Computers and Electronics in Agriculture 142 (2017) 283–297

Contents lists available at ScienceDirect

Computers and Electronics in Agriculture

journal homepage: www.elsevier.com/locate/compag

Review

Review of IoT applications in agro-industrial and environmental fields

Jesús Martín Talavera a, Luis Eduardo Tobón b, Jairo Alejandro Gómez b,⇑, María Alejandra Culman a,
Juan Manuel Aranda c, Diana Teresa Parra a, Luis Alfredo Quiroz b, Adolfo Hoyos b, Luis Ernesto Garreta b
a
Universidad Autónoma de Bucaramanga, Avenida 42 No. 48-11, Bucaramanga, Colombia
b
Pontificia Universidad Javeriana, Calle 18 No. 118-250, Cali, Colombia
c
Pontificia Universidad Javeriana, Carrera 7 No. 40-62, Bogotá D.C, Colombia

a r t i c l e i n f o a b s t r a c t

Article history: This paper reviews agro-industrial and environmental applications that are using Internet of Things (IoT).
Received 28 March 2017 It is motivated by the need to identify application areas, trends, architectures and open challenges in
Received in revised form 26 July 2017 these two fields. The underlying survey was developed following a systematic literature review using
Accepted 9 September 2017
academic documents written in English and published in peer-reviewed venues from 2006 to 2016.
Available online 18 September 2017
Selected references were clustered into four application domains corresponding to: monitoring, control,
logistics, and prediction. Implementation-specific details from each selected reference were compiled to
Keywords:
create usage distributions of sensors, actuators, power sources, edge computing modules, communication
Internet of things
IoT
technologies, storage solutions, and visualization strategies. Finally, the results from the review were
Agro-industry compiled into an IoT architecture that represents a wide range of current solutions in agro-industrial
Environmental monitoring and environmental fields.
Systematic literature review Ó 2017 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
2. Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
3. Conduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
4. Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
4.1. Answer to the first research question . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
4.2. Answer to the second research question . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
5. Recent works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
5.1. Communications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
5.2. Energy management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
5.3. Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
5.4. Logistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
6. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
6.1. Limitations and open challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
6.2. Proposed architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
7. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295

⇑ Corresponding author at: Center of Excellence and Appropriation on Internet of Things, Pontificia Universidad Javeriana, Calle 18 No. 118-250, Cali, Colombia.
E-mail addresses: [email protected] (J.M. Talavera), [email protected] (L.E. Tobón), [email protected] (J.A. Gómez), mculman@
unab.edu.co (M.A. Culman), [email protected] (J.M. Aranda), [email protected] (D.T. Parra), [email protected] (L.A. Quiroz), adolfo.
[email protected] (A. Hoyos), [email protected] (L.E. Garreta).

http://dx.doi.org/10.1016/j.compag.2017.09.015
0168-1699/Ó 2017 Elsevier B.V. All rights reserved.
284 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297

1. Introduction data mining and synthesis methodologies. For this study, the two
research questions considered were:
The widespread of Internet in the last two decades brought
countless benefits to citizens and organizations around the world. 1. What are the main technological solutions of the Internet of
Arguably the most important benefit was the ability to consume Things in agro-industrial and environmental fields?
and produce data and services in real time. Recently, the Internet 2. Which infrastructure and technology are using the main solu-
of Things is promising to bring the same benefits to everyday tions of IoT in agro-industrial and environmental fields?
objects, giving us a way to extend our perception and our ability
to modify the environment around us. To collect information, authors performed an Internet search
In this context, agro-industrial and environmental fields are using various academic digital libraries and search engines.
ideal candidates for the deployment of IoT solutions because they Obtained results were manually compiled in order to select the
occur in wide areas that need to be continuously monitored and best information sources to answer the two research questions.
controlled. At the same time, IoT opens new opportunities beyond After analyzing the results, digital libraries and search engines
ground floor automation when the collected data are used to feed described in Table 1 were chosen based on their scientific and tech-
machine learning algorithms to provide predictions (Saville et al., nical content, as well as their close relationship to areas of knowl-
2015), easing decision planning and decision making for owners, edge associated with the objective of this paper.
managers, and policy makers. The next step was to define search terms and a consistent pro-
IoT can be used at different levels in the agro-industrial produc- cedure to seek scientific and technical documentation in the digital
tion chain (Medela et al., 2013). It can help to evaluate field vari- libraries and search engines. To define the search terms, a set of
ables such as soil state, atmospheric conditions, and biomass of keywords was selected from the research questions to create two
plants or animals. It can also be used to assess and control variables groups of words which are shown in Table 2. Each group contained
such as temperature, humidity, vibration, and shock during the consolidated expressions with synonyms or terms with related
product transport (Pang et al., 2015). It can be used to monitor meaning. Group 1 included words associated with the Internet of
and predict the product state and its demand on shelves or inside Things, while Group 2 contained a set of terms related to the
refrigerators. In addition, it can provide information to the final agro-industry and environment. Logical operators supported by
user/consumer about the origin and properties of the product. the advanced search of digital libraries were used to construct
The IoT applied to the agro-industry can contribute to create an search strings, based on the two research questions, combining
informed, connected, developed and adaptable rural community. terms from Groups 1 and 2 of Table 2. The general structure of
Under the IoT paradigm, low-cost electronic devices can improve the search queries that were applied to the information sources
human interaction with the physical world, and the computing is presented in Table 3.
power and software available on the Internet can provide valuable In order to ensure the quality of papers, only those that passed
analytics. In summary, IoT can be an important tool in the years to the following criteria were considered in the reviewing process.
come for people interacting within an agro-industrial system: sup-
pliers, farmers, technicians, distributors, business men, consumers,  Documents published in peer-reviewed conferences, peer-
and government representatives. reviewed journals, papers from computer science or engineer-
IoT can be incorporated into environmental applications to pro- ing organizations, patents, or technical reports.
duce dense and real-time maps of air and water pollution, noise  Documents published in English.
level (Torres-Ruiz et al., 2016; Hachem et al., 2015), temperature,  Documents published between 2006 and 2016 (both years
and harmful radiation among others. It can be used to collect and inclusive).
store environmental records, check the compliance of environmen-
tal variables with local policies, trigger alerts, or send recommen- If the main topic of a given paper was irrelevant or if it was out-
dation messages to citizens and authorities (Liu et al., 2013). side the scope of this study, it was deleted. Then, a selection crite-
Once the data reach the cloud, governments can feed predictive rion was applied in order to reduce the number of papers found
models to forecast environmental variables, and identify and track during the search and to get a small number of high-quality
pollution sources over time and space, ultimately leading to faster sources that could be used to answer the research questions. This
and better decisions to ensure a safe and healthy environment for
all citizens.
Based on the potential of IoT applications in agro-industrial and Table 1
environmental fields described in the previous paragraphs, this Information sources used for the search phase.
paper aims to identify the current state of solutions in these fields, Source Type URL
as well as the trends, architectures, technologies and open chal-
IEEE Xplore Digital http://ieeexplore.ieee.org/Xplore/
lenges. This paper uses a Systematic Literature Review (SLR) based Library home.jsp
on a methodology proposed by Kitchenham and Charters (2007), in Science Direct Digital http://www.sciencedirect.com/
order to make it unbiased in terms of information selection, pro- Library
ACM Digital Library Digital http://dl.acm.org/dl.cfm
cessing, and presentation of results.
Library
The paper is structured as follows. Sections 2–4 describe the Citeseer library Digital http://citeseer.ist.psu.edu/
stages of planning, conduction, and results of the SLR. Section 5 Library advanced_search
outlines some recent works that were published online after the Sensors Digital http://www.mdpi.com/journal/
SLR was concluded. Section 6 includes a discussion of the obtained Library sensors
Scopus Search http://www.scopus.com/
results, and Section 7 presents the conclusions from this study.
Engine
Microsoft Academic Search http://academic.research.
Search Engine microsoft.com/
2. Planning Microsoft Academic Search https://academic.microsoft.com/
Engine
Google Scholar Search https://scholar.google.com/
During this stage of the SLR, the protocol was defined. This
Engine
included: research questions, search strategies, selection criteria,
 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297 285

Table 2
Words used for the search query.

Group 1:
Internet of Things, Web of Things.
Group 2:
Agricultural industry, Agricultural products, Agriculture, Agribusiness, Agroindustry, Air pollution, Apiculture, Aquaculture, Product Traceability, Smart Agriculture,
Greenhouses, Harvesting, Horticulture, Husbandry, Irrigation, Livestock, Climate, Feeding, Fertilizers, Forestry, Weather, Animal production, Animal sensing, Animal
tracking, Animal trade control, Avalanche, Bio-fuel, Biological production, Bio-monitoring, Breeding, Cereals, Crop, Dairy, Drones, Drought, Earthquake sensor,
Environmental monitoring, Equipment status, Farm, Farming, Feed production, Fish, Fishery, Flooding, Food chain, Food production, Forecast, Forest fire, Freeze,
Fruit, Fruit storage, Grassland, Heating, Landslide, Meat, Pest, Plant, Poultry, Seed, Vegetable, Waste, Water.

