See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/377766728

Integration of Wireless Sensor Networks, Internet of Things, Artificial

Intelligence, and Deep Learning in Smart Agriculture: A Comprehensive
Survey

Article · January 2024

CITATIONS READS

6 186

5 authors, including:

Mushtaque Ahmed Rahu Sarang Karim

Quaid-e-Awam University of Engineering, Science and Technology Nawab Shah. Quaid-e-Awam University of Engineering, Science and Technology
17 PUBLICATIONS 118 CITATIONS 28 PUBLICATIONS 341 CITATIONS

SEE PROFILE SEE PROFILE

Sayed Mazhar Ali Shah

Mehran University of Engineering and Technology SZAB Campus
14 PUBLICATIONS 54 CITATIONS

SEE PROFILE

All content following this page was uploaded by Sayed Mazhar Ali Shah on 29 January 2024.

The user has requested enhancement of the downloaded file.

 Original Article

Available online at
https://jiicet.gnt.com.pk/index.php/jiicet/

Journal of Innovative Intelligent Computing and

Emerging Technologies (JIICET)
Vol. 1, No. 1, 2024

Integration of Wireless Sensor Networks, Internet of Things, Artificial

Intelligence, and Deep Learning in Smart Agriculture: A Comprehensive
Survey
Mushtaque Ahmed Rahu1, Sarang Karim2, Sayed Mazhar Ali3, Ghullam Murtaza Jatoi1, Najamu Din Sohu5
1
Department of Electronic Engineering, QUEST, Nawabshah 67450, Pakistan, email: rahumushtaque@gmail.com (M.A.R) (G.M.J)
2
Department of Telecommunication Engineering, QUEST, Nawabshah 67450, Pakistan, email sarangkarim@hotmail.com (S.K)
3
Department of Electrical Engineering MUET SZAB Khairpur Mirs. email mazharlakyari@gmail.com (S.M.A)
4
Government College University Hyderabad. email najam_sohu@yahoo.com (N.S)
* Correspondence email: rahumushtaque@gmail.com

Article History Abstract

Received 01 November 2023 This survey explores th e synergistic integration of Wi reless Sensor
Revised 21 December 2023 Networks (WSNs), Internet of Th ings (IoT) , Artifici al Intelligence (AI), and
Accepted 31 December 2023 Deep Learning (DL) in the realm of smart agricultu re (SA). The agricultu ral
Available Online 28 JAN 2024 sector is u ndergoing a transformative paradigm shift , leveraging advanced
technologies to enhance efficiency, productivity, and sustainab ility. WSNs
Keywords: serve as the backbone, fac ilitating real -time data acq uisition from various
WSN sensors deployed in the field. IoT seamlessly connects these sensor nodes,
IoT creating a dynamic and i nterconnected agricultural ec osystem.
AI The survey delves into the applic ation of AI and DL t echniques to process
DL the vast datasets generated by WSNs and IoT devices. Machine Learning
SA (ML) algorithms enable pre dictive analytics for crop management, disease
detection, and optimal res ource utilization. DL model s, with their ability to
extract int ricate patterns from dat a, pl ay a piv otal role in image
recognition for crop monit oring and yield p rediction.
Furthermore, the survey outlines the key challenges and opportunities in
deploying these technologies in SA, includ ing energy efficiency, scalability ,
and data secu rity. It discu sses current trends, emerg ing technologies, and
potential future developments in th is interd isciplin ary field.
In conclusion, this compre hensive survey p rovides a holistic overview of
the integration of WSNs, IoT, AI , and DL in S A, highl ighting the
transformative i mpact on farming practices. The synthesis of these
technologies holds the promise of ushering in a n ew era of p recision
agriculture , fostering s ustainable p ractices and ens uring food security for
a growing global populatio n.

Copyright: © 2024 by the authors. This is an open-access article distributed under the
terms of the Creative Commons Attribution-NonCommercial 4.0 International License.
(https://creativecommons.org/licenses/by-nc/4.0/)
1. Introduction technologies represents a transformative shift,
redefining traditional farming practices [1].
In an agricultural domain, significant The widespread adoption of IoT systems on a
technological advancements have ushered in a global scale signals a shift towards innovative
new era, incorporating cutting -edge novelties methodologies that leverage data generated by
like IoT, WSNs, Wireless Network Protocols, devices to enhance overall agricultural
Unmanned Aerial Vehicles (UAVs), AI, productivity. IoT facilitates seamless
Agricultural Robotics, Big Data Analytics, ML connections between machines and humans,
and Blockchain systems. This amalgamation of enabling extensive real -time information
exchange across disparate networks. In this

DOI: 10.60084/ijds.XXXX.XX Page | 8

 Journal of Innovative Intelligent Computing and Emerging Technologies (JIICET), Vol 1, No 1, 2024

