 1–7.
Agriculture, vol. 11, no. 6, p. 475, May 2021.
[106] S. Mandal, I. Ali, and S. Saha, ‘‘IoT in agriculture: Smart farming using
[87] F. Bertone, G. Caragnano, S. Ciccia, O. Terzo, and E. Cremonese, ‘‘Green MQTT protocol through cost-effective heterogeneous sensors,’’ in Proc.
data platform: An IoT and cloud infrastructure for data management Int. Conf. Frontiers Comput. Syst. Cham, Switzerland: Springer, 2021,
and analysis in agriculture 4.0,’’ in Complex, Intelligent and Software pp. 903–913.
Intensive Systems. Cham, Switzerland: Springer, 2021, pp. 365–374. [107] E. Symeonaki, K. Arvanitis, and D. Piromalis, ‘‘A context-aware
[88] M. A. Guillén, A. Llanes, B. Imbernón, R. Martínez-España, A. Bueno- middleware cloud approach for integrating precision farming facilities
Crespo, J.-C. Cano, and J. M. Cecilia, ‘‘Performance evaluation of into the IoT toward agriculture 4.0,’’ Appl. Sci., vol. 10, no. 3, p. 813,
edge-computing platforms for the prediction of low temperatures in Jan. 2020.
agriculture using deep learning,’’ J. Supercomput., vol. 77, no. 1, [108] V. Romanov, I. Galelyuka, O. Voronenko, O. Kovyrova, S. Dzyadevych,
pp. 818–840, Jan. 2021. and L. Shkotova, ‘‘Smart sensors and computer devices for agriculture,
[89] H. Khalid, S. J. Hashim, S. M. S. Ahmad, F. Hashim, and food production process control and medicine,’’ in Proc. 29th Int. Conf.
M. A. Chaudhary, ‘‘Robust multi-gateway authentication scheme for Comput. Theory Appl. (ICCTA), Oct. 2019, pp. 9–13.
agriculture wireless sensor network in society 5.0 smart communities,’’ [109] P. Verma, S. Bhutani, S. Srividhya, D. Karthikeyan, and D. C. S. Tong,
Agriculture, vol. 11, no. 10, p. 1020, Oct. 2021. ‘‘Review of Internet of Things towards sustainable development in
[90] M. H. Thi, L. Hoang Son, N. T. Quoc Vinh, and N. Thi Huong Quynh, agriculture,’’ J. Crit. Rev., vol. 7, no. 3, p. 2020, 2019.
‘‘Computing infrastructure of IoT applications in smart agriculture: A [110] M. A. Zamora-Izquierdo, J. Santa, J. A. Martínez, V. Martínez, and
systematical review,’’ in Proc. 6th Int. Conf. Innov. Technol. Intell. Syst. A. F. Skarmeta, ‘‘Smart farming IoT platform based on edge and cloud
Ind. Appl. (CITISIA), Nov. 2021, pp. 1–9. computing,’’ Biosystems Eng., vol. 177, pp. 4–17, Jan. 2019.
[91] A. Tsipis, A. Papamichail, G. Koufoudakis, G. Tsoumanis, [111] S. Gutiérrez, I. Martínez, J. Varona, M. Cardona, and R. Espinosa,
S. E. Polykalas, and K. Oikonomou, ‘‘Latency-adjustable cloud/fog ‘‘Smart mobile LoRa agriculture system based on Internet of Things,’’
computing architecture for time-sensitive environmental monitoring in in Proc. IEEE 39th Central Amer. Panama Conv. (CONCAPAN XXXIX),
olive groves,’’ AgriEngineering, vol. 2, no. 1, pp. 175–205, Mar. 2020. Nov. 2019, pp. 1–6.
[92] M. A. Uddin, M. Ayaz, A. Mansour, E.-H.-M. Aggoune, Z. Sharif, [112] S. V. Mukherji, R. Sinha, S. Basak, and S. P. Kar, ‘‘Smart agriculture using
and I. Razzak, ‘‘Cloud-connected flying edge computing for smart Internet of Things and MQTT protocol,’’ in Proc. Int. Conf. Mach. Learn.,
agriculture,’’ Peer Peer Netw. Appl., vol. 14, no. 6, pp. 3405–3415, Big Data, Cloud Parallel Comput. (COMITCon), Feb. 2019, pp. 14–16.
Nov. 2021. [113] O. Elijah, T. A. Rahman, I. Orikumhi, C. Y. Leow, and M. N. Hindia, ‘‘An
[93] M. A. Rasheed, M. A. Rasheed, and A. M. Mohammed, ‘‘Infrastructure overview of Internet of Things (IoT) and data analytics in agriculture:
for smart agriculture for cotton crops based on the Internet of Things,’’ Benefits and challenges,’’ IEEE Internet Things J., vol. 5, no. 5,
IOP Conf. Ser., Earth Environ. Sci., vol. 761, no. 1, May 2021, pp. 3758–3773, Oct. 2018.