Table 3
Algorithm: search query-(Group 1) AND (Group 2).

TITLE-ABS-KEY (‘‘Internet of Things” OR ‘‘Web of Things”) AND (‘‘Agricultural industry” OR ‘‘Agricultural products” OR agriculture OR agribusiness OR agroindustry OR
‘‘Air pollution” OR ‘‘Apiculture” OR aquaculture OR ‘‘Product Traceability” OR greenhouses OR harvesting OR horticulture OR husbandry OR irrigation OR livestock
OR climate OR feeding OR fertilizers OR forestry OR weather OR ‘‘Animal production” OR ‘‘Animal sensing” OR ‘‘Animal tracking” OR ‘‘Animal trade control” OR
avalanche OR biofuel OR ‘‘Biological production” OR biomonitoring OR breeding OR cereals OR crop OR dairy OR drones OR drought OR ‘‘Earthquake sensor” OR
‘‘Environmental monitoring” OR ‘‘Equipment status” OR farm OR farming OR ‘‘Feed production” OR fish OR fishery OR flooding OR ‘‘Food chain” OR ‘‘Food
production” OR forecast OR ‘‘Forest fire” OR freeze OR fruit OR ‘‘Fruit storage” OR grassland OR heating OR landslide OR meat OR pest OR plant OR poultry OR seed OR
vegetable OR waste OR water)

Table 4
 IC based on quality analysis: in this phase, a quality analysis
Form used to extract data for each study.
was applied to remaining papers and those that did not comply
Data retrieved Description any of the following four quality criteria (QC) were discarded:
Title Title of the main study – QC1: Does the study present a comprehensive solution of IoT
Year Publication year of the study for agro-industry or environment?
Institution Name of institution(s) leading the research
– QC2: Does the paper show details of the infrastructure and/
Country Country that developed the research
Source Conference, journal, or book containing the main or technologies used to implement the proposed solution?
study – QC3: Does the paper present a state of the art or related
Solution Name of the IoT solution described work?
Domain and Area of agro-industry or environment where IoT was – QC4: Does the paper present an analysis of the results?
subdomain applied
Architecture model Description of the architecture used, its scope and
limitation The next stage of the SLR was data mining and synthesis. The
Sensors Information about sensor type and sensor count per goal here was to extract the information needed to answer the
node in the solution research questions in an objective manner. The information fields
Power source Mechanisms used to power IoT devices
extracted for each study are presented in Table 4.
Edge computing Information about computing platforms, hardware
architecture, the number of nodes, topology
(homogeneous vs. heterogeneous).
Connectivity and Technologies used for transmitting data
3. Conduction
communication
Data storage Techniques used for storing data (locally, distributed, The protocol described in the previous section was used to
and cloud-based), as well as data access search, select and evaluate preliminary papers. For the search pro-
methodologies
cess, the query defined in Table 3 was passed to information
Data processing and Algorithms and methodologies for processing and
visualization analyzing data (data aggregation, data fusion, sources given in Table 1. The search was limited to title, abstract
machine learning, pattern recognition, big data), and and keywords.
models to visualize them Fig. 1 illustrates the conduction process discriminated by the
Deployment scenario Characteristics of the deployment site for the IoT academic database and search engine used, highlighting the key
solution
steps followed to select relevant studies for this review. Initially,
3578 studies were recovered from electronic databases. Firstly,
involved using inclusion criteria (IC) and quality criteria (QC), duplicates were excluded, i.e. studies available in more than one
which were defined in a three-phase process. database, eliminating 849 copies. Out of the 2729 remaining stud-
ies, 2652 were initially screened based on inclusion and exclusion
 IC based on abstracts: in this phase, authors discarded papers criteria applied to the title, abstract, and keywords. These papers
found in the search stage based on the information provided were marked to be downloaded, and references that could not be
in their abstracts. Papers that satisfied the first inclusion crite- retrieved were discarded. Afterward, these studies were evaluated
rion were kept for further processing, i.e. papers that discussed using quality criteria obtaining 720 studies. These studies were
IoT solutions applied to agro-industry and environment. Papers used to extract the data defined in Table 4. Finally, only 72 main
with little relevant information in their abstract were temporar- studies were selected based on their quality for the final conduc-
ily kept in the list and were processed in the next stage. It is tion phase and used to extract results presented in the next
important to highlight that quality criteria were not considered section.
in this phase. It is worth to note that more than 90% of included papers were
 IC based on full reading: in this phase, papers that did not retrieved from two sources: IEEExplore (76.4%) and Scopus (13.9%).
address the search terms shown in Table 2 were removed. This In contrast, the least effective sources of information were Micro-
means that even though those papers contained the search soft Academic Search and Microsoft Academic. They retrieved
terms in their abstract, they only represented minor aspects of 668 papers during the first stage of the conduction phase (repre-
them. senting 25.2% of all retrieved papers, and only behind IEEExplore
286 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297

Fig. 1. Process followed in the SLR to select main studies.

Fig. 2. Distribution of papers selected by publication year.

with 45%). However, only 3.1% of them were included for the next 4. Results
reviewing phase, a number well below the 39.8% of papers
included from IEEExplore. These facts can be explained because This phase presents results of the SLR in order to answer the
IEEExplore and Scopus have complete and usable advanced search two research questions based on the information extracted from
systems and they have been operating continuously unlike Micro- main studies selected.
soft’s counterpart (Sinha et al., 2015a).
Fig. 2 enumerates the number of primary studies classified by
publication year. It can be seen that most of the selected papers
were published between 2012 and 2016. It should be highlighted 4.1. Answer to the first research question
that the small number of papers shown in 2016 can be explained
because the initial search was made in April of that year. To identify the main technological solutions of IoT in agro-
Fig. 3 summarizes the country of origin of selected papers. industry and environmental fields, studies were grouped into four
Every continent of the world is represented by at least one research technological domains, corresponding to: (1) monitoring, (2) con-
work. China is the country that contributed with the largest num- trol, (3) prediction and (4) logistics. Results are summarized in
ber of papers. Asia has more than half of contributions and America Table 5 and illustrated in Fig. 4. From this figure, it can be seen that
has less than ten percent of them, showing a huge potential for this most of the selected studies were focused on monitoring (62%), fol-
continent. lowed by control (25%), logistics (7%), and prediction (6%).
 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297 287

Fig. 3. Distribution of papers selected by country.

Table 5
Clustering of main studies by application domain.

Domain Main study

Monitoring (Hussain et al., 2006; Lu et al., 2010; Pokrić et al., 2014;
Postolache et al., 2014; Sawant et al., 2014; Ehsan et al., 2012;
Langendoen et al., 2006; Chen et al., 2014; Liu et al., 2013; Islam
et al., 2014; Kuroda et al., 2015; Fourati et al., 2014; Kar and Kar,
2015; Chen et al., 2015; Medela et al., 2013; Zou, 2014;
Diedrichs et al., 2014; Mittal et al., 2012; De La Concepcion
et al., 2014; Jardak et al., 2009; Vo et al., 2013; Tarange et al.,
2015; Kodali et al., 2014; Sinha et al., 2015b; Eom et al., 2014;
Sun et al., 2012; Hakala et al., 2008; Jain et al., 2008;
Watthanawisuth et al., 2009; Nguyen et al., 2015; Lee et al.,
2013; Ma et al., 2012; Jayaraman et al., 2015a; Jayaraman et al.,
2015b; Soontranon et al., 2014; Hashim et al., 2015; Zhao and
Zhu, 2015; Mathurkar et al., 2014; Kiyoshi et al., 2008;
Postolache et al., 2013; Mafuta et al., 2012; Feng et al., 2012;
Xijun et al., 2009; Gutiérrez et al., 2014; Sarangi et al., 2016;
Fang et al., 2014)
Control (Yoo et al., 2007; Kanoun et al., 2014; Sales et al., 2015; Chavez-
Burbano et al., 2014; Ryu et al., 2015; Pahuja et al., 2013; Xu
et al., 2015; Ye et al., 2013; Jiao et al., 2014; Jiber et al., 2011; Fig. 4. Distribution of papers selected by application domain.
Shuwen and Changli, 2015; Culibrina and Dadios, 2015;
Kaewmard and Saiyod, 2014; Li et al., 2014; Tao et al., 2014;
Smarsly, 2013; Roy et al., 2015) of information without an operator and its transmission to a server
or data center for processing and visualization. Integrated monitor-
Logistics (Pang et al., 2015; Li et al., 2013; Jiang and Zhang, 2013;
Charoenpanyasak et al., 2011; Marino et al., 2010) ing tools made it possible to maintain a continuous communica-
tion with the deployed WSN, and access stored data through the
Prediction (Khandani and Kalantari, 2009; Saville et al., 2015; Lee et al.,
2012; Luan et al., 2015) Internet. Hence, smart agriculture based on IoT adds value to farm-
ers by helping them to collect relevant data from crops and farms
using sensor devices. Some IoT setups could display, process and
Selected papers grouped in the monitoring domain dealt with analyze remote data applying cloud services in order to provide
remote sensing of physical and environmental parameters gath- new insights and recommendations for better decision-making.
ered in scenarios such as crops and farms using a Wireless Sensor IoT solutions categorized in monitoring domain can be divided
Network (WSN). The main goal of this domain was the acquisition into three architectural layers (Zou, 2014): (i) a perception layer
288 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297