context, the deployment of intelligent and applications. Here's the compressed outline
computational sensors captures real -time data, of their key contributions [7]:
effortlessly transmitting valuable insights to Irrigation Systems: With the ability to closely
individuals worldwide via the internet, monitor water usage, WSNs enable farmers to
irrespective of time or geographical constraints. effectively manage water resources and avoid
This interconnected framework not only over-irrigation. The detection of regions with
augments the efficiency of agricultural drainage problems or inadequate irrigation
processes but also opens up new possibilities distribution is made possible by real -time soil
for informed decision -making and resource moisture content monitoring .
optimization in the farming ecosystem [2]. Soil Moisture Monitoring Systems: Continuous
The agricultural landscape is undergoing a measurement and transmission of soil moisture
profound transformation propelled by the data at countless pits by WSNs empower
integration of cutting -edge technologies, farmers to optimize irrigation practices. This
namely WSNs, IoT, AI, and DL. This helps in addressing challenges related to under
amalgamation marks a paradigm shift in and over-watering.
traditional farming practices, ushering in an era Fertilizer Optimization and Control: WSNs are
of precision agriculture where data -driven useful instruments for tracking soil nutrient
insights and intelligent decision -making levels and providing current data on the
converge to optimize productivity, resource nutritional state of the soil. WSNs help refine
utilization, and sustainability [3]. fertilizer applications by precisely
WSNs form the foundational infrastructure of recommending amounts and ideal timings
this revolution, acting as the sensory nerve based on thorough study of soil data .
system of SA. These networks consist of Early Stage Control of Pest and Crop Diseases:
spatially distributed sensors strategically WSNs gather vital information about
deployed across fields to capture real -time data temperature, humidity, and other elements that
on environmental conditions, soil quality, and affect the growth of pests and diseases. This
crop health. The seamless connectivity information helps farmers prevent or minimize
facilitated by the IoT interlinks these sensor crop loss by enabling early diagnosis and
nodes, creating a dynamic network that intervention.
transcends the limitations of conventional Energy Saving and Power Consumption: WSNs
farming approaches [4]. make it easier to keep an eye on how much
AI, with its ability to process vast datasets and energy is used in irrigation systems and other
discern complex patterns, emerges as a key agricultural operations. With the help of this
enabler in this transformative landscape. M L capability, farmers may pinpoint places where
algorithms analyze the data generated by WSNs they could save energy and optimize power use
and IoT devices, providing actionable insights for more sustainable and productive farming
for crop management, disease detection, and methods.
precision irrigation. DL, a subset of AI, further In essence, the integration of WSNs with IoT, AI,
enhances the capabilities through sophisticated ML, and DL technologies enhances the
techniques like image recognition for crop precision, efficiency, and sustainability of
monitoring and yield prediction [5]. agricultural practices. The real -time data and
This comprehensive survey navigates through insights provided by these systems empower
the intricate integration of these technologies farmers to make informed decisions, leading to
in SA, offering a panoramic view of their improved resource management and increased
collective impact. By examining current productivity in SA [8].
applications, challenges, and emerging trends, The research contributes to the field in the
the survey sheds light on the potential of this following ways:
interdisciplinary approach to revolutionize I. The study extensively reviews existing
farming practices. In essence, the synthesis of literature related to SA, with a primary focus on
WSNs, IoT, AI, and DL not only propels IoT, WSNs, and wireless communication
agriculture towards unprecedented efficiency technologies. This entails a careful examination,
but also contributes to global efforts for grouping publications according to the year of
sustainable and resilient food production publication starting in 2019 and focusing on
systems [6]. network protocols and IoT -WSN application
WSNs incorporated with IoT, AI , ML, and DL play domains. The analysis sheds light on the IoT -
a pivotal role in revolutionizing SA by supplying WSN architectures and related network
crucial information tailored to definite systems protocols for SA systems .

Page | 9
 Journal of Innovative Intelligent Computing and Emerging Technologies (JIICET), Vol 1, No 1, 2024

II. Wireless communication protocols for SA, hardware prototype to test the system. The overall
such as ZigBee, Wi -Fi, SigFox, and LoRa WAN, system is developed in conjunction with automated
are examined in detail. The emphasis is on processes and systems operating on IoT through a
understanding their features and useful remotely operated portable gadget and an Internet-
applications [10]. connected computer. A robotic machine with different
III. A comprehensive survey discusses five key actuators and sensor systems (camera, object detector,
applications within SA: irrigation systems, sprayer, alarm, and cutter) is remotely controlled
energy optimisation, insect and crop disease through GPS location data. These monitors are used to
control, fertilizer optimisation, and soil keep a watch on the crop and ensure that animals
moisture monitoring. The report also looks at or birds do not harm it. Another strategy involves
how these applications have been integrated greenhouse management. Its purpose is to sense
with wireless communication technologies since temperature, humidity, and motion and to operate the
2019. light, water pump, and heater. The final mechanism is
IV. The paper explores the current problems an intelligent wireless humidity sensor node that
and unresolved concerns in SA technology. IoT - transmits gathered data to the greenhouse control
WSN scalability and reliability, data privacy and unit for the water reservoir to be actuated working on
confidentiality, network security, intrusion real-time data collected. The projects are based on the
detection, data integrity and authenticity, user device types named as follows: Raspberry Pi, ZigBee
privacy and consent, resilience to failures and modules, and the AVR AT Mega microcontroller.
attacks, location privacy, power consumption, The authors of the study [13] presented a system that
cost and standardization are all included in functioned on IoT for assessing the quality of water. The
these concerns. Possible remedies for these presented hardware model meets WHO standards for
issues are also carefully considered. [11]. water quality parameters. The gathered data is
V. Moreover, the study offers a wide range of transferred to a cloud server for storage in real-time and
future recommendations covering sophisticated analysis. The primary factor of this study concerns
architectures, such as blockchain technology, the prediction, using a ML model to approximate
5G and 6G networks, agri -robotics, artificial metrics for water quality through the cloud system.
intelligence (AI) and artificial general Sughapriya et al. [15] planned a system for analyzing
intelligence (AGI) systems, and renewable water quality using IoT and multiple sensor modules.
energy. Future farmers stand to gain from these This system monitors water quality by detecting pH,
developments if they open up new possibilities conductivity, temperature, and turbidity using
for economical, sustainable, and user -friendly different sensors. Information from the sensors is then
agricultural systems [9]. retrieved by the Arduino controller. The acquired data is
evaluated using IoT, and water pollution can be
1.1 Paper Structure examined using a stringent mechanism. Furthermore,
the proposed system provides messages and alerts to
The structure of this paper unfolds as follows. concerned municipalities and specialists regarding the
In Section II, a comprehensive exploration of quality of water. The presented model is comprised of
relevant study is undertaken through a many sensors that analyze water quality data in real-
thorough review and analysis of prior research. time for quick deployment. Furthermore, the
Moving to Section III, a detailed discussion presented model is accurate, cost-effective, and requires
about SA parameters and studies that have fewer workforces. In another research by Krishnan et al.
employed various technologies to scrutinize [14] different approaches involving DL methods are
these metrics. Section IV is dedicated to an in - discussed and their importance in the field of water
depth discussion of the issues and challenges management techniques.
faced during course of a research, accompanied Jerom B. et al. [15] suggested an Intelligent Water
by suggestions for potential avenues to extend Quality Evaluating System that operated on IoT that
this project in the future. Section V shows integrates DL and Cloud techniques to examine the
Future Directions, ultimately Section VI water quality of different water resources. Existing
encapsulates the report, drawing together the monitoring methods are capable of manually collecting
findings and insights gleaned throughout the water samples from various water resources, followed
paper, culminating in a coherent and conclusive by monitoring and evaluation in a laboratory. This
summary. procedure is often ineffective because it is tough,
requires a long time, and does not produce results in
2. Literature Review
real time. Consistent evaluation of water quality is
The authors in [12] presented a multi-objective required to ensure the safe delivery of water to
intelligent agriculture system. They prepared a consumers from any water resources or supplies. Thus,
manufacturing and implementing a cost-