Art. no. 012018. [114] W. Xue-Fen, D. Xing-Jing, B. Wen-Qiang, L. Le-Han, Z. Jian, Z. Chang,
[94] K. T. Chew, V. Raman, and P. H. H. Then, ‘‘Fog-based WSAN for Z. Ling-Xuan, Y. P. Yu-Xiao, and Y. Yi, ‘‘Smartphone accessible
agriculture in developing countries,’’ in Proc. IEEE Int. Conf. Smart agriculture IoT node based on NFC and BLE,’’ in Proc. IEEE Int. Symp.
Internet Things (SmartIoT), Aug. 2021, pp. 289–293. Consum. Electron. (ISCE), Nov. 2017, pp. 78–79.
[115] J. Bišćan, M. Has, K. Tržec, and M. Kušek, ‘‘Selecting IoT commu-
[95] P. Bhale, S. Biswas, and S. Nandi, ‘‘LIENE: Lifetime enhancement for
nication technologies for sensing in agriculture,’’ in Proc. Int. Conf.
6lowpan network using clustering approach use case: Smart agriculture,’’
Broadband Commun. Next Gener. Netw. Multimedia Appl. (CoBCom),
in Proc. Int. Conf. Innov. Community Services, Bamberg, Germany.
Jul. 2022, pp. 1–7.
Cham, Switzerland: Springer, May 2021, pp. 59–75.
[116] K. S. Pratyush Reddy, Y. M. Roopa, K. Rajeev L. N., and N. S. Nandan,
[96] A. Caruso, S. Chessa, S. Escolar, J. Barba, and J. C. López, ‘‘Collection ‘‘IoT based smart agriculture using machine learning,’’ in Proc. 2nd Int.
of data with drones in precision agriculture: Analytical model and LoRa Conf. Inventive Res. Comput. Appl. (ICIRCA), Jul. 2020, pp. 130–134.
case study,’’ IEEE Internet Things J., vol. 8, no. 22, pp. 16692–16704,
[117] G. Lavanya, C. Rani, and P. GaneshKumar, ‘‘An automated low cost
Nov. 2021.
IoT based fertilizer intimation system for smart agriculture,’’ Sustain.
[97] P. K. Reddy Maddikunta, S. Hakak, M. Alazab, S. Bhattacharya, Comput., Informat. Syst., vol. 28, Dec. 2020, Art. no. 100300.
T. R. Gadekallu, W. Z. Khan, and Q.-V. Pham, ‘‘Unmanned aerial vehi- [118] R. Chakravarthy, ‘‘An experimental study of IoT-based topologies on
cles in smart agriculture: Applications, requirements, and challenges,’’ MQTT protocol for agriculture intrusion detection,’’ Meas., Sensors,
IEEE Sensors J., vol. 21, no. 16, pp. 17608–17619, Aug. 2021. vol. 24, Dec. 2022, Art. no. 100470.
[98] F. Susanto, P. Nurtantio, M. A. Soeleman, P. A. Sunarya, and [119] F.-H. Tseng, H.-H. Cho, and H.-T. Wu, ‘‘Applying big data for
E. Nursasongko, ‘‘P2P for prototype embedded system information on intelligent agriculture-based crop selection analysis,’’ IEEE Access,
agriculture result using IoT,’’ in Proc. Int. Seminar Appl. Technol. Inf. vol. 7, pp. 116965–116974, 2019.
Commun. (iSemantic), Sep. 2021, pp. 348–352. [120] V. D. P. N. Kwizera, Z. Li, V. E. Lumorvie, F. Nambajemariya, and X. Niu,
[99] J. Yuan, W. Liu, J. Wang, J. Shi, and L. Miao, ‘‘An efficient framework ‘‘IoT based greenhouse real-time data acquisition and visualization
for data aggregation in smart agriculture,’’ Concurrency Comput., Pract. through message queuing telemetry transfer (MQTT) protocol,’’ Adv.
Exper., vol. 33, no. 10, May 2021, Art. no. e6160. Internet Things, vol. 11, no. 2, pp. 77–93, 2021.

61018 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

[121] D. Glaroudis, A. Iossifides, and P. Chatzimisios, ‘‘Survey, comparison [143] S. K. Smmarwar, G. P. Gupta, and S. Kumar, ‘‘Deep malware detection
and research challenges of IoT application protocols for smart farming,’’ framework for IoT-based smart agriculture,’’ Comput. Electr. Eng.,
Comput. Netw., vol. 168, Feb. 2020, Art. no. 107037. vol. 104, Dec. 2022, Art. no. 108410.