supported by a WSN; (ii) a network layer where the sensor infor-  Animal monitoring: This subdomain referred to animal track-
mation travels a long distance using different protocols and Gate- ing for both wildlife and animal husbandry activities. A research
ways, and (iii) an application layer that includes a web server and a belonging to this subdomain was a delay-tolerant WSN for the
database. Moreover, IoT solutions grouped in this domain are inter- monitoring and tracking of six horses presented in Ehsan et al.
ested in monitoring several types of physical variables depending (2012). For this purpose, authors developed necklaces that
on the subdomain to which they belong. Specifically, the following acquired information about horses’ position and speed at a
subdomains were identified: air monitoring (34.5%), soil monitor- given time, and transmitted such logs to fixed nodes when they
ing (27.3%), water monitoring (16.4%), plant monitoring (10.9%), were close to its coverage area. Another example of animal
and others (10.9%) which include areas such as aquaculture and monitoring was given by Jain et al. (2008), where an IoT solu-
animal monitoring. It is worth to highlight that most of the tion was responsible for monitoring the behavior and migration
selected studies retrieved in this SLR can be categorized in more patterns of Swamp Deers, obtaining information of the animal
than one subdomain. For instance, the system proposed in Zou position and the climate at the same time.
(2014) is used for online crop growth monitoring and it captures
different types of variables such as: temperature, humidity, soil Papers selected and grouped under the domain of control use
moisture, CO2, luminosity, pH of water, and images. Some repre- remote actuator devices deployed on-site. Unlike monitoring
sentative examples of IoT applications categorized in the monitor- domain applications, which handle information in one-way, appli-
ing domain are described below. cations categorized in control use a two-way information channel.
This means that a new level of communication was added, and
 Air monitoring: this subdomain aimed to provide periodic or commands could be sent back to the field. In this case, information
continuous measurements, evaluating and determining envi- from the server or data center traveled to a Wireless Sensor and
ronmental parameters or pollution levels in order to prevent Actuator Network (WSAN) in order to control a set of actuator
negative and damaging effects. It also included the forecasting devices to modify the state of the process or environment. Com-
of possible changes in the ecosystem or the biosphere as a mands were sent through a human–computer interface or as a
whole. For instance, in Watthanawisuth et al. (2009) authors result of a decision algorithm supported by analytic modules. Actu-
described an agricultural IoT solution which can be categorized ator devices included valves, pumps, humidifiers, and alarms
in the air monitoring subdomain. In this solution, authors pro- among others. Many of these systems aimed to optimize the usage
posed a real-time monitoring system of micro climate based of water, fertilizers, and pesticides based on information provided
on a WSN. The solution included temperature and relative by weather prediction systems and on-site WSN. Solutions in this
humidity sensors (SHT15) powered by solar panels and sup- domain could help farmers to reduce water consumption and
ported by ZigBee communication technology. Another air mon- waste by scheduling irrigation times and quantities according to
itoring IoT solution is GEMS (Lu et al., 2010), which proposed an the state of the crop and its growth cycle. Control systems were
environmental monitoring system based on GPRS technology programmed to be adaptive, for instance, switching off sprinkler
for monitoring apple orchards. This system was tested on five if rain was detected. Overall, solutions with control systems could
different regions of China over a 2-year period by monitoring save money to the farmer and provide at the same time valuable
variables such as relative humidity, temperature, and radiation. insights about the consumption of water, fertilizers, pesticides,
 Soil monitoring: papers classified in this subdomain such as and electricity.
(Chen et al., 2014 and Mafuta et al., 2012) proposed systems Actuator devices used by IoT solutions grouped in the control
for monitoring multi-layer soil temperature and moisture in a domain depended heavily on the subdomain to which they
farmland fields using WSN. These systems are supported by belonged. In this paper, the following subdomains were consid-
communication technologies such as ZigBee, GPRS and Internet, ered: irrigation (72.22%), fertilizers (5.56%), pesticides (5.56%), illu-
where user interaction with the system is handled by a web mination (5.56%), and access control (5.56%). During the review, it
application. was found that some studies used actuators in the domain of logis-
 Water monitoring: primary studies categorized in this subdo- tics (5.56%). Representative examples of IoT applications catego-
main intend to monitor water pollution or water quality by rized in the control domain are described next.
sensing chemicals, pH, and temperature, which can alter the
natural state of water. An example of this subdomain is pre-  Irrigation control: A precision irrigation solution based on
sented in Postolache et al. (2013), where authors proposed wireless sensor network was proposed by Kanoun et al.
an IoT solution for water quality assessment through the mea- (2014). The main challenge of that study was to create an auto-
surement of conductivity, temperature, and turbidity. The mated irrigation system which could reduce water waste, sav-
solution is based on a WSN architecture that combines low- ing energy, time, and money. This system was built using
cost sensing devices and monitoring of multiple parameters three nodes based on the TelosB mote: (i) a node to measure soil
of water quality of shallow waters (lakes, estuaries, rivers) in moisture and soil temperature; (ii) a node to measure environ-
urban areas. Similarly, (Xijun et al., 2009) proposed a WSN mental parameters such as air temperature, air humidity, wind
system for monitoring water level and rainfall in irrigation speed and brightness; and (iii) a node that was connected to a
systems. valve for irrigation control. Data were transmitted to a base sta-
 Plant monitoring: The LOFAR-agro Project (Langendoen et al., tion for storage and were sent to the farmer’s PC to allow him to
2006) is an example of plant or crop monitoring. This project take action. Another precision irrigation IoT system was pro-
aimed to protect a potato crop against phytophthora (a genus posed by Jiao et al. (2014). This included an environmental
of water mold) by monitoring the microclimate (humidity and monitoring system for agricultural management, as well as
temperature) using a large-scale WSN. The system intended to the implementation of precision dripping. The system consid-
generate a policy to protect the crop against the fungal disease ered an IoT ecosystem divided into three layers corresponding
based on the collected data. In Fourati et al. (2014), authors pro- to sensing, transmission, and application. A WSN was used to
pose a Web-based decision support system communicating perceive environmental information in real time within a
with a WSN for irrigation scheduling in olive fields. For this pur- tomato greenhouse, to later transmit the data to a remote server
pose, authors use sensors to measure humidity, solar radiation, management system. In Shuwen and Changli (2015) research-
temperature, and rain. ers described a remote farmland irrigation monitoring solution
 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297 289