Page | 10
 Journal of Innovative Intelligent Computing and Emerging Technologies (JIICET), Vol 1, No 1, 2024

efficient method for real-time analysis of water quality and the IoT device transfers them to the server for
parameters implementing IoT is now required. The analysis of the water pollutant emissions. In a study [20],
prototype developed for this study detects a variety of a system working on IoT for monitoring the quality of
intoxicants in the water. Different sensing devices are water is presented in which sensors are utilized to
used to analyse several different factors to determine determine the pH and temperature levels of water
the water quality from water resources. The collected to assess the parametric aspects of the water.
data is stored in the Cloud, and DL algorithms are In a paper presented by Wai et al. [21], a review was
applied to analyse if the water under test is safe to drink conducted to discuss the DL methods application in the
or not. field of water management. This paper helps the
Anuradha et al. [16] used IoT to develop an affordable authorities in decision-making regarding a sustainable
system that monitored the quality of water in real-time. approach in the field of agriculture. According to [22],
The strategy describes a Water Quality Testing approach the proposed system is composed of water quality
based on sensors that analyse the chemical and physical testing of pH, Turbidity, and Temperature sensors, an
metrics of water. Water properties are monitored Arduino controller data processing unit, an information-
using sensors and interpreted using a controller, in this provided module, a monitoring centre, and other
case, Raspberry Pi. Lastly, the acquired measurement equipment. Throughout the day, turbidity, pH, water,
from the sensing device is visible on the internet using and temperature are automatically sensed by an
the Thing Speak API. The water monitoring system individual microcontroller. The data is collected by a
devised in this project has numerous returns, including single chip, which then functions and analyses it. If the
good mobility, high frequency, and minimal power water quality is not up to the standards, the data is
consumption. Quality factors such as hardness, transferred to the monitoring center, and the users
conductivity, ammonia, iron, fluoride, and chloride are alerted instantly. This makes it easier for supervisors
content can also be tested for quality of water, and the to take appropriate steps on time and to monitor real-
values obtained are used to test the purity of the water time water quality conditions remotely. Similarly, [23]
for several different purposes like every day needs for proposes the system of an IoT-based water quality
industrial sectors and water consumption by humans analysis that examines water quality on a real-time
for drinking. basis. This system includes sensors that detect water
Demetillo et al. [17] offered a low-cost, real-time water quality constraints like pH, dissolved oxygen, turbidity,
quality assessment method for distant rivers, ponds, conductivity, and temperature. The obtained data
and other natural water resources. The main collected by the sensors are processed by the
components of the system are a microcontroller, basic microcontroller to make them Zigbee module compliant.
sensors operating on an electrochemical approach, a This computed data is remotely delivered to the call
purposely built buoy, and a system of controller through the Zigbee network. Finally, by using
wireless communication technology. The designed cloud computing, sensor data can be accessed on an
structure can monitor pH, water temperature, internet browser application.
and dissolved oxygen at pre-set periods. To best serve ML models are currently being employed all over the
the authorized customers, the suggested system sends world to make devices smart by executing predictive
the gathered data in charts and graphs styles to a analysis. A study [24], presents a drinking water
tailored web-based server and authorized mobile predictive study in which the quality of water is assessed
phones. To assess the performance of the system, the using pH, dissolved solids, and turbidity. To calculate the
buoy's durability in unfavourable environmental correlation between calculated parameters, a linear
conditions, the power usage of the overall system, regression model is implemented for the measured
efficiency in data transmission, and information display model metrics. Similarly, in [25], a Fuzzy Neural Network
in a web-based software tool were carefully analyzed. is deployed to a three-year dataset to estimate water
The outcome of the research demonstrated that the quality utilizing water quality metrics.
designed method had better prediction and could be In another approach, Tace et al. [26] present a technique
used for practical management of the environment by based on ML for smart irrigation that can be
providing the users with important and timely statistics implemented in different regions. They made sure the
for improved action plans. system was cost-effective and had good power
In [18], a practical monitoring method for surface water consumption.
quality is presented employing an In 2019, Mohammad Rezapour et al. evaluated
affordable seawater sensor capable of sensing water evapotranspiration (ET) by analyzing and contrasting
conductivity, temperature, and turbidity. Another three different models: SVM, and adaptive neuro-
strategy based on IoT is described in [19], which uses fuzzy inference system (ANFIS). The effective
drone machinery in combination with sensors operable evapotranspiration for semi-arid areas was estimated by
in water to check quality. The measured values of the all three models [27]. ET simulations were executed on
real-time metrics collected by the sensors are data spanning 1970 to 2010, with input values from five
transmitted to the main drone over the RF transceivers, various combinations in southern Iran. The SVM, an ML-

Page | 11
 Journal of Innovative Intelligent Computing and Emerging Technologies (JIICET), Vol 1, No 1, 2024