[122] S. Aydin and M. N. Aydin, ‘‘A sustainable multi-layered open data [144] Q. Yan, J. Lou, M. C. Vuran, and S. Irmak, ‘‘Scalable privacy-preserving
processing model for agriculture: IoT based case study using semantic geo-distance evaluation for precision agriculture IoT systems,’’ ACM
web for hazelnut fields,’’ Adv. Sci., Technol. Eng. Syst. J., vol. 5, no. 2, Trans. Sensor Netw., vol. 17, no. 4, pp. 1–30, Nov. 2021.
pp. 309–319, 2020. [145] J. Liu, L. Shu, X. Lu, and Y. Liu, ‘‘Survey of intelligent agricultural IoT
[123] J. Yang, A. Sharma, and R. Kumar, ‘‘IoT-based framework for smart based on 5G,’’ Electronics, vol. 12, no. 10, p. 2336, May 2023.
agriculture,’’ Int. J. Agricult. Environ. Inf. Syst., vol. 12, no. 2, pp. 1–14, [146] A. Alabdulatif, N. N. Thilakarathne, Z. K. Lawal, K. E. Fahim, and
Apr. 2021. R. Y. Zakari, ‘‘Internet of Nano-Things (IoNT): A comprehensive review
[124] S. Parween, R. S. Hameed, and K. Sinha, ‘‘IoT and its real-time from architecture to security and privacy challenges,’’ Sensors, vol. 23,
application in agriculture,’’ in Handbook of Research on Knowledge and no. 5, p. 2807, Mar. 2023.
Organization Systems in Library and Information Science. Hershey, PA, [147] S. Sontowski, M. Gupta, S. S. Laya Chukkapalli, M. Abdelsalam,
USA: IGI Global, 2021, pp. 103–123. S. Mittal, A. Joshi, and R. Sandhu, ‘‘Cyber attacks on smart farming
[125] F. Dahlqvist, M. Patel, A. Rajko, and J. Shulman, Growing Opportunities infrastructure,’’ in Proc. IEEE 6th Int. Conf. Collaboration Internet
in the Internet of Things. Chicago, IL, USA: McKinsey & Company, Comput. (CIC), Dec. 2020, pp. 135–143.
[148] S. Nakhodchi, A. Dehghantanha, and H. Karimipour, ‘‘Privacy and
2019, pp. 1–6.
security in smart and precision farming: A bibliometric analysis,’’
[126] Ö. Köksal and B. Tekinerdogan, ‘‘Architecture design approach for IoT-
in Handbook Big Data Privacy. Springer, 2020, pp. 305–318.
based farm management information systems,’’ Precis. Agricult., vol. 20, [149] M. A. Khan and K. Salah, ‘‘IoT security: Review, blockchain solutions,
no. 5, pp. 926–958, Oct. 2019. and open challenges,’’ Future Gener. Comput. Syst., vol. 82, pp. 395–411,
[127] T. Ojha, S. Misra, and N. S. Raghuwanshi, ‘‘Internet of Things for May 2018.
agricultural applications: The state of the art,’’ IEEE Internet Things J., [150] S. Namasudra, G. C. Deka, P. Johri, M. Hosseinpour, and A. H. Gandomi,
vol. 8, no. 14, pp. 10973–10997, Jul. 2021. ‘‘The revolution of blockchain: State-of-the-art and research challenges,’’
[128] M. Gupta, M. Abdelsalam, S. Khorsandroo, and S. Mittal, ‘‘Security and Arch. Comput. Methods Eng., vol. 28, no. 3, pp. 1497–1515, May 2021.
privacy in smart farming: Challenges and opportunities,’’ IEEE Access, [151] A. Spurr and M. Ausloos, ‘‘Challenging practical features of Bitcoin
vol. 8, pp. 34564–34584, 2020. by the main altcoins,’’ Qual. Quantity, vol. 55, no. 5, pp. 1541–1559,
[129] A. Chakraborty, A. Banerjee, and S. Saha, ‘‘Sensors in precision Oct. 2021.
agriculture: A survey on application, security and privacy,’’ Adv. Appl. [152] S. K. T. Mehedi, A. A. M. Shamim, and M. B. A. Miah, ‘‘Blockchain-
Math. Sci., vol. 18, no. 9, pp. 949–957, Jul. 2019. based security management of IoT infrastructure with Ethereum
[130] A. Yazdinejad, B. Zolfaghari, A. Azmoodeh, A. Dehghantanha, transactions,’’ Iran J. Comput. Sci., vol. 2, no. 3, pp. 189–195, Sep. 2019.