based on ZigBee. The system included a solar-powered irriga- in agriculture include: adding value to agricultural products, sav-
tion control system that also monitored air temperature, ing money in distribution costs, improving shipping efficiency,
humidity and soil temperature. reducing unnecessary losses, and to some extent, avoiding risks
 Fertilizer and pesticide control: IoT solutions categorized in (Liping, 2012). Primary studies in logistics were further divided
this subdomain applied conservation practices to improve in: production (55.6%), commerce (22.2%) and transport(22.2%).
nutrient usage, efficiency, crop quality, overall yield, and eco- The next paragraphs include representative studies of each
nomic return while reducing off-site transport of nutrients. In subdomain.
Pahuja et al. (2013), authors developed an online micro-
climate monitoring and control system for greenhouses. The  Production: in Feng et al. (2012) researchers proposed an intel-
system was supported by a WSN to gather and analyze plant- ligent system for monitoring an apple orchard that imple-
related sensor data to produce actions to control the climate, mented suggestions based on data. The system aimed to
fertilization, irrigation, and pests. reduce management costs of apple orchards, improve apple
 Illumination control: authors in Yoo et al. (2007) described an quality, and provide detailed, comprehensive and accurate elec-
automated agriculture system based on WSN for monitoring tronic information for planting works, pest warnings, and
greenhouses used to grow melons and cabbages. The system production-quality tracking of apples. The system included
monitored the growing process of crops and controlled the WSN using Zigbee, GPRS, and IoT providing detailed monitoring
greenhouse’s environment. Some of the variables measured data of apple growth for agricultural cooperatives, to support
included ambient light, temperature, and humidity. For the for decision making in farming.
greenhouse with melons, the system could control the illumina-  Commerce: (Li et al., 2013) presented an information system
tion by changing the light state through a relay. for agriculture based on IoT which used a distributed architec-
 Access control: An agricultural intrusion detection system was ture. In that study, tracking and tracing of the whole agricul-
presented in Roy et al. (2015). The proposed system generated tural production process were made with distributed IoT
alarms in the farmers house and sent a text message to the servers. Moreover, an information-discovery system was
farmer’s mobile phone when an intruder entered the crop field. designed to implement, capture, standardize, manage, locate,
and query business data from agricultural production. The sys-
Selected papers categorized in the prediction domain were tem also allowed consumers to query information of agricul-
focused on providing knowledge and tools to farmers to support tural products to verify their authenticity and quality.
decision making. They had specific modules for these tasks in their  Transport: A representative example of this subdomain is
architecture, and their predicted variables were grouped as fol- presented in Pang et al. (2015), where an IoT architecture
lows: environmental conditions (42.86%), production estimation was proposed for the food-production and commercialization
(42.86%), and crop growth (14.29%). chain. This paper dealt with logistics involved in the trans-
portation of melons from Brazil to Sweden in a journey that
 Environmental conditions: A representative example of envi- takes 46 days. Sensor nodes measured conditions in the envi-
ronmental condition prediction is proposed in Khandani and ronment including oxygen, carbon dioxide, ethylene, temper-
Kalantari (2009), where authors described a design methodol- ature, humidity, and mechanical stress, such as vibrations,
ogy to determine the spatial sampling of humidity sensors for tilts, and shocks.
the soil within a WSN. They used a historical database of dense
soil-humidity measurements to determine the behavior of the Fig. 5 summarizes the distribution of each application domain
2D correlation that exists between the measurements of nearby into its corresponding subdomains described in the previous
sensors. This was used later to find the largest spatial sampling paragraphs.
that ensured a user-defined variance for the estimation on any
given point of interest in the space. Authors found that the spa- 4.2. Answer to the second research question
tial correlation function decays exponentially with the distance
between sensors. Another example of the prediction of environ- Infrastructure and technology used by selected IoT solutions in
mental conditions was presented in Luan et al. (2015), which agro-industrial and environmental fields were organized in seven
described a system that integrates drought monitoring and groups, corresponding to: (i) sensing variables, (ii) actuator
forecasting as well as irrigation prediction using IoT. devices, (iii) power sources, (iv) communication technologies, (v)
 Production estimation: Authors in Lee et al. (2013) presented edge computing technologies (Shi et al., 2016), (vi) storage strate-
an IoT-based agricultural production system for stabilizing sup- gies, and (vii) visualization strategies.
ply and demand of agricultural products. They achieved this
goal by sensing environmental variables and by developing a  Sensing variables: about 26% of analyzed studies sense temper-
prediction system for the growth and yield of crops. In a differ- ature, followed by humidity, physicochemical properties, and
ent application, (Saville et al., 2015) introduced a real-time esti- radiation with 16%, 11%, and 10%, respectively. Particularly,
mation system for fixed-net fishery using ultrasonic sensors and temperature and physicochemical sensors are distributed in
supervised learning. all subdomains as it can be seen in Fig. 6. Similarly, 55% of sen-
 Crop growth: a dynamic analysis of farmlands using mobile sors are used for air monitoring. Thus, air temperature and
sensors was presented in Lee et al. (2012). The developed sys- humidity, soil moisture and solar radiation, can be considered
tem aimed to establish growth-control plans for grapes, and universal variables in agricultural applications.
viticulture activities.  Actuator devices: the distribution of actuators used in selected
studies is shown in Fig. 7. It can be stated that there are far
The last domain used to categorize selected studies was logis- fewer actuator devices than sensors currently being used in
tics. Logistics in agriculture refers to the physical flow of entities these studies and that most of them are concentrated in appli-
and related information from producer to consumer to satisfy con- cations of control and logistics. In fact, more than 60% of actua-
sumer demand. It includes: agricultural production, acquisition, tors reported were found in irrigation processes.
transportation, storage, loading and unloading, handling, packag-
ing, distribution, and related activities. Some objectives of logistics
290 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297

Fig. 5. Distribution of papers selected by application subdomain.

Fig. 6. Types of sensing variables collected in the monitoring domain.

 Power sources: currently, most monitoring applications prefer vibration harvesters were not found in selected studies showing
rechargeable batteries connected to solar panels, which offer a that these approaches must mature and gain popularity for
simple but sustainable energy supply. In contrast, control appli- agricultural and environmental applications.
cations that typically have demanding energy requirements  Communication technologies: Fig. 9 shows that most studies
prefer the electrical grid. These trends can be appreciated in (40%) used Wireless Personal Area Network (WPAN) protocols
Fig. 8. Recent power sources, such as electromagnetic or such as Bluetooth and ZigBee, followed by Wireless
 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297 291

Fig. 7. Type of actuator device used.

Fig. 8. Power sources.

Metropolitan Area Network (WMAN) with 36% of the studies can be stated that web-based solutions were the preferred
mainly supported by cellular technologies (GPRS/GSM/3G/4G). strategy to visualize reports in all subdomains of applications.
Meanwhile, the near-field communication, which is relatively
new, has started to emerge in some field applications. Most of the selected works do not address security issues
 Edge computing technologies: microcontroller platforms were explicitly and leave them on a side. However, some efforts in this
chosen in more than half of the applications reviewed. Interest- domain were found. For instance, (Jardak et al., 2009) described
ingly, Single Board Computers (SBC) are not yet appropriate for the design of a WSN that implemented a RANdom SAmple Consen-
edge computing in IoT agricultural applications. The complete sus (RANSAC) filter to eliminate inconsistent sensor-node data due
distribution of edge computing technologies is shown in Fig. 10. to the presence of faulty or malicious nodes in the network. Sun
 Storage strategies: reviewing Fig. 11, it is clear that even et al. (2012) presented a dam monitoring system where users
though Cloud storage represents a key service for IoT systems, needed to sign in through the main interface in order to validate
only 7.32% of selected studies used it. This shows that most their credentials. Tao et al. (2014) selected AppWeb as the embed-
researchers preferred their own data-storage implementation. ded Web server for the IoT Gateway of an intelligent granary man-
 Visualization strategies: Fig. 12 shows the distribution of three agement system because it could add the Secure Sockets Layer
different visualization strategies: web, mobile and local, in four (SSL) protocol to enable encrypted data connection. This was valu-
subdomains: monitoring, control, prediction, and logistics. It able because the network information was vulnerable as it came
292 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297

Fig. 9. Communication technologies.

Fig. 11. Storage strategy.

from a wireless channel. Kuroda et al. (2015) proposed a WSN with

in a recent survey by Sinha et al. (2017), authors found that LoRa is
easy-to-use secure communication that was implemented using
the best option for smart agriculture applications. In Lukas et al.
Zero-admin encrypt/decrypt functions at the MAC level with the
(2015), authors designed a long-range water level monitoring sys-
Advanced Encryption Standard (AES-128), which enabled auto-
tem for troughs using a WSN based on LoRa transceivers, allowing
matic encryption/decryption of messages between each sensor
the cattleman to observe water availability for livestock even when
node and the coordinator node.
the barn was 1 or 3 km away. In a different application, (Pham
et al., 2016) proposed an IoT framework to contribute to rural
5. Recent works development implementing agricultural applications supported
by open-source hardware and long-range communication devices.
The following paragraphs are devoted to introducing some The first deployment of this solution used LoRa transceivers since
recent and representative works that were available online rural villages were located in remote areas and it was convenient
between May 2016 and July 2017, beyond the initial scope of the to have a low-cost and non-proprietary infrastructure.
SLR process described so far. They cover areas such as communica-
tions, energy management, monitoring and logistics for agro-
5.2. Energy management
industrial and environmental applications.
One of the main requirements for devices used in IoT projects is
5.1. Communications that they must be energy-efficient according to Borgia (2014). This
is particularly important for pervasive solutions deployed outdoors
Low-power WAN (LPWAN) technologies such as SigFox, LoRa, that can not be powered from the electric grid nor regularly main-
narrowband IoT and others are becoming popular within IoT appli- tained because they are installed in difficult or remote environ-
cations due to its reduced energy requirements, wide coverage ments. In WSN scenarios, the current challenge is to develop
range, and low-cost when compared to other long-distance tech- multi-source energy harvesters and ultra-efficient sensors to cre-
nologies according to Barrachina-Muñoz et al. (2017). For example, ate battery-free solutions, (Shaikh and Zeadally, 2016). These con-

Fig. 10. Edge computing technologies.