based model outperformed the other two methods, estimate the average heat required for this process. This
with daylight hours, average air temperature, relative is in addition to the sensors' convenience of
humidity, and air pressure as input data for the model. manufacture. Alam, Clyne, and Deen [33] deployed
Rokade et.al [28], presented an innovative approach for Wheatstone bridge sensors to provide high-precision
the smart farming industry. This paper has outlined an temperature readings with minimum variability ranging
efficient smart informatics system of farming with between 0 and 50 C. There are four connecting
predictive information analytics on measuring terminals of which two are made of Silicon wafers of P-
characteristics in an intelligent farming system using a type, which offers increased Coefficient Resistance of
supervised ML method. Temperature (TCR). This is the computation of
Nikoo and Mahjouri [29], used a Probabilistic Support comparative variation in resistance as each degree of
Vector Machines (PSVMs) model in combination with a temperature shift takes place. The remaining two
GIS approach to manage the categorization and terminals are made up of polystyrene.
circulation of shallow and underground water in Iran. According to the authors, the sensor was also integrated
They claimed using these two methodologies would into the Arduino framework using Android platforms
deliver reliable data for effective research into water- integrating programs. A sensor connected to resistive
conserving initiatives. In many case studies, Heddam devices, Wu et al. designed for analysis of heat utilizing
[30] used artificial neural networks (ANN) to assess platinum (Pt) film, which is a great conductor and offers
water quality components. He claimed that AI features that support the measurement of heat factor.
methodologies are capable of modeling and forecasting Finally, Simic et al. [34] proposed another approach for
the integrated relationship between water quality resistive temperature sensor measurement, this time
parameters and exhibiting their periods. employing a budget-friendly and commonly available
sensor (LM35). They calibrated the gadget in the
3. Few Parameters used in Smart Agriculture research laboratory and achieved a precision of 0.23 C.
Lastly, Huang et al. [35] observed temperature using
3.1 Temperature optical fiber. Despite its high price, this approach is
typically adopted to monitor temperature when the
Temperature sensors, akin to pH sensors, are integral
optical fiber is also utilized to determine other factors.
components in a variety of multi-parametric sensing
Two insulated optical fiber terminals were deployed
devices. This prevalence arises from the pivotal role that
since the characteristics the researchers were
temperature plays in determining water quality, given
monitoring are highly sensitive to temperature changes.
that numerous parameters are intricately linked to
Thus, by varying the various central wavelengths, a
temperature variations (e.g., bioactivity, pH,
linear relationship with temperature was discovered,
conductivity, and dissolved oxygen). Moreover, the ease
enabling the device to be calibrated.
of monitoring temperature is underscored by its strong
linear relationship with resistivity and electromotive 3.2 pH
force. This makes temperature a key factor in ensuring
the accuracy and reliability of water quality Due to the vital role of pH in guaranteeing optimal water
assessments, as it significantly influences various critical quality, it is regularly assessed and incorporated into
parameters. The inclusion of temperature sensors in nearly all multiparameter tools. Various techniques,
sensing devices thus becomes imperative, not only due including visual inspection, potentiometric, and
to the interdependence of multiple water quality photometric methods, are employed to determine the
indicators on temperature but also owing to the pH of a water sample. The visual method, utilizing
practical advantages offered by its straightforward and specific materials like litmus paper and relying on color
linear monitoring capabilities. changes as a pH indicator, is characterized by lower
Water temperature can be determined by using range of accuracy levels and provides only approximate pH value
technologies, consisting thermal expansion of medium, estimations.
thermoelectric reactions, optical fiber, semiconductors, Nernst equation is the basis of the potentiometric
electrical resistance, and capacitance [31]. The approach, which estimates the variation in hydrogen ion
application of thermoelectric tools and/or resistive levels resulting from chemical experiments.
sensors, on the other hand, is the most typical less Spectrophotometry is the fundamental core of the
costly temperature measurement technique. These photometric approach that offers data on the varying
techniques are popular due to their accuracy, affordable pH samples as the shifts in wavelength absorption
price for the required range of temperatures for water occur.
monitoring, stability, and ease of use [32]. PH sensors were designed in various ways by Deen,
The resistive technique is the most widely used method Alam, and Clyne [36], Wu et al. [37], and Simic et al. [38].
for determining temperature. The reason is that In a polyimide substrate, Alam et al. [39] employed a
thermoelectric sensing equipment, particularly specific palladium (Pd) ink with silver-chloride diodes
thermistors, generally uses resistive sensor nodes to (AgCl) that worked as a baseline. Wu et al. measured pH

Page | 12
 Journal of Innovative Intelligent Computing and Emerging Technologies (JIICET), Vol 1, No 1, 2024

using ruthenium (Ru) redox; in comparison to Pd, Ru Wang et al. also experimented with cost-effective device
has less impurity risk, is simple to prepare, and has for measuring turbidity and a web-based water quality
better chemical resistance. The researchers were able to assessment project that incorporated an 850 nm
assess pH with an accuracy of 1.02% between 4.01 and infrared LED, a customized IoT platform for
10.87. Finally, Simic et al. [39] designed a pH sensor by communication, and dual orthogonal photodetectors.
applying titanium dioxide (TiO2) as the core component The study found that the device could measure turbidity
linked to a digital communication electronic circuit with good precision and reliability equal to existing
(using an AD5933 adapter). sensors. In another experiment, Rahman et al. [45] used
Hossain et al. [40] deployed a multi-parametric device to a device that worked with LED for water measurement
examine water quality parameters in their work. For the of turbidity and also observed how it responded to
analysis of the pH of water, a Photo-Induced Electron different visible light colors used for the task and
Transfer (PET) technique using 4-aminonapthlimide was determined the best photo detector in terms of
used as a color agent. Dutta, Nath, and Sarma [41] power fluctuation in the ON/OFF state. The writers
attempted to implement photometric evaluation demonstrated that the white light offers the greatest
on transparent fluids, that is, without the need for color efficiency with lower than 8% experimental errors in
agents for testing. The sensor captures the visual most measures, after which follows a UV LED, however,
spectrum and conducts the pH measurement by optical both beams were appropriate for measuring the
pre-processing and subsequent exchange into intensity turbidity of water fluctuating between 0 to 1000 NTU. In
and wavelength scattering. another study, Schima et al. [46] designed a
To calculate the pH of the samples, Silva et al. [42] photosensitive method for real-time evaluation of
utilized a colorimetric-based device that paired the turbidity that used sensors in the ultraviolet band of the
camera of a smartphone with a microfluidic gadget electromagnetic field and showed significant precision
based on paper. For the approach, 3d printing supports when compared to laboratory criteria. Furthermore, the
were designed to assure the device's durability, and the Python script running on a Raspberry Pi controller was
system was capable of determining pH in the 4.7-12 liable for communicating with a detector, demonstrating
range. that open-source technology can be vital to robust and
reliable systems even in the laboratory phase.
3.3 Turbidity
3.4 Total Dissolved Solids (TDS)
Water turbidity is a measure of the extent to which
incident light is intercepted when passing through "Dissolved solids" refers to minerals, saline compounds,
water, primarily caused by the presence of dissolved or metals that are in a dissolved state within water. Total
particles like inorganic and organic compounds that lead Dissolved Solids (TDS) encompass organic compounds,
to a cloudy appearance. Consequently, turbidity serves predominantly consisting of calcium, potassium,
as a fundamental indicator for evaluating water quality, magnesium, bicarbonate, sodium, chlorides, and
helping ascertain its suitability for consumption and, in sulphates, along with trace amounts of organic
turn, acting as a preventive measure against waterborne compounds dispersed in the water.
infections. The TDS sensor module is plug-and-play, user-friendly,
Considering the significance of turbidity, various and compatible with IoT devices. Figure 5 shows this
turbidity sensors widely available in the market can be sensor. We can easily construct a TDS detector to
coupled with other water quality indicators to create determine the TDS value of water.
IoT-based online systems for analysis, as implemented The concentrations of TDS can increase as a result of
and addressed in extensive research. To optimize and wash-off from salted roads in the winter season.
reduce the costs related to monitoring turbidity, a Increasing amounts of nitrate or phosphate ions may be
recent study, such as that by Azman et al. [43], has produced by wastewater treatment projects that use
devised a cost-effective technique that works on a organic content. When TDS intensity increases,
turbidity sensor that is nephelometric for consistent especially when dissolved salts are involved, many
measurement of the quality of water. The species of aquatic life suffer. The dissolved
experimenters claim that the functionality of electrical compounds dehydrate animal hides. TDS concentrations
sensing devices relies on the density of reflected light by in rivers and streams are frequently found to
light dispersing in fluids and solids employing an LDR range between 50 and 250 mg/L. In areas with mostly
(Light Dependent Resistor) as receiver, LED (Light hard water or incredibly high salinity, its amounts can
Emitting Diode) as a transmitter, and RS232 device for extend up to 500 mg/L. It is a water quality metric
linkage between sensing device and desktop. Arifin et al. derived from the total suspended particle loss on
[44] explored construction of a sensing device for ignition. It is vital in the treatment of water and
measuring water turbidity employing an LED, a wastewater.
photodetectors a polymer optical fiber as major As per Sibal and Espino [49], most commonly used lab-
materials, and, promising sensitivity values. based methods for analyzing TDS in water are atomic