H. Karimipour, E. Fraser, A. G. Green, C. Russell, and E. Duncan, ‘‘A [153] R. Norvill, B. B. F. Pontiveros, R. State, and A. Cullen, ‘‘Visual emulation
review on security of smart farming and precision agriculture: Security for Ethereum’s virtual machine,’’ in Proc. NOMS IEEE/IFIP Netw. Oper.
aspects, attacks, threats and countermeasures,’’ Appl. Sci., vol. 11, no. 16, Manage. Symp., Apr. 2018, pp. 1–4.
p. 7518, Aug. 2021. [154] Z. A. Abduljabbar, V. O. Nyangaresi, H. M. Jasim, J. Ma, M. A. Hussain,
[131] S. Khan, M. A. Bagiwa, A. W. A. Wahab, A. Gani, and A. Abdelaziz, Z. A. Hussien, and A. J. Y. Aldarwish, ‘‘Elliptic curve cryptography-based
‘‘Understanding link fabrication attack in software defined network using scheme for secure signaling and data exchanges in precision agriculture,’’
formal methods,’’ in Proc. IEEE Int. Conf. Informat., IoT, Enabling Sustainability, vol. 15, no. 13, p. 10264, Jun. 2023.
Technol. (ICIoT), Feb. 2020, pp. 555–562. [155] S. Padhy, M. Alowaidi, S. Dash, M. Alshehri, P. P. Malla, S. Routray, and
[132] T. Aishwarya Lakshmi, B. Hariharan, and P. Rekha, ‘‘A survey on energy H. Alhumyani, ‘‘AgriSecure: A fog computing-based security framework
efficient routing protocol for IoT based precision agriculture,’’ in Proc. for agriculture 4.0 via blockchain,’’ Processes, vol. 11, no. 3, p. 757,
Int. Conf. Commun. Electron. Syst. (ICCES), Jul. 2019, pp. 1284–1288. Mar. 2023.
[133] A. Deroussi, I. Alihamidi, L. A. Charaf, A. A. Madi, and A. Addaim, [156] S. Alam, ‘‘Security concerns in smart agriculture and blockchain-based
‘‘Routing protocols for WSN: A survey precision agriculture case study,’’ solution,’’ in Proc. OPJU Int. Technol. Conf. Emerg. Technol. Sustain.
in Proc. IEEE 2nd Int. Conf. Electron., Control, Optim. Comput. Sci. Develop. (OTCON), Feb. 2023, pp. 1–6.
[157] Raad. M. Khaleefah, N. A. M. Al-isawi, M. K. Hussein, and
(ICECOCS), Dec. 2020, pp. 1–7.
N. A. M. Alduais, ‘‘Optimizing IoT data transmission in smart agri-
[134] H. M. Rouzbahani, H. Karimipour, E. Fraser, A. Dehghantanha,
culture: A comparative study of reduction techniques,’’ in Proc. 5th
E. Duncan, A. Green, and C. Russell, ‘‘Communication layer secu-
Int. Congr. Hum.-Comput. Interact., Optim. Robotic Appl. (HORA),
rity in smart farming: A survey on wireless technologies,’’ 2022,
Jun. 2023, pp. 1–5.
arXiv:2203.06013.
[158] I. Abunadi, A. Rehman, K. Haseeb, L. Parra, and J. Lloret, ‘‘Traffic-
[135] A. Gerodimos, L. Maglaras, M. A. Ferrag, N. Ayres, and I. Kantzavelou,
aware secured cooperative framework for IoT-based smart monitoring in
‘‘IoT: Communication protocols and security threats,’’ Internet Things
precision agriculture,’’ Sensors, vol. 22, no. 17, p. 6676, Sep. 2022.
Cyber-Phys. Syst., vol. 3, pp. 1–13, Jan. 2023. [159] M. A. Ali, B. Balamurugan, R. K. Dhanaraj, and V. Sharma, ‘‘IoT and
[136] A. S. Musleh, G. Chen, and Z. Y. Dong, ‘‘A survey on the detection blockchain based smart agriculture monitoring and intelligence security
algorithms for false data injection attacks in smart grids,’’ IEEE Trans. system,’’ in Proc. 3rd Int. Conf. Comput., Autom. Knowl. Manage.
Smart Grid, vol. 11, no. 3, pp. 2218–2234, May 2020. (ICCAKM), Nov. 2022, pp. 1–7.
[137] D. Ye and T.-Y. Zhang, ‘‘Summation detector for false data-injection [160] S. Anand and A. Sharma, ‘‘Comprehensive analysis of services towards
attack in cyber-physical systems,’’ IEEE Trans. Cybern., vol. 50, no. 6, enhancing security in IoT-based agriculture,’’ Meas., Sensors, vol. 24,
pp. 2338–2345, Jun. 2020. Dec. 2022, Art. no. 100599.