 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297 293

Fig. 12. Visualization strategies.

siderations are very important for IoT solutions for agro-industrial As an additional result, authors found a strong correlation between
and environmental problems as recharging batteries is not practi- the beehive status and the short-term rain forecast. Overall, this
cal and ambient energy sources are usually available. study is relevant for agriculture because crop pollination depends
In terms of smart energy control for IoT projects, (Wang et al., on honey bees. Sarangi et al. (2016) presented a framework for
2016) proposed a novel energy management strategy for solar an automated crop-disease advisory service that integrates the
powered devices that intend to power the load directly from the interoperability of an IoT web repository with an agricultural advi-
solar cell, avoiding power converters and energy storage elements sory call center. The implemented system processes images of the
that contribute to energy losses, greater weight/volume ratio, and diseased plant sent by the farmer, and then it provides the plant
higher price. Another trend that is likely to continue is the develop- diagnosis and the corresponding management recommendation
ment of self-power devices, such as the soil water content sensor for the disease.
for an autonomous landslide surveillance system designed by Lu
et al. (2016). In this case, the sensor used the soil moisture to 5.4. Logistics
power itself making it suitable for large scale deployments.
Marjanović et al. (2016) described a cloud-based decision-making Food safety and quality control in logistics are emerging as IoT
mechanism for managing sensor data acquisition that is applicable agribusiness areas in response to the demand from businesses and
to collaborative sensing solutions using distributed sensors, like end consumers to obtain real-time information about food supply
mobile devices, to efficiently monitor large geographical areas. chain and ‘‘farm-to-fork” traceability. For instance, (Ruan and Shi,
The system selected which sensors had to upload the information 2016) presented an IoT framework to assess the fruit freshness
to the cloud to prevent the acquisition of redundant information on e-commerce deliveries, which is a non-traditional retail service
from other nearby sensors for a specific coverage area, maintaining that faces unique challenges in transportation due to the product
a spatial sampling quality and reducing in this way the battery perishability and expensive logistics. Similarly, (Liu et al., 2016)
depletion of the devices. introduced a pilot project using IoT to monitor food safety through-
out the product life cycle, helping authorities and consumers to
5.3. Monitoring trace the food and make better decisions before buying it. In a
related work, (Wang and Yue, 2017) proposed an early-warning
Recent environmental monitoring solutions are now offering system for food safety that automatically warns about product
additional capabilities in terms of decision making and manage- quality risks and incidents by sharing and centralizing information
ment. For example, (Giorgetti et al., 2016) proposed a custom- among supply chains. Lastly, (Capello et al., 2016) developed a
made landslide risk monitoring system based on a WSN that allows business-to-business monitoring service based on IoT that pro-
fast deployments in hostile environments without human inter- vides geo-located information (humidity and temperature) about
vention because the system is able to deal with node failure and food storage and transportation without a vendor lock-in
poor-quality communication links reorganizing the network by infrastructure.
itself. Wong and Kerkez (2016) presented a Web service and
real-time data architecture that includes an adaptive controller 6. Discussion
that updates the parameters of each sensing node within a WSN
based on a previously defined policy. Zheng et al. (2016) proposed 6.1. Limitations and open challenges
an IoT management system to protect the ecological and environ-
mental quality while building an artificial river where nature and After analyzing the difficulties and limitations described in
city converge. The system monitored key elements like soil, water, selected papers from the SLR, the following list summarizes a
atmosphere, and wind at a high spatial resolution over a large area. few insights that aim to contribute to the mass adoption of IoT
Edwards-Murphy et al. (2016) introduced a beehive monitoring solutions in agricultural and environmental fields.
system that collects internal and external data to describe the sta-
tus of the bee colony from a set of possible states using a classifi-  Stronger standardization: it will help to improve compatibility
cation algorithm based on decision trees. This information was among different vendors and to ensure stronger security mea-
used to determine if a visit to the beehive was required or not. sures across the entire IoT stack, starting from field devices all
294 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297

the way up to cloud providers and end-user interfaces (Pang not true for high-quality sensors and actuators. In order to
et al., 2015). deploy IoT solutions with hundreds and possibly thousands
 Better power management: it will increase the endurance of of nodes, the overall hardware, Internet access and interna-
IoT solutions because nowadays the main factor limiting the tional data roaming costs have to be reduced even further
lifespan of IoT deployments is energy depletion (Jain et al., (Pang et al., 2015).
2008; Chen et al., 2014; Islam et al., 2014; Diedrichs et al.,  Aim for a good compatibility with legacy infrastructure: sim-
2014). The lifespan can be improved by lowering the power ilarly to what has happened in industrial automation, it is
consumption of each electronic module, including energy har- important to deliver IoT solutions that can be integrated with
vesters, and using alternative power storage mechanisms as the customer’s existing infrastructure such as specialized equip-
replacements of rechargeable batteries, which affect the expira- ment, field machines, and software.
tion date of deployed devices.  Consider scalability early on: with an increasing number of
 Security: a major challenge in the realization of the IoT in agri- devices in large deployments, data synchronization and data
culture is the security problem (Jiang and Zhang, 2013), and the reliability become critical (Diedrichs et al., 2014).
few works that consider it only incorporate fragmented strate-  Adopt good practices of software engineering: as the scale
gies to mitigate it. Therefore, it is evident that there is a need for and endurance of deployed IoT solutions grow, the time and
agro-industrial and environmental IoT solutions that address effort devoted to analyzing generated data, refining the code,
end-to-end information security and physical integrity of field and adding new features will explode unless the software is
devices. well designed and documented (Hussain et al., 2006;
 Design using modular hardware and software: it will enable a Jayaraman et al., 2015a).
greater degree of reuse and customization for the end user  Improve robustness for field deployments: commercial IoT
(Pang et al., 2015). solution should be able to handle strong changes in tempera-
 Improve unit cost: even though the cost of embedded com- ture, humidity, and illumination to deal with seasonal changes
puting platforms have been decreasing sharply, the same is and worldwide climate variability.

Monitoring Control Prediction Logistics

Application
Layer

Communication Storage Analytics Visualization Security Service

Services Layer
GPRS 3G 4G 5G
NB-IoT Field
IoT
Gateway
Gateway

Communication
Internet Ethernet Layer
NFC

Field
WSN Gateway
Weather
Control

Station Illumination Irrigation

Sensing

Pesticide Harvesting Fertilization

Solar Panel RFID
Physical
Robot Layer

Soil Moisture Temperature Humidity Electrical Grid

Perception Layer Control Layer

Fig. 13. Proposed IoT architecture for agro-industrial and environmental applications.
 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297 295

 User-centered design: the installation and management of cor- The temperature and humidity of the air, as well as the soil
responding IoT nodes should be straight forward for non-expert moisture and solar radiation can be recognized as universal vari-
users. Additionally, the hardware must require very little or ables measured in agricultural applications based on selected stud-
none human maintenance during its lifespan, and the underly- ies. Similarly, actuators such as valves, pumps, motors, sprinklers,
ing communication network should be intelligent enough to humidifiers, and lamps were widely used in irrigation, fertilization,
reconfigure or heal itself in the case of a node failure. pesticide management, and illumination control. It was also
 Contribute to the IoT the ecosystem: there is a noticeable void observed that new energy sources and Cloud storage have not been
in the literature on how to improve and adapt IoT solutions for widely adopted, showing that there are opportunities for research
real-world applications beyond simple prototypes (Chen et al., and development in these areas.
2015). Studies included in this paper provide a compact view of solu-
 Sustainable practices: even if the most humble predictions tions proposed for agro-industrial and environmental problems
about the worldwide adoption of IoT devices become a reality, during the last decade. It was found that most of them relied heav-
recycling strategies will have to be taken into account for new ily on heterogeneous components and wireless sensor networks.
solutions deployed on the field, as an integral part of the pro- However, it seems reasonable to assume that future solutions will
duct life cycle to reduce the environmental impact. need to fully embrace Cloud services and new ways of connectivity
in order to get the benefits of a truly connected and smart IoT
ecosystem.
6.2. Proposed architecture