Page | 13
 Journal of Innovative Intelligent Computing and Emerging Technologies (JIICET), Vol 1, No 1, 2024

emission spectrometry (AES), atomic absorption 3.6 Color

spectrometry (AAS), inductively coupled plasma MS (ICP-
MS), mass spectrometry (MS), X-ray fluorescence (XRF), Prior to technological advancements and the
and optical technologies. According to Garciá-Miranda introduction of remote monitoring tools, color of water,
Ferrari et al.[47], these procedures are complex and defined by reflection of light in minute particles from
costly and may involve the usage of preconcentration organic or mineral sources, served as a longstanding
and extraction procedures for high-analysis enactment. indicator for water quality analysis. However, with the
Methods using electrochemical sensors, on the other progression of technology in society, increased research
hand, may be more effective, due to their small size, low attention has been directed toward understanding and
cost, simple installation, and simple to use. Phyllis et al. characterizing water color. Edwards constructed sensor
[51]presents a review of the effects that total dissolved to assess color and turbidity of water sources
solids have on marine life and which species are more through four-beam frequency benefit method for
sensitive and affected by the TDS intensity are different precise assessment. At the time, this prototype, which
life stages. was utilized in water treatment unit, was considered to
be very innovative research.
3.5 Salinity Studies like Murphy et al. [54] planned cost-effective
optical sensor that helps in evaluating the water quality
Salinity is the measurement of the dissolved salt content based on a multi-wavelength source of light and has two
in water, typically expressed in units such as parts per photodiode sensors that can measure light transmission
thousand (ppt) or percentage (%). For freshwater and sideways scattering at the top of the detector to
sourced from rivers, the salinity level is generally 0.5 ppt determine the metrics of color and turbidity.
or lower. Procedure was carried out in a research center, but
The salinity of the oceans differs, but the relative research team wishes to employ sensor in long run as
quantities of the most essential dissolved elements the real-time major water contamination evaluation
remain almost steady, regardless of the existence of technique.
lower percentages of other salt compounds in seawater,
sodium (Na+) and chloride (Cl-) ions makeup
approximately 91% of all seawater ions. Freshwater has
far fewer salt ions. Electrical conductivity (EC) testing is
frequently used to calculate salinity. EC is calculated by
flowing an electric current through a water sample
between two metallic plates or electrodes and observing
how quickly current flows between the plates.
Two types of sensors are used as salinity sensors:
Interferometers and fiber gratings. Cong et al. [47] and
Liu et al. [48] modeled the salinity sensor with a fiber
grating, with a salinity sensitivity of 10.4 pm/%.
Gentleman and Booksh [49] used surface plasmon
resonance to determine the salinity of a liquid
deploying a multimode optical fiber. Sensitivity of the Figure 1. Number of studies reported each year on the topic
optical fiber increased to 200 pm/% compared to the [53]
standard prism refractive index approach. Hussain et al.
Considering the significance of gaining knowledge of the
[47] proposed two techniques for measuring salinity of
connection between color and other metrics used in
water using smartphones, the possibility was proven,
water quality assessment, Yang [55] devised a
and the mobility of the salinity sensor was enhanced.
multisensory approach for assessing water quality
In the research [52], a sample of saline wastewater from
factors (pH value, temperature, ammonia, nitrogen, and
different sources was collected that contained different
color) for fish farming using algorithmically improved
kinds of salt contents, pollutants, organic and inorganic
sensing devices, with the evaluation of color constraint
compounds, and many others. Keeping in view the
in water sent in real-time through Zigbee Network
damage these cause to the environmental system and
communication standards. In addition, Saravanan et al.
their contribution to land degradation, various methods
[56] discussed an IoT-based monitoring system for the
of treatment of saline water were proposed in this
quality of water that involves color as one of the factors
paper. The most commonly used techniques to
to be monitored in real-time.
desalinate the wastewater into the freshwater discussed
in this research are reverse osmosis, membrane
distillation, freezing process, electro dialysis, and 3.7 Nitrogen
desalination process.

Page | 14
 Journal of Innovative Intelligent Computing and Emerging Technologies (JIICET), Vol 1, No 1, 2024