[138] G. Idoje, T. Dagiuklas, and M. Iqbal, ‘‘Survey for smart farming [161] Md. R. A. Asif, K. F. Hasan, M. Z. Islam, and R. Khondoker,
technologies: Challenges and issues,’’ Comput. Electr. Eng., vol. 92, ‘‘STRIDE-based cyber security threat modeling for IoT-enabled precision
Jun. 2021, Art. no. 107104. agriculture systems,’’ in Proc. 3rd Int. Conf. Sustain. Technol. Ind. 4.0
[139] K. Demestichas, N. Peppes, and T. Alexakis, ‘‘Survey on security threats (STI), Dec. 2021, pp. 1–6.
in agricultural IoT and smart farming,’’ Sensors, vol. 20, no. 22, p. 6458, [162] V. K. V. V. Bathalapalli, S. P. Mohanty, E. Kougianos, V. P. Yanambaka,
Nov. 2020. B. K. Baniya, and B. Rout, ‘‘A PUF-based approach for sustainable
[140] L. Barreto and A. Amaral, ‘‘Smart farming: Cyber security challenges,’’ cybersecurity in smart agriculture,’’ in Proc. 19th OITS Int. Conf. Inf.
in Proc. Int. Conf. Intell. Syst. (IS), Sep. 2018, pp. 870–876. Technol. (OCIT), Dec. 2021, pp. 375–380.
[141] P. N. Bahrami, A. Dehghantanha, T. Dargahi, R. M. Parizi, [163] M. A. Ferrag, L. Shu, H. Djallel, and K.-K.-R. Choo, ‘‘Deep
K.-K. R. Choo, and H. H. S. Javadi, ‘‘Cyber kill chain-based taxonomy learning-based intrusion detection for distributed denial of service
of advanced persistent threat actors: Analogy of tactics, techniques, and attack in agriculture 4.0,’’ Electronics, vol. 10, no. 11, p. 1257,
procedures,’’ J. Inf. Process. Syst., vol. 15, no. 4, pp. 865–889, 2019. May 2021.
[142] E. M. Dovom, A. Azmoodeh, A. Dehghantanha, D. E. Newton, [164] R. Kloibhofer, E. Kristen, and A. Ameri E, ‘‘LoRaWAN with HSM
R. M. Parizi, and H. Karimipour, ‘‘Fuzzy pattern tree for edge malware as a security improvement for agriculture applications-evaluation,’’ in
detection and categorization in IoT,’’ J. Syst. Archit., vol. 97, pp. 1–7, Proc. Int. Conf. Comput. Saf., Rel., Secur. Cham, Switzerland: Springer,
Aug. 2019. Sep. 2021, pp. 128–140.

VOLUME 12, 2024 61019

 A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

[165] K. A. Awan, I. U. Din, A. Almogren, and H. Almajed, ‘‘Agritrust— [184] P. R. Emparanza, N. Hongkarnjanakul, D. Rouquette, C. Schwob, and
A trust management approach for smart agriculture in cloud-based L. Mezeix, ‘‘Land cover classification in Thailand’s eastern economic
Internet of Agriculture Things,’’ Sensors, vol. 20, no. 21, p. 6174, corridor (EEC) using convolutional neural network on satellite images,’’
2020. Remote Sens. Appl., Soc. Environ., vol. 20, Nov. 2020, Art. no. 100394.
[166] K. Haseeb, I. Ud Din, A. Almogren, and N. Islam, ‘‘An energy [185] P. Bharati and A. Pramanik, ‘‘Deep learning techniques-R-CNN to
efficient and secure IoT-based WSN framework: An application to smart mask R-CNN: A survey,’’ in Computational Intelligence in Pattern
agriculture,’’ Sensors, vol. 20, no. 7, p. 2081, Apr. 2020. Recognition: Proceedings of CIPR 2019. Springer, 2020, pp. 657–668.
[167] J. Song, Q. Zhong, W. Wang, C. Su, Z. Tan, and Y. Liu, ‘‘FPDP: [186] K. Seetharaman and T. Mahendran, ‘‘Leaf disease detection in banana
Flexible privacy-preserving data publishing scheme for smart plant using Gabor extraction and region-based convolution neural
agriculture,’’ IEEE Sensors J., vol. 21, no. 16, pp. 17430–17438, network (RCNN),’’ J. Inst. Eng. (India), Ser. A, vol. 103, no. 2,
Aug. 2021. pp. 501–507, Jun. 2022.
[168] M. Shyamala Devi, R. Suguna, A. S. Joshi, and R. A. Bagate, ‘‘Design [187] V. N. Thanh Le, G. Truong, and K. Alameh, ‘‘Detecting weeds from crops
of IoT blockchain based smart agriculture for enlightening safety and under complex field environments based on faster RCNN,’’ in Proc. IEEE
security,’’ in Proc. Int. Conf. Emerg. Technol. Comput. Eng. Cham, 8th Int. Conf. Commun. Electron. (ICCE), Jan. 2021, pp. 350–355.