To summarize the findings of this study, authors proposed the Acknowledgements

IoT architecture for agro-industrial and environmental applications
that is illustrated in Fig. 13. This encapsulates most of the studies Authors would like to acknowledge the support of all partners
analyzed in this paper. The architecture has four main layers: phys- within the Center of Excellence and Appropriation on the Internet
ical, communication, service, and application. The physical layer of Things (CEA-IoT), as well the Colombian Ministry for the Infor-
includes perception and control. In perception, the main objective mation and Communication Technologies (MinTIC), and the
is to produce valuable data sensing field variables using a WSN. Colombian Administrative Department of Science, Technology
Data produced are sent to the communication layer through field and Innovation (Colciencias) through the project ID: FP44842-
gateways. Devices in the perception layer can be powered by bat- 502-2015 from the National Trust for Funding Science, Technology
teries for short-term deployments or by solar panels because of and Innovation Francisco José de Caldas.
their low-power consumption. In contrast, the control layer acts
as a data sink, receiving information from a communication layer
References
or a perception layer in the simplest case. Information received
in the control layer alters the state of field actuators frequently Barrachina-Muñoz, S., Bellalta, B., Adame, T., Bel, A., 2017. Multi-hop
requiring power from the electrical grid. In the middle of the per- communication in the uplink for LPWANs. Comput. Netw. 123, 153–168.
ception and control layers there is a mobile robot that can be used http://dx.doi.org/10.1016/j.comnet.2017.05.020.
Borgia, E., 2014. The internet of things vision: key features, applications and open
when fixed devices are not the best option. In the communication issues. Comput. Commun. 54, 1–31. http://dx.doi.org/10.1016/
layer, the objective is to move the information from the physical j.comcom.2014.09.008.
layer to the Internet, collecting data from IoT gateways based Capello, F., Toja, M., Trapani, N., 2016. A real-time monitoring service based on
industrial internet of things to manage agrifood logistics. In: 6th International
either on Ethernet or mobile networks (e.g: GPRS/3G/4G/NB-IoT
Conference on Information Systems, Logistics and Supply Chain, pp. 1–8.
and eventually 5G). This layer includes field gateways acting as Charoenpanyasak, S., Suntiamorntut, W., Phatthanatraiwat, T., Ruksachum, J., 2011.
interfaces between IoT gateways and transceivers using ZigBee, Smart shrimp hatchery using mikros platform. In: 4th Joint IFIP Wireless and
Bluetooth, NFC, WiFi, LoRA, or Sigfox. The service layer handles Mobile Networking Conference (WMNC). IEEE, pp. 1–5.
Chavez-Burbano, P., Marin-Garcia, I., Muñoz-Arcentales, A., 2014. Ad-hoc network
data ingestion from the communication layer, as well as their stor- implementation and experimental testing using low cost and COTS
age, analytics, visualization, and security. Finally, the application components: an ecuatorian case study. In: International Work Conference on
layer consumes services from the previous layer in the architecture Bio-inspired Intelligence (IWOBI). IEEE, pp. 133–137.
Chen, K.T., Zhang, H.H., Wu, T.T., Hu, J., Zhai, C.Y., Wang, D., 2014. Design of
and allows the user to handle monitoring, control, prediction, and monitoring system for multilayer soil temperature and moisture based on WSN.
logistics. In: International Conference on Wireless Communication and Sensor Network
(WCSN). IEEE, Wuhan, pp. 425–430. http://dx.doi.org/10.1109/WCSN.2014.9.
Chen, Y., Chanet, J.-P., Hou, K.-M., Shi, H., de Sousa, G., 2015. A scalable context-
aware objective function (SCAOF) of routing protocol for agricultural low-
7. Conclusions power and lossy networks (RPAL). Sensors 15, 19507–19540. http://dx.doi.org/
10.3390/s150819507.
Culibrina, F.B., Dadios, E.P., 2015. Smart farm using wireless sensor network for data
This paper presented an updated review of IoT applications for acquisition and power control distribution. International Conference on
agro-industrial and environmental fields. It was guided by a sys- Humanoid, Nanotechnology, Information Technology, Communication and
tematic literature review, and therefore the methodology and Control, Environment and Management (HNICEM). IEEE, pp. 1–6.
De La Concepcion, A.R., Stefanelli, R., Trinchero, D., 2014. A wireless sensor network
intermediate results obtained during the stages of planning, con- platform optimized for assisted sustainable agriculture. In: Global
duction, and results were reported in great detail. From 3578 initial Humanitarian Technology Conference (GHTC). IEEE, pp. 159–165. http://dx.
studies extracted from electronic sources, 72 main studies were doi.org/10.1109/GHTC.2014.697027.
Diedrichs, A.L., Tabacchi, G., Grünwaldt, G., Pecchia, M., Mercado, G., Antivilo, F.G.,
selected based on their relevance to answer two research 2014. Low-power wireless sensor network for frost monitoring in agriculture
questions. research. In: Biennial Congress of Argentina (ARGENCON). IEEE, pp. 525–530.
Selected studies came from five continents, and Asian countries http://dx.doi.org/10.1109/ARGENCON.2014.686854.
Edwards-Murphy, F., Magno, M., Whelan, P.M., O’Halloran, J., Popovici, E.M., 2016. b
contributed to more than half of them. During this study, it was
+WSN: smart beehive with preliminary decision tree analysis for agriculture
discovered that most of the research still focuses on monitoring and honey bee health monitoring. Comput. Electron. Agric. 124, 211–219.
applications (62%); however there is a growing interest in closing http://dx.doi.org/10.1016/j.compag.2016.04.008.
the loop by doing control (25%), and there are some preliminary Ehsan, S., Bradford, K., Brugger, M., Hamdaoui, B., Kovchegov, Y., Johnson, D.,
Louhaichi, M., 2012. Design and analysis of delay-tolerant sensor networks for
solutions in logistics and prediction (13%) for agro-industrial and monitoring and tracking free-roaming animals. IEEE Trans. Wireless Commun.
environmental applications using IoT. 11, 1220–1227. http://dx.doi.org/10.1109/TWC.2012.012412.111405.
296 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297