Nitrogen is prevalent in various environmental sources, poses a substantial challenge, requiring standardized
including decomposed plants, wildlife, human protocols and robust integration frameworks.
excrement, manure, and pesticides. Monitoring nitrogen 5. Lack of Standardization: The absence of standardized
levels can be accomplished through diverse techniques protocols across different devices and platforms
like chromatography, electrochemistry, and hampers the seamless integration of technologies in SA.
spectroscopy techniques. Standardization is essential for ensuring compatibility,
Chromatography is utmost effective approach when data consistency, and the interoperability of various
there is organic material in testers as it does not solutions.
undergo involvement from other composites; however, 6. Skill Gaps and Education: The effective deployment of
it is considered an expensive and complicated strategy advanced technologies in agriculture requires a skilled
of use because it requires unique methods and specific workforce. There is a notable gap in agricultural
components in the procedure [43]. Spectroscopy can education and training programs that cover WSNs, IoT,
also encounter intervention, but it is more convenient to AI, ML, and DL. Bridging this gap is essential for farmers
utilize than the other two techniques and generates and agricultural professionals to harness the full
results faster. Electrochemistry monitoring has a potential of these technologies.
significant potential for inexpensive applications, but it 7. Cost Considerations: The initial investment and
can be affected by organic and ion compounds in trials. ongoing costs associated with deploying and
An evolution of inexpensive sensors for the real- maintaining WSNs, IoT devices, and AI systems can be a
time identification of nitrogen has attracted significant barrier, especially for smaller or resource-
consideration in the paper because it is a key factor constrained farms. Achieving cost-effectiveness without
from an evaluating point of view. For example, Akhter et compromising functionality is a persistent challenge.
al. [57] designed a sensor system in which the substrate 8. Environmental Impact: While SA aims to
used is poly dimethyl siloxane, and the conductive enhance sustainability, the environmental
medium is tough outer carbon nanotubes. Grapheme impact of deploying technology -intensive
has good electrical characteristics, but large-scale solutions should be carefully considered. Issues
manufacture is complicated, which may adversely such as electronic waste, energy consumption,
impact the predictive ability of the designed sensor. and the ecological footprint of technology
implementations need attention.
4. Issues and challenges in WSN IoT AI ML and DL in 9. Ethical Considerations: The use of AI and ML in
smart agriculture decision-making processes for agriculture raises ethical
concerns. Issues related to transparency, accountability,
An addition of WSNs, IoT, AI, ML, and DL in SA introduce and biases in algorithmic decision-making must be
a myriad of opportunities, but it is not without its share addressed to ensure fair and responsible use of technology
of challenges. This section explores the key issues and in farming practices.
hurdles faced in deploying these advanced technologies 10. Connectivity in Remote Areas: In agricultural regions
in the agricultural landscape. with limited connectivity, ensuring reliable network
1. Scalability and Network Coverage: One of the primary access for WSNs and IoT devices becomes challenging.
challenges in WSNs and IoT deployment is ensuring This is particularly relevant in remote rural areas where
scalability and adequate network coverage. Large agriculture is a primary economic activity.
agricultural areas may pose difficulties in maintaining a In conclusion, addressing these challenges is crucial for
reliable and expansive network, leading to gaps in data the successful and sustainable implementation of WSNs,
collection and communication. IoT, AI, ML, and DL in SA. Collaborative efforts from
2. Data Privacy and Security: The vast amount of data technology developers, policymakers, and the
generated by WSNs, IoT devices, and AI systems in SA agricultural community are essential to overwhelmed
raises significant concerns about data privacy and these hurdles and expose occupied prospective of SA
security. Protecting sensitive information related to crop for a more efficient and sustainable future.
yields, soil conditions, and farm practices is crucial to
prevent unauthorized access and potential misuse. 5. Future Directions
3. Energy Efficiency: Many WSNs and IoT devices in
agriculture operate in remote or challenging Cutting -edge digital technologies have made
environments. Ensuring energy efficiency for these remarkable advancements, seamlessly
devices, especially those reliant on batteries or integrating with Internet of Things and Wireless
renewable sources, is crucial for maintaining Sensor Networks (IoT -WSNs) to enhance the
uninterrupted data collection and communication. sustainability of applications in SA. This
4. Integration Complexity: Integrating diverse integration brings about substantial
technologies like WSNs, IoT, AI, ML, and DL can be optimization across whole spectrum of
complex. Achieving seamless interoperability and agricultural procedures, spanning from
effective communication between these components cultivation to harvest and revolutionizing whole

Page | 15
 Journal of Innovative Intelligent Computing and Emerging Technologies (JIICET), Vol 1, No 1, 2024

agricultural sector. These forward -thinking once-in-a-lifetime chance to improve resource

enterprises mark a pivotal moment in reshaping efficiency, optimize farming methods, and
an agricultural background. dramatically increase agricultural yield. The
Tailoring of digitization to meet specific sections that follow provide a thorough analysis
requirements demands considerable financial of the several AI -based tools that can be used
investment to align with individual necessities to advance sustainable agriculture . These
of farmers. To fortify consistency of this discussions delve into the potential impact and
digitization, it is imperative to embrace contributions these technologies can make to
government-backed initiatives, leverage grants, reshape and advance the agricultural
foster strategic public -private partnerships, and landscape.
implement open data policies. Such initiatives
should be complemented by research efforts II. ML in Smart Agriculture
tailored to regional nuances, strengthening
their impact and effectiveness. As a component of AI, ML is specifically dedicated to
Achieving precision in tailoring digitization to fostering the development of computer systems that
unique requirements demands substantial possess the ability to learn and adapt through
financial commitment that caters to the specific experiences. When employed in the context of Internet
demands of individual farmers. Equally crucial of Things and Wireless Sensor Networks (IoT-WSNs)
is the establishment of transparent data within agriculture, ML algorithms undertake a thorough
policies, ensuring reinforcement by local analysis of historical and real-time data. Primary goal is
research activities. An organized tactic involves to refine irrigation patterns, anticipate and proactively
particular implementation of well -structured manage crop diseases, and automate diverse
roadmap for development of SA systems. This agricultural processes. This integration plays a pivotal
journey commences with inaugurating initial role in substantially augmenting competence and total
architecture that comprises vital constituents output within an agricultural domain.
and streamlined functionalities, setting the Artificial General Intelligence (AGI) stands at forefront,
stage for a comprehensive transformation in showcasing significant prospective to exert profound
agricultural practices. impact across diverse sectors, with agriculture
prominently featured. AGI, particularly in domains such
I. DL in Smart agriculture as Natural Language Processing (NLP) and Agri-Robotics,
holds the capability to improve crop yields, minimize
DL, an advanced technique within the realm of wastage, and advance sustainable farming practices.
ML, employs multi -layered neural networks to This distinctive capability uniquely positions AGI as likely
mimic complex structure of human brain. DL solution to complex contests encountered by an
algorithms play an indispensable role in context agricultural division.
of Internet of Things and Wireless Sensor
Networks (IoT -WSNs) specifically tailored for SA. III. Agri-ROBOTS
These algorithms excel in cultured data
analysis, particularly in areas such as image Agri-robotics entails incorporating robotic
recognition and Natural Language Processing systems and automation technologies into
(NLP), contributing significantly to profound agricultural operations. When coupled with IoT -
understanding of agricultural procedures . WSNs, these robotic systems gain additional
In this regard, computer vision becomes a capabilities through the integration of sensor
crucial component, enabling machines to data, optimizing tasks like accurate planting,
decode and understand visual data taken from monitoring, and harvesting. This integration
pictures or movies. By carefully analyzing leads to improved efficiency and productivity in
imagery data, its incorporation into IoT -WSNs agriculture.
for agricultural applications increases the The precision of these robots stems from their
potential for monitoring crop health, quickly specialized sensing and actuation capabilities,
identifying illnesses, and precisely estimating potentially reducing the need for labor while
growth stages. This integration makes it easier enhancing various agricultural processes.
to make educated, timely adjustments to Drones play a vital role in tasks such as
agricultural methods, which eventually pesticide spraying, irrigation, crop harvesting,
increases productivity and promotes efficiency . seed sowing, and soil cultivation, playing a
Modern AI technologies combined with wireless pivotal role in transforming conventional
sensor networks (WSNs) have emerged in recent farming practices.
years as a promising new direction for the
agriculture industry. This integration offers a IV. 5G and 6G in smart agriculture

Page | 16
 Journal of Innovative Intelligent Computing and Emerging Technologies (JIICET), Vol 1, No 1, 2024