Switzerland: Springer, 2019, pp. 7–19. [188] H. N. Saha and R. Roy, ‘‘ML-based smart farming using LSTM,’’
[169] A. Sultan, M. A. Mushtaq, and M. Abubakar, ‘‘IoT security issues via in Smart Agriculture Automation Using Advanced Technologies: Data
blockchain: A review paper,’’ in Proc. Int. Conf. Blockchain Technol., Analytics and Machine Learning, Cloud Architecture, Automation and
Mar. 2019, pp. 60–65. IoT. Springer, 2021, pp. 89–111.
[170] C. Nartey, E. T. Tchao, J. D. Gadze, E. Keelson, G. S. Klogo, [189] P. S. B. C. Suryo, W. Mustika, I, O. Wahyunggoro, and H. S. Wasisto,
B. Kommey, and K. Diawuo, ‘‘On blockchain and IoT integration ‘‘Improved time series prediction using LSTM neural network for smart
platforms: Current implementation challenges and future perspec- agriculture application,’’ in Proc. 5th Int. Conf. Sci. Technol. (ICST),
tives,’’ Wireless Commun. Mobile Comput., vol. 2021, pp. 1–25, vol. 1, Jul. 2019, pp. 1–4.
Apr. 2021. [190] E. A. Devi, S. Gopi, U. Padmavathi, S. R. Arumugam, S. P. Premnath,
[171] A. A. Alfa, J. K. Alhassan, O. M. Olaniyi, and M. Olalere, ‘‘Blockchain and D. Muralitharan, ‘‘Plant disease classification using CNN-LSTM
technology in IoT systems: Current trends, methodology, problems, techniques,’’ in Proc. 5th Int. Conf. Smart Syst. Inventive Technol.
applications, and future directions,’’ J. Reliable Intell. Environ., vol. 7, (ICSSIT), Jan. 2023, pp. 1225–1229.
no. 2, pp. 115–143, Jun. 2021. [191] S. Sharma, S. Rai, and N. C. Krishnan, ‘‘Wheat crop yield prediction
[172] A. Ali, H. A. Rahim, M. F. Pasha, R. Dowsley, M. Masud, J. Ali, using deep LSTM model,’’ 2020, arXiv:2011.01498.
and M. Baz, ‘‘Security, privacy, and reliability in digital healthcare [192] T. de Camargo, M. Schirrmann, N. Landwehr, K.-H. Dammer, and
systems using blockchain,’’ Electronics, vol. 10, no. 16, p. 2034, M. Pflanz, ‘‘Optimized deep learning model as a basis for fast UAV
Aug. 2021. mapping of weed species in winter wheat crops,’’ Remote Sens., vol. 13,
no. 9, p. 1704, Apr. 2021.
[173] B. K. Mohanta, D. Jena, S. Ramasubbareddy, M. Daneshmand, and
A. H. Gandomi, ‘‘Addressing security and privacy issues of IoT [193] M. T, K. Makkithaya, and N. V. G, ‘‘A federated learning-based crop yield
using blockchain technology,’’ IEEE Internet Things J., vol. 8, no. 2, prediction for agricultural production risk management,’’ in Proc. IEEE
pp. 881–888, Jan. 2021. Delhi Sect. Conf. (DELCON), Feb. 2022, pp. 1–7.
[194] S. Islam, S. Jamwal, M. H. Mir, and Q. R. Khan, ‘‘IoT-smart agriculture:
[174] C. Yang, L. Tan, N. Shi, B. Xu, Y. Cao, and K. Yu, ‘‘Auth-
Comparative study on farming applications and disease prediction of
PrivacyChain: A blockchain-based access control framework with
apple crop using machine learning,’’ Iraqi J. Sci., vol. 63, pp. 5520–5533,
privacy protection in cloud,’’ IEEE Access, vol. 8, pp. 70604–70615,
Dec. 2022.
2020.
[195] K. S. Kumari, S. L. A. Haleem, G. Shivaprakash, M. Saravanan,
[175] S. Malik, V. Dedeoglu, S. S. Kanhere, and R. Jurdak, ‘‘TrustChain:
B. Arunsundar, and T. K. S. Pandraju, ‘‘Agriculture monitoring system
Trust management in blockchain and IoT supported supply chains,’’
based on Internet of Things by deep learning feature fusion with classifi-
in Proc. IEEE Int. Conf. Blockchain (Blockchain), Jul. 2019,
cation,’’ Comput. Electr. Eng., vol. 102, Sep. 2022, Art. no. 108197.
pp. 184–193.
[196] K. Ramana, R. Aluvala, M. R. Kumar, G. Nagaraja, A. V. Krishna, and
[176] M. Ayaz, M. Ammad-Uddin, Z. Sharif, A. Mansour, and E. M. Aggoune,
P. Nagendra, ‘‘Leaf disease classification in smart agriculture using deep
‘‘Internet-of-Things (IoT)-based smart agriculture: Toward making
neural network architecture and IoT,’’ J. Circuits, Syst. Comput., vol. 31,
the fields talk,’’ IEEE Access, vol. 7, pp. 129551–129583,
no. 15, Oct. 2022, Art. no. 2240004.