Eom, K.-H., Hyun, K.-H., Lin, S., Kim, J.-W., 2014. The meat freshness monitoring japanese consumer under sensor Asia. In: SICE Annual Conference, pp. 2582–
system using the smart RFID tag. Int. J. Distrib. Sensor Networks 2014, 1–10. 2585. http://dx.doi.org/10.1109/SICE.2008.4655101.
Fang, S., Da Xu, L., Zhu, Y., Ahati, J., Pei, H., Yan, J., Liu, Z., 2014. An integrated system Kodali, R.K., Rawat, N., Boppana, L., 2014. WSN sensors for precision agriculture. In:
for regional environmental monitoring and management based on internet of Region 10 Symposium. IEEE, pp. 651–656. http://dx.doi.org/10.1109/
things. IEEE Trans. Ind. Inform. 10, 1596–1605. TENCONSpring.2014.686311.
Feng, C., Wu, H.R., Zhu, H.J., Sun, X., 2012. The design and realization of apple Kuroda, M., Ibayashi, H., Mineno, H., 2015. Affordable 400 MHz long-haul sensor
orchard intelligent monitoring system based on internet of things technology. network for greenhouse horticulture. In: International Conference on
Advanced Materials Research, vol. 546. Trans Tech Publ, pp. 898–902. Information Networking (ICOIN). IEEE, Cambodia, pp. 19–24. http://dx.doi.org/
Fourati, M.A., Chebbi, W., Kamoun, A., 2014. Development of a web-based weather 10.1109/ICOIN.2015.705785.
station for irrigation scheduling. In: 3rd International Colloquium in Langendoen, K., Baggio, A., Visser, O., 2006. Murphy loves potatoes experiences from
Information Science and Technology (CIST). IEEE, pp. 37–42. http://dx.doi.org/ a pilot sensor network deployment in precision agriculture. 20th International
10.1109/CIST.2014.701659. Parallel and Distributed Processing Symposium (IPDPS), vol. 2006. IEEE, Rhodes
Giorgetti, A., Lucchi, M., Tavelli, E., Barla, M., Gigli, G., Casagli, N., Chiani, M., Dardari, Island, pp. 1530–2075. http://dx.doi.org/10.1109/IPDPS.2006.163941.
D., 2016. A robust wireless sensor network for landslide risk analysis: system Lee, J., Kang, H., Bang, H., 2012. Dynamic crop field analysis using mobile sensor
design, deployment, and field testing. IEEE Sens. J. 16, 6374–6386. http://dx.doi. node. In: International Conference on ICT Convergence (ICTC), pp. 7-11. http://
org/10.1109/JSEN.2016.2579263. dx.doi.org/10.1109/ICTC.2012.6386766.
Gutiérrez, J., Villa-Medina, J.F., Nieto-Garibay, A., Porta-Gándara, M.Á., 2014. Lee, M., Hwang, J., Yoe, H., 2013. Agricultural production system based on IoT. In:
automated irrigation system using a wireless sensor network and GPRS 16th International Conference on Computational Science and Engineering (CSE).
module. IEEE Trans. Instrum. Meas. 63, 166–176. IEEE, pp. 833–837.
Hachem, S., Mallet, V., Ventura, R., Pathak, A., Issarny, V., Raverdy, P.-G., Bhatia, R., Li, M., Chen, G., Zhu, Z., 2013. Information service system of agriculture IoT.
2015. Monitoring noise pollution using the urban civics middleware. In: First Automatika - J. Control, Meas. Electron. Comput. Commun. 54, 415–426.
International Conference on Big Data Computing Service and Applications. IEEE, Li, R.-A., Sha, X., Lin, K., 2014. Smart greenhouse: a real-time mobile intelligent
pp. 52–61. monitoring system based on WSN. In: International Wireless Communications
Hakala, I., Tikkakoski, M., Kivel, I., 2008. Wireless sensor network in environmental and Mobile Computing Conference (IWCMC). IEEE, pp. 1152–1156.
monitoring - case foxhouse. In: 2nd International Conference on Sensor Liping, W., 2012. Study on agricultural products logistics mode in Henan Province of
Technologies and Applications (SENSORCOMM), pp. 202–208. http://dx.doi. China. In: Software Eng. Knowledge Eng.: Theory Practice. Springer, pp. 635–
org/10.1109/SENSORCOMM.2008.27. 640.
Hashim, N., Mazlan, S., Aziz, M.A., Salleh, A., Ja’afar, A., Mohamad, N., 2015. Liu, Y., Han, W., Zhang, Y., Li, L., Wang, J., Zheng, L., 2016. An internet-of-things
Agriculture monitoring system: a study. J. Teknologi 77, 53–59. http://dx.doi. solution for food safety and quality control: a pilot project in China. J. Ind.
org/10.11113/jt.v77.4099. Inform. Integrat. 3, 1–7. http://dx.doi.org/10.1016/j.jii.2016.06.001.
Hussain, S., Schofield, N., Matin, A.W. 2006. Design of a web-based application for Liu, Z., Huang, J., Wang, Q., Wang, Y., Fu, J., 2013. Real-time barrier lakes monitoring
wireless sensor networks. In: 17th International Workshop on Database and and warning system based on wireless sensor network. In: International
Expert Systems Applications (DEXA), pp. 319–326. http://dx.doi.org/10.1109/ Conference on Intelligent Control and Information Processing (ICICIP). IEEE,
DEXA.2006.50. Beijing, pp. 551–554. http://dx.doi.org/10.1109/ICICIP.2013.656813.
Islam, A., Islam, T., Syrus, M.A., Ahmed, N., 2014. Implementation of flash flood Lu, S., Duan, M., Zhao, P., Lang, Y., Huang, X., 2010. GPRS-based environment
monitoring system based on wireless sensor network in Bangladesh. 3rd monitoring system and its application in apple production. International
International Conference on Informatics, Electronics & Vision. IEEE, Dhaka, pp. Conference on Progress in Informatics and Computing (PIC), vol. 1. IEEE, pp.
1–6. http://dx.doi.org/10.1109/ICIEV.2014.685075. 486–490. http://dx.doi.org/10.1109/PIC.2010.568757.
Jain, V.R., Bagree, R., Kumar, A., Ranjan, P., 2008. wildCENSE: GPS based animal Lu, T.-C., Huang, L.-R., Lee, Y., Tsai, K.-J., Liao, Y.-T., Cheng, N.-C., Chu, Y.-H., Tsai, Y.-
tracking system. In: International Conference on Intelligent Sensors, Sensor H., Chen, F.-C., Chiueh, T.-C., 2016. Invited – wireless sensor nodes for
Networks and Information Processing (ISSNIP), pp. 617–622. http://dx.doi.org/ environmental monitoring in internet of things. In: 53rd Design Automation
10.1109/ISSNIP.2008.4762058. Conference (DAC). ACM, pp. 1–5. http://dx.doi.org/10.1145/2897937.289860.
Jardak, C., Rerkrai, K., Kovacevic, A., Riihijarvi, J., Mahonen, P., 2009. Email from the Luan, Q., Fang, X., Ye, C., Liu, Y., 2015. An integrated service system for agricultural
vineyard. In: 5th International Conference on Testbeds and Research drought monitoring and forecasting and irrigation amount forecasting. In: 23rd
Infrastructures for the Development of Networks & Communities and International Conference on Geoinformatics. IEEE, pp. 1–7.
Workshops (TridentCom). IEEE, pp. 1–6. http://dx.doi.org/10.1109/ Lukas, Tanumihardja, W.A., Gunawan, E., 2015. On the application of IoT:
TRIDENTCOM.2009.497624. monitoring of troughs water level using WSN. In: Conference on Wireless
Jayaraman, P.P., Palmer, D., Zaslavsky, A., Georgakopoulos, D., 2015a. Do-it-yourself Sensors (ICWiSe). IEEE, pp. 58–62. http://dx.doi.org/10.1109/
digital agriculture applications with semantically enhanced IoT platform. 10th ICWISE.2015.738035.
International Conference on Intelligent Sensors, Sensor Networks and Ma, D., Ding, Q., Li, Z., Li, D., Wei, Y., 2012. Prototype of an aquacultural information
Information Processing (ISSNIP). IEEE, pp. 1–6. system based on internet of things E-Nose. Intell. Automat. Soft Comput. 18,
Jayaraman, P.P., Palmer, D., Zaslavsky, A., Salehi, A., Georgakopoulos, D., 2015b. 569–579.
Addressing information processing needs of digital agriculture with OpenIoT Mafuta, M., Zennaro, M., Bagula, A., Ault, G., Gombachika, H., Chadza, T., 2012.
platform. In: Interoperability and Open-Source Solutions for the Internet of Successful Deployment of a Wireless Sensor Network for Precision Agriculture
Things. Springer, pp. 137–152. in Malawi. In: 3rd International Conference on Networked Embedded Systems
Jiang, R., Zhang, Y., 2013. Research of agricultural information service platform for Every Application (NESEA). IEEE, pp. 1–7.
based on internet of things. In: 12th International Symposium on Distributed Marino, P., Fontán, F.P., Domínguez, M.Á., Otero, S., 2010. An experimental Ad-hoc
Computing and Applications to Business, Engineering Science (DCABES), pp. WSN for the instrumentation of biological models. IEEE Trans. Instrum. Meas.
176–180. http://dx.doi.org/10.1109/DCABES.2013.39. 59, 2936–2948.
Jiao, J., Ma, H., Qiao, Y., Du, Y., Kong, W., Wu, Z., 2014. Design of farm environmental Marjanović, M., Skorin-Kapov, L., Pripužić, K., Antonić, A., Podnar Žarko, I., 2016.
monitoring system based on the internet of things. Adv. J. Food Sci. Technol. 6, Energy-aware and quality-driven sensor management for green mobile crowd
368–373. sensing. J. Network Comput. Appl. 59, 95–108. http://dx.doi.org/10.1016/j.
Jiber, Y., Harroud, H., Karmouch, A., 2011. Precision agriculture monitoring jnca.2015.06.023.
framework based on WSN. In: 7th International Wireless Communications Mathurkar, S.S., Patel, N.R., Lanjewar, R.B., Somkuwar, R.S., 2014. Smart sensors
and Mobile Computing Conference, pp. 2015–2020. http://dx.doi.org/10.1109/ based monitoring system for agriculture using field programmable gate array.
IWCMC.2011.5982844. In: International Conference on Circuit, Power and Computing Technologies
Kaewmard, N., Saiyod, S., 2014. Sensor data collection and irrigation control on (ICCPCT). IEEE, pp. 339–344.
vegetable crop using smart phone and wireless sensor networks for smart farm. Medela, A., Cendón, B., González, L., Crespo, R., Nevares, I., 2013. IoT Multiplatform
In: Conference on Wireless Sensors (ICWiSE). IEEE, pp. 106–112. networking to monitor and control wineries and vineyards. In: Future Network
Kanoun, O., Khriji, S., El Houssaini, D., Viehweger, C., Jmal, M.W., Abid, M., 2014. and Mobile Summit. IEEE, pp. 1–10.
Precision irrigation based on wireless sensor network. IET Sci. Meas. Technol. 8, Mittal, A., Chetan, K.P., Jayaraman, S., Jagyasi, B.G., Pande, A., Balamuralidhar, P.,
98–106. http://dx.doi.org/10.1049/iet-smt.2013.0137. 2012. mKRISHI wireless sensor network platform for precision agriculture. In:
Kar, A., Kar, A., 2015. A novel design of a portable double beam-in-time 6th International Conference on Sensing Technology (ICST). IEEE, pp. 623–629.
spectrometric sensor platform with cloud connectivity for environmental http://dx.doi.org/10.1109/ICSensT.2012.646175.
monitoring applications. In: 3rd International Conference on Computer, Nguyen, T.-D., Thanh, T.T., Nguyen, L.-L., Huynh, H.-T., 2015. On the design of energy
Communication, Control and Information Technology (C3IT), pp. 1–6. http:// efficient environment monitoring station and data collection network based on
dx.doi.org/10.1109/C3IT.2015.7060228. ubiquitous wireless sensor networks. In: International Conference on
Khandani, S.K., Kalantari, M., 2009. Using field data to design a sensor network. In: Computing & Communication Technologies-Research, Innovation, and Vision
43rd Annual Conference on Information Sciences and Systems (CISS), pp. 219– for the Future (RIVF). IEEE, pp. 163–168.
223. http://dx.doi.org/10.1109/CISS.2009.5054720. Pahuja, R., Verma, H., Uddin, M., 2013. A wireless sensor network for greenhouse
Kitchenham, B., Charters, S., 2007. Guidelines for performing systematic literature climate control. IEEE Pervasive Comput. 12, 49–58.
reviews in software engineering. In: EBSE Technical Report. EBSE-2007-01. pp. Pang, Z., Chen, Q., Han, W., Zheng, L., 2015. Value-centric design of the internet-of-
1–50. things solution for food supply Chain: value creation, sensor portfolio and
Kiyoshi, H., Shrestha, A., Chinnachodteeranun, R., Mizoguchi, M., Shimamura, H., information fusion. Inform. Syst. Front. 17, 289–319. http://dx.doi.org/10.1007/
Kameoka, T., 2008. Spinach field monitoring for bridging thai producer and s10796-012-9374-9.
 J.M. Talavera et al. / Computers and Electronics in Agriculture 142 (2017) 283–297 297