Blockchain in SA offers a decentralized and

The evolution of 5G technologies and the distributed ledger that records transactions and
anticipated revolution ushered in by 6G within information at each stage of the agricultural
the context of IoT -WSNs holds significant process. This ledger, comprised of blocks linked
consequences for an agricultural region. 5G, in a secure chain, ensures data integrity and
marking 5 t h generation of wireless technology, transparency. Each block contains a time
signifies the substantial advancement; stamped record of transactions, making it
introducing key improvements likewise virtually impossible to alter historical data
accelerated data speeds, reduced latency, and without consensus from whole network.
enhanced connectivity. These enhancements One notable application of blockchain in SA is
translate into heightened real -time monitoring supply chain management. It allows for
capabilities and faster transmission of sensor traceability of agricultural products from farm
data, ultimately refining coordination among to consumer, providing a secure and
devices in agricultural solicitations . unalterable record of each step in the
Looking towards the future, the developmental production and distribution process. This
phase of 6G technology holds a potential to transparency enhances food safety by quickly
revolutionize data collection and analysis, identifying and isolating the source of
potentially achieving speeds in terabits -per - contamination or other issues.
second choice. This ensures exceptionally Smart contracts , self-executing contracts with
timely and precise insights for farmers. The terms of an agreement directly written into
integration of 5G and an intended capability of code, further enhance functionality of
6G within IoT -WSNs for agriculture present blockchain in agriculture. These contracts
massive potential. This combination facilitates automate and enforce predefined rules,
real-time monitoring of critical factors, facilitating seamless and trustless transactions.
including soil conditions, crop health, weather For instance, smart contracts can be employed
patterns, and equipment performance, to automate payments between farmers and
empowering optimized decision -making. suppliers based on predefined conditions,
Current generation of intelligent agricultural streamlining financial transactions and
applications, trusting on the relatively limited reducing the risk of disputes.
number of wireless sensors, demands improved Moreover, blockchain technology enhances data
accuracy and effectiveness. Conversely, an security and privacy in SA. Farmers can securely
ongoing development of 6G communication store and share sensitive data, such as crop
technologies lays foundation for forthcoming of yields or soil quality, with authorized parties
intelligent and sustainable agriculture. 6G without compromising the integrity of the
technology initiates to allow an interconnection information. This decentralized approach
of general sensors, granting farmers capacity to mitigates the risks associated with centralized
gather intricate, plant -specific information. This data storage systems, reducing the likelihood of
transformative potential fueled by 6G data breaches or unauthorized access.
technology is poised to revolutionize SA, In conclusion, the integration of blockchain
ensuring precise data collection, advanced technology in SA brings forth a paradigm shift
robotics, and precision agriculture in remote by introducing transparency, security, and
locations, thereby making agriculture efficiency across the agricultural supply chain.
considerably more efficient, ecological, and This innovative approach not only enhances
economical. traceability and food safety but also
streamlines transactions and ensures the
V. Block chain Systems in Smart Agriculture integrity of sensitive agricultural data. As an
agricultural industry continues to hold digital
transformation, blockchain stands as a robust
The integration of blockchain technology in SA
and promising solution to address key
represents a significant advancement with far -
challenges and propel SA into a more secure
reaching implications. Blockchain, originally
and transparent future.
devised as a decentralized and secure ledger
for crypto currencies , has found applications in
various industries, including agriculture. In the 6. Conclusion
context of SA , blockchain serves as a
In conclusion, the comprehensive survey on the
transparent and tamper -resistant system for
integration of WSNs, IoT, AI, and DL in S A
managing, recording, and verifying data across
reveals a transformative landscape for the
the agricultural supply chain.
agricultural sector. The amalgamation of these

Page | 17
 Journal of Innovative Intelligent Computing and Emerging Technologies (JIICET), Vol 1, No 1, 2024