2019. [197] M. F. Aslan, A. Durdu, K. Sabanci, E. Ropelewska, and S. S. Gültekin,
[177] T. Pavithra. P and Baranidharan. B, ‘‘An analysis on application ‘‘A comprehensive survey of the recent studies with UAV for precision
of deep learning techniques for precision agriculture,’’ in Proc. agriculture in open fields and greenhouses,’’ Appl. Sci., vol. 12, no. 3,
3rd Int. Conf. Inventive Res. Comput. Appl. (ICIRCA), Sep. 2021, p. 1047, Jan. 2022.
pp. 1538–1545. [198] N. Fatima, S. A. Siddiqui, and A. Ahmad, ‘‘IoT-based smart greenhouse
[178] O. Jouini, K. Sethom, and R. Bouallegue, ‘‘The impact of the with disease prediction using deep learning,’’ Int. J. Adv. Comput. Sci.
application of deep learning techniques with IoT in smart agriculture,’’ Appl., vol. 12, no. 7, pp. 1–11, 2021.
in Proc. Int. Wireless Commun. Mobile Comput. (IWCMC), Jun. 2023, [199] G. Garg, S. Gupta, P. Mishra, A. Vidyarthi, A. Singh, and A. Ali,
pp. 977–982. ‘‘CROPCARE: An intelligent real-time sustainable IoT system for crop
[179] M. Altalak, M. A. Uddin, A. Alajmi, and A. Rizg, ‘‘Smart agriculture disease detection using mobile vision,’’ IEEE Internet Things J., vol. 10,
applications using deep learning technologies: A survey,’’ Appl. Sci., no. 4, pp. 2840–2851, Feb. 2023.
vol. 12, no. 12, p. 5919, Jun. 2022. [200] R. Narayan and N. Nageshwari, ‘‘A Wi-Fi based smart energy efficient
[180] M. H. Saleem, J. Potgieter, and K. M. Arif, ‘‘Automation in wireless sensor network for yield prediction in precision agriculture,’’
agriculture by machine and deep learning techniques: A review of in Proc. Int. Conf. Circuits, Controls Commun. (CCUBE), Dec. 2021,
recent developments,’’ Precis. Agricult., vol. 22, no. 6, pp. 2053–2091, pp. 1–5.
Dec. 2021. [201] Y. Osman, R. Dennis, and K. Elgazzar, ‘‘Yield estimation and visualiza-
[181] S. M. Hassan and A. K. Maji, ‘‘Plant disease identification using tion solution for precision agriculture,’’ Sensors, vol. 21, no. 19, p. 6657,
a novel convolutional neural network,’’ IEEE Access, vol. 10, Oct. 2021.
pp. 5390–5401, 2022. [202] P. Kaur, S. Harnal, R. Tiwari, S. Upadhyay, S. Bhatia, A. Mashat, and
[182] A. Allmendinger, M. Spaeth, M. Saile, G. G. Peteinatos, and R. Gerhards, A. M. Alabdali, ‘‘Recognition of leaf disease using hybrid convolutional
‘‘Precision chemical weed management strategies: A review and a design neural network by applying feature reduction,’’ Sensors, vol. 22, no. 2,
of a new CNN-based modular spot sprayer,’’ Agronomy, vol. 12, no. 7, p. 575, Jan. 2022.
p. 1620, Jul. 2022. [203] S. M. H. Rizvi, A. Naseer, S. U. Rehman, S. Akram, and V. Gruhn,
[183] D. Indira, J. Goddu, B. Indraja, V. M. L. Challa, and B. Manasa, ‘‘A review ‘‘Revolutionizing agriculture: Machine and deep learning solutions
on fruit recognition and feature evaluation using CNN,’’ Mater. Today, for enhanced crop quality and weed control,’’ IEEE Access, vol. 12,
Proc., vol. 80, pp. 3438–3443, Jan. 2023. pp. 11865–11878, 2024.

61020 VOLUME 12, 2024

A. Naseer et al.: Systematic Literature Review of the IoT in Agriculture

ASMA NASEER received the M.S. and Ph.D. AWAIS AHMAD received the Ph.D. degree in
degrees in computer science from the National computer science and engineering from Kyung-
University of Computer and Emerging Science pook National University, Daegu, South Korea.