Pham, C., Rahim, A., Cousin, P., 2016. Low-cost, long-range open IoT for smarter Sun, E., Zhang, X., Li, Z., 2012. The internet of things (IOT) and cloud computing (CC)
Rural African villages. In: International Smart Cities Conference (ISC2). IEEE, pp. based tailings dam monitoring and pre-alarm system in mines. Safety Sci. 50,
1–6. http://dx.doi.org/10.1109/ISC2.2016.758082. 811–815. http://dx.doi.org/10.1016/j.ssci.2011.08.028.
Pokrić, B., Krčo, S., Drajić, D., Pokrić, M., Jokić, I., Stojanović, M.J., 2014. ekoNET - Tao, R., Yang, S., Tan, W., Zhang, C., 2014. Secure gateway of internet of things based
environmental monitoring using low-cost sensors for detecting gases, on AppWeb and secure sockets layer for intelligent granary management
particulate matter, and meteorological parameters. In: Eighth International system. In: International Conference on Computer and Computing Technologies
Conference on Innovative Mobile and Internet Services in Ubiquitous in Agriculture. Springer, pp. 78–89.
Computing (IMIS), pp. 421–426. http://dx.doi.org/10.1109/IMIS.2014.57. Tarange, P.H., Mevekari, R.G., Shinde, P.A., 2015. Web based automatic irrigation
Postolache, O., Pereira, J.D., Girão, P.S., 2014. Wireless sensor network-based system using wireless sensor network and embedded linux board. In:
solution for environmental monitoring: water quality assessment case study. International Conference on Circuit, Power and Computing Technologies
IET Sci., Meas. Technol. 8, 610–616. http://dx.doi.org/10.1049/iet- (ICCPCT), pp. 1–5. http://dx.doi.org/10.1109/ICCPCT.2015.7159327.
smt.2013.0136. Torres-Ruiz, M., Juárez-Hipólito, J.H., Lytras, M.D., Moreno-Ibarra, M., 2016.
Postolache, O., Pereira, M., Gir ao, P., 2013. Sensor network for environment Environmental noise sensing approach based on volunteered geographic
monitoring: water quality case study. In: 4th Symposium on Environmental information and spatio-temporal analysis with machine learning. In:
Instrumentation and Measurements, pp. 30–34. International Conference on Computational Science and Its Applications.
Roy, S.K., Roy, A., Misra, S., Raghuwanshi, N.S., Obaidat, M.S., 2015. AID: A prototype Springer, pp. 95–110.
for agricultural intrusion detection using wireless sensor network. In: Vo, T.T., Nguyen, T.D., Vo, M.T., 2013. Ubiquitous sensor network for development of
International Conference on Communications (ICC). IEEE, pp. 7059–7064. climate change monitoring system based on solar power supply. In:
Ruan, J., Shi, Y., 2016. Monitoring and assessing fruit freshness in IoT-Based E- International Conference on Advanced Technologies for Communications, pp.
commerce delivery using scenario analysis and interval number approaches. 121–124. http://dx.doi.org/10.1109/ATC.2013.6698090.
Inf. Sci. 373, 557–570. http://dx.doi.org/10.1016/j.ins.2016.07.014. Wang, J., Yue, H., 2017. Food safety pre-warning system based on data mining for a
Ryu, M., Yun, J., Miao, T., Ahn, I.-Y., Choi, S.-C., Kim, J., 2015. Design and sustainable food supply Chain. Food Control 73, 223–229. http://dx.doi.org/
implementation of a connected farm for smart farming system. In: In Sensors. 10.1016/j.foodcont.2016.09.048.
IEE, pp. 1–4. Wang, Y., Liu, Y., Wang, C., Li, Z., Sheng, X., Lee, H.G., Chang, N., Yang, H., 2016.
Sales, N., Remédios, O., Arsenio, A., 2015. Wireless sensor and actuator system for Storage-less and converter-less photovoltaic energy harvesting with maximum
smart irrigation on the cloud. In: 2nd World Forum on Internet of Things (WF- power point tracking for internet of things. IEEE Trans. Comput. Aided Des.
IoT). IEEE, pp. 693–698. Integr. Circuits Syst. 35, 173–186. http://dx.doi.org/10.1109/
Sarangi, S., Umadikar, J., Kar, S., 2016. Automation of agriculture support systems TCAD.2015.2446937.
using Wisekar: case study of a crop-disease advisory service. Comput. Electron. Watthanawisuth, N., Tuantranont, A., Kerdcharoen, T., 2009. Microclimate real-time
Agric. 122, 200–210. http://dx.doi.org/10.1016/j.compag.2016.01.009. monitoring based on zigbee sensor network. In: Sensors. IEEE, pp. 1814–1818.
Saville, R., Hatanaka, K., Wada, M., 2015. ICT application of real-time monitoring Wong, B.P., Kerkez, B., 2016. Real-time environmental sensor data: an application to
and estimation system for set-net fishery. In: OCEANS, pp. 1–5. water quality using web services. Environ. Modell. Software 84, 505–517.
Sawant, S.A., Adinarayana, J., Durbha, S.S., 2014. KrishiSense: a semantically aware http://dx.doi.org/10.1016/j.envsoft.2016.07.020.
web enabled wireless sensor network system for precision agriculture Xijun, Y., Limei, L., Lizhong, X., 2009. The application of wireless sensor network in
applications. In: Geoscience and Remote Sensing Symposium. IEEE, pp. 4090– the irrigation area automatic system. In: International Conference on Networks
4093. http://dx.doi.org/10.1109/IGARSS.2014.694738. Security, Wireless Communications and Trusted Computing (NSWCTC), vol. 1.
Shaikh, F.K., Zeadally, S., 2016. Energy harvesting in wireless sensor networks: a IEEE, pp. 21–24.
comprehensive review. Renew. Sustain. Energy Rev. 55, 1041–1054. http://dx. Xu, J., Zhang, J., Zheng, X., Wei, X., Han, J., 2015. Wireless sensors in farmland
doi.org/10.1016/j.rser.2015.11.010. environmental monitoring. In:International Conference on Cyber-Enabled
Shi, W., Cao, J., Zhang, Q., Li, Y., Xu, L., 2016. Edge computing: vision and challenges. Distributed Computing and Knowledge Discovery, pp. 372–379. http://dx.doi.
IEEE Internet Things J. 3, 637–646. http://dx.doi.org/10.1109/ org/10.1109/CyberC.2015.17.
JIOT.2016.2579198. Ye, J., Chen, B., Liu, Q., Fang, Y., 2013. A precision agriculture management system
Shuwen, W., Changli, Z., 2015. Study on farmland irrigation remote monitoring based on internet of things and WebGIS. In: 21st International Conference on
system based on ZigBee. In: International Conference on Computer and Geoinformatics, pp. 1–5. http://dx.doi.org/10.1109/Geoinformatics.2013.
Computational Sciences (ICCCS). IEEE, pp. 193–197. 6626173.
Sinha, A., Shen, Z., Song, Y., Ma, H., Darrin Eide, B.-J.P.H., Wang, K., 2015a. An Yoo, S.E., Kim, J.E., Kim, T., Ahn, S., Sung, J., Kim, D., (2007). A2S automated
overview of microsoft academic service (MAS) and applications. In: 24th agriculture system based on WSN. In: IEEE International Symposium on
International Conference on World Wide Web. ACM, pp. 243–246. Consumer Electronics, pp. 1–5. http://dx.doi.org/10.1109/ISCE.2007.4382216.
Sinha, N., Pujitha, K.E., Alex, J.S.R., 2015b. Xively Based sensing and monitoring Zhao, L., Zhu, X., 2015. The development of remote monitoring system for
system for IoT. In: International Conference on Computer Communication and cultivation environment of pleurotus eryngii. In: International Conference on
Informatics (ICCCI), pp. 1–6. http://dx.doi.org/10.1109/ICCCI.2015.7218144. Information and Automation. IEEE, pp. 2643–2648.
Sinha, R.S., Wei, Y., Hwang, S.-H., 2017. A Survey on LPWA technology: LoRa and NB- Zheng, R., Zhang, T., Liu, Z., Wang, H., 2016. An EIoT system designed for ecological
IoT. ICT Express 3, 14–21. http://dx.doi.org/10.1016/j.icte.2017.03.004. and environmental management of the Xianghe segment of china’s grand canal.
Smarsly, K., 2013. Agricultural ecosystem monitoring based on autonomous sensor Int. J. Sustain. Dev. World Ecol. 23, 372–380. http://dx.doi.org/10.1080/
systems. In: 2nd International Conference on Agro-Geoinformatics (Agro- 13504509.2015.1124470.
Geoinformatics). IEEE, pp. 402-407. Zou, C.-J., 2014. Research and implementation of agricultural environment
Soontranon, N., Tangpattanakul, P., Srestasathiern, P., Rakwatin, P., 2014. An monitoring based on internet of things. In: 5th International Conference on
agricultural monitoring system: field server data collection and analysis on Intelligent Systems Design and Engineering Applications (ISDEA). IEEE, pp. 748–
paddy field. In: 14th International Symposium on Communications and 752. http://dx.doi.org/10.1109/ISDEA.2014.17.
Information Technologies (ISCIT). IEEE, pp. 597–601.