(NUCES), Lahore, Pakistan, in 2008 and 2019, In 2014, he was a Researcher with INTEL-
respectively. NTU, National Taiwan University, Taiwan, where
She is currently an Associate Professor with he was working on Wukong Project (Smart
NUCES. Prior to joining NUCES, she was a Home). He is currently an Assistant Professor
Faculty Member with the Department of Computer with the Department of Information Systems,
Science, University of Management and Technol- College of Computer and Information Sciences,
ogy (UMT)from 2010 to 2021. She is a dedicated Imam Muhammad Ibn Saud Islamic University,
Artificial Intelligence (AI) and Machine Learning (ML) Expert, with almost Riyadh, Saudi Arabia. Since 2013, he has been published in more than
15 years of experience. She has been awarded full scholarships from the 170 international journals (Cumulative Impact Factor: more than 300)/con-
Higher Education Commission (HEC), Pakistan, and other local and foreign ferences/book chapters in various reputed IEEE TRANSACTIONS, IEEE
bodies for the master’s and Ph.D. studies. Magazines, ACM Transactions, Elsevier, and Springer journals, whereas in
leading conferences, such as IEEE GLOBECOM, IEEE INFOCOM, IEEE
LCN, and IEEE ICC, respectively. His research interests include cyber
MUHAMMAD SHMOON received the degree
security, deep learning, machine learning, artificial intelligence, denoising
in computer science from the National University
and demosacking, big data analytics, and the Internet of Things. He was
of Computer and Emerging Science (NUCES),
a recipient of four prestigious awards, such as the IEEE Best Research
Lahore, Pakistan, in 2023. He is currently a
Paper Award from the International Workshop on Ubiquitous Sensor Systems
Research Officer with NUCES and a Software
(UWSS 2015), in conjunction with the Smart World Congress (SWC
Engineer with System Ltd. His expertise lies in
2015), Beijing, China, August; the Research Award from President of
AI, machine learning, deep learning, and data
Bahria University Islamabad, Pakistan, in 2011; the Best Paper Nomination
science. In addition to using software development
Award from WCECS 2011 at UCLA, USA; and the Best Paper Award
expertise, his dynamic work includes contributing
from the 1st Symposium on CS&E, Moju Resort, South Korea, in 2013.
to innovative research initiatives.
He was also awarded as a Best Outgoing Researcher of CCMP Labs.
He is serving as a Guest Editor in various Elsevier and Springer journals,
TANZEELA SHAKEEL received the B.S. degree such as Future Generation Computer Systems (Elsevier), Sustainable Cities
in computer engineering and the M.S. degree in and Society (Elsevier), and Computational Intelligence and Complexity
electrical engineering from COMSATS University (Springer), Multimedia Tools and Applications (Springer), IEEE ACCESS,
Islamabad (CUI), Lahore, Pakistan, in 2012 and Journal of Real-Time Image Processing (Springer), and Current Opinion in
2016, respectively. She is currently employed as Biomedical Engineering (Elsevier). Moreover, he served as a TPC member or
a Research Officer with the National University reviewer in more than 20 international conferences and workshops, including
of Computer and Emerging Science (NUCES), IEEE Globecom, IEEE ICC, IEEE Infocom, and ACM SAC.
Lahore, and the Lead Industrial IoT Engineer
with AMS IoT Ltd. Before joining NUCES
and Industry, she was a Faculty Member with
the Department of Computer Science, University of Management and
Technology (UMT), from 2012 to 2022. She is a devoted IoT and control
system professional with almost 12 years of experience. Her research
interests include the IoT, AI, ML, embedded systems, cybersecurity, and
control systems.

SHAFIQ UR REHMAN received the M.S. degree

in computer science from Dresden University of VOLKER GRUHN received the M.S. and Ph.D.
Technology, Dresden, Germany, and the Ph.D. degrees in computer science from the Univer-
degree in computer science from the Department sity of Dortmund, Germany, in 1987 and 1991,
of Software Engineering, University of Duisburg- respectively. Then, he was with the Fraunhofer
Essen, Germany, in 2020. He was a Consultant Institute for Software and System Technology.
(Requirements Engineer) in well-renowned inter- In 1997, he co-founded adesso AG. Since 2010,
national organizations in Germany. He is currently he has been the Chair of Software Engineering
an Assistant Professor with the College of Com- with the University of Duisburg-Essen, Germany.
puter and Information Sciences, Imam Moham- He supervised several Ph.D. students. He has
mad Ibn Saud Islamic University (IMSIU), Riyadh, Saudi Arabia. He has published more than 300 research papers in
published several research papers in high-ranked international conferences ISI-indexed journals and high-ranked international conferences. His research
and ISI-indexed journals. He is involved in different international-funded interests include methods for industrial software engineering and the effects
projects in the field of cyber-physical systems and cybersecurity. His research of digital transformation on enterprises. He is the Supervisory Board
interests include AI, cyber-physical systems, cybersecurity, and requirements Chairperson.
engineering.

VOLUME 12, 2024 61021