cutting-edge technologies offers a holistic and Conflicts of Interest: “All the authors declare that there are no
data-driven approach to farming practices, conflicts of interest.”
bringing about significant advancements in
References
efficiency, productivity, and sustainability.
The synergy of WSNs and IoT provides real -time 1. Rahu, M. A., Chandio, A. F., Aurangzeb, K., Karim, S., Alhussein,
monitoring and data acquisition capabilities, M., & Anwar, M. S. (2023). Towards design of Internet of Things
enabling farmers to make informed decisions and machine learning-enabled frameworks for analysis and
prediction of water quality. IEEE Access.
based on accurate and timely information. AI,
2. Shaikh, F. K., Karim, S., Zeadally, S., & Nebhen, J. (2022). Recent
with its ability to analyze vast datasets, trends in internet of things enabled sensor technologies for
optimizes various agricultural processes such smart agriculture. IEEE Internet of Things Journal.
as irrigation, soil management, and pest 3. Mirani, A. A., Memon, M. S., Rahu, M. A., Bhatti, M. N., & Shaikh,
U. R. (2019). A review of agro-industry in IoT: applications and
control, contributing to resource efficiency and
challenges. Quaid-E-Awam University Research Journal of
yield improvement. Engineering, Science & Technology, Nawabshah, 17(01), 28-33..
The incorporation of DL, particularly in image 4. Das, B., Ali, S. M., Shaikh, M. Z., Chandio, A. F., Rahu, M. A.,
recognition and data analysis, enhances the Pabani, J. K., & Khalil, M. U. R. (2023, January). Linear
precision of monitoring and decision -making Regression Based Crop Suggestive System for Local Pakistani
Farmers. In 2023 Global Conference on Wireless and Optical
within SA. This holds promising implications for
Technologies (GCWOT) (pp. 1-6). IEEE.
crop health assessment, disease detection, and
5. Mowla, M. N., Mowla, N., Shah, A. S., Rabie, K., & Shongwe, T.
overall crop management. The survey (2023). Internet of Things and Wireless Sensor Networks for
underscores the potential of these technologies Smart Agriculture Applications-A Survey. IEEE Access.
to revolutionize traditional farming methods, 6. Lashari, M. H., Karim, S., Alhussein, M., Hoshu, A. A.,
paving the way for more intelligent, data - Aurangzeb, K., & Anwar, M. S. (2023). Internet of Things-based
driven, and sustainable agricultural practices. sustainable environment management for large indoor
facilities. PeerJ Computer Science, 9, e1623.
However, amidst the promises, challenges such 7. Rahu, M. A., Karim, S., Shams, R., Soomro, A. A., & Chandio, A.
as scalability, data security, and standardization F. (2022). Wireless Sensor Networks-based Smart Agriculture:
need attention for the seamless integration of Sensing Technologies, Application and Future
Directions. Sukkur IBA Journal of Emerging Technologies, 5(2), 18-
these technologies. Collaborative efforts from
32.
researchers, policymakers, and industry 8. Mirani, A. A., Memon, M. S., Bhati, M. N., Soomro, M. A., &
stakeholders are essential to address these Rahu, M. A. (2017, December). Taxonomy of ubiquitous
challenges and unlock full potential of SA. computing: Applications and challenges. In 2017 International
Conference On Information And Communication Technologies
In essence, the survey illuminates the
(Icict) (pp. 202-208). IEEE.
transformative power of integrating WSNs, IoT, 9. Rahu, M. A., Kumar, P., Karim, S., & Mirani, A. A. (2018).
AI, and DL in agriculture. The synergy of these Agricultural Environmental Monitoring: A WSN
technologies has the potential to usher in a new Perspective. University of Sindh Journal of Information and
Communication Technology, 2(1), 17-24.
era of precision farming, where data -driven
10. Mirani, A. A., Memon, E. M. S., Chohan, R., Sodhar, I. N., & Rahu,
insights contribute to more efficient resource M. A. (2021). Irrigation scheduling, water pollution monitoring
utilization, reduced environmental impact, and in IoT: A Review.
11. Rahu, M. A. (2018). Energy Harvesting for Water Quality
increased resilience in the face of evolving
Monitoring using Floating Sensor Networks: A Generic
agricultural challenges. As SA continues to Framework. Sukkur IBA Journal of Emerging Technologies, 1(2),
evolve, the integration of these technologies 19-32.
remains a cornerstone in shaping the future of 12. N. Gondchawa and P. D. R. S. Kawitka, "IoT based Smart
Agriculture," International Journal of Advanced Research in
sustainable and intelligent farming practices.
Computer and Communication Engineering, vol. 5, no. 6, pp. 1-
5, 2016
Author Contributions: “Conceptualization, M.A.R. and S.M.A.; 13. Shafi, U., Mumtaz, R., Anwar, H., Qamar, A. M., & Khurshid, H.
(2018, October). Surface water pollution detection using
methodology, S.K; and N.S writing—original draft preparation,
internet of things. In 2018 15th international conference on
M.A.R.; writing—review and editing, G.M.J.; visualization, M.A.R.; smart cities: improving quality of life using ICT & IoT (HONET-
supervision, S.K.; All authors have read and agreed to the ICT) (pp. 92-96). IEEE.
published version of the manuscript.” 14. Krishnan, S. R., Nallakaruppan, M. K., Chengoden, R., Koppu, S.,
Iyapparaja, M., Sadhasivam, J., & Sethuraman, S. (2022). Smart
Funding: “This study does not receive external funding.”. water resource management using Artificial Intelligence—A
review. Sustainability, 14(20), 13384.
Ethical Clearance: “Not applicable”. 15. Ajith, J. B., Manimegalai, R., & Ilayaraja, V. (2020, February). An
IoT based smart water quality monitoring system using cloud.
Informed Consent Statement: “Not applicable.” In 2020 International conference on emerging trends in
information technology and engineering (ic-ETITE) (pp. 1-7). IEEE.
16.
Data Availability Statement: “Not applicable.” Anuradha, T., Bhakti, C. R., & Pooja, D. (2018). IoT based low
cost system for monitoring of water quality in real time. Int. Res. J.
Eng. Technol.(IRJET), 5(5).
Acknowledgments: Thanks Co-authors for support and help. 17. Demetillo, A. T., Japitana, M. V., & Taboada, E. B. (2019). A
system for monitoring water quality in a large aquatic area using
wireless sensor network technology. Sustainable Environment
Research, 29, 1-9.

Page | 18
 Journal of Innovative Intelligent Computing and Emerging Technologies (JIICET), Vol 1, No 1, 2024

18. Adamo, F., Attivissimo, F., Carducci, C. G. C., & Lanzolla, A. M. L. monitoring system," IEEE International Conference on
(2014). A smart sensor network for sea water quality Automatic Control and Intelligent Systems, pp. 202-207, 2016.
monitoring. IEEE Sensors Journal, 15(5), 2514-2522. 41. A. Arifin, I. Irwan and B. Abdullah, "Design of sensor water
19. Agarwal, A., Shukla, V., Singh, R., Gehlot, A., & Garg, V. (2018). turbidity based on polymer optical fiber," 2017 International
Design and development of air and water pollution quality
monitoring using IoT and quadcopter. In Intelligent Seminar on Sensors, Instrumentation, Measurement, and
Communication, Control and Devices: Proceedings of ICICCD Metrology (ISSIMM), 2017.
2017 (pp. 485-492). Springer Singapore. 42. M. F. A. Rahman, A. H. A. Samah and S. Z. Yahaya, "Performance
20. Rose, K., Eldridge, S., & Chapin, L. (2015). The internet of things: evaluation of LED Based sensor for water turbidity measurement," 12th
An overview. The internet society (ISOC), 80, 1-50. International Conference on Sensing Technology (ICST), pp. 20-24,
21. Wai, K. P., Chia, M. Y., Koo, C. H., Huang, Y. F., & Chong, W. C. 2018.
(2022). Applications of deep learning in water quality 43. R. Wagner, S. Krüger and J. Bumberger, "Mobile Monitoring-
management: A state-of-the-art review. Journal of Hydrology, Open-Source Based Optical Sensor System for Service-
128332. Oriented Turbidity and Dissolved Organic Matter Monitoring,"
22. P. M, M. J. P. R., and P. V., "The Real-Time Monitoring of Water Frontiers in Earth Science, p. 184, 2019.
Quality in IoT Environment," International Journal of Innovative
Research in Science, Engineering, and Technology, vol. 5, no. 3, 44. A. G.-M. Ferrar, P. Carrington, and S. J. Rowley-Neale, "Recent
pp. 1-6, 2016. advances in portable heavy metal electrochemical sensing
23. Roy, A., Mukhopadhyay, S., & Roy, S. (2022, September). IoT platforms," Environmental Science: Water Research and
Based Water Quality Monitoring System. In 2022 Second Technology, no. 10, 2020.
International Conference on Computer Science, Engineering and 45. P. K. Weber-Scannell and L. Duffy, "Effects of Total Dissolved Solids
Applications (ICCSEA) (pp. 1-4). IEEE. on Aquatic Organisms: A Review of Literature and Recommendation for
24. Geetha, S., & Gouthami, S. J. S. W. (2016). Internet of things Salmonid Species," American Journal of Environmental Sciences, 2007.
enabled real time water quality monitoring system. Smart 46. Y. Ahsan, A. W. Qurashi and R. Yasmeen, "Desalination of Saline
Water, 2(1), 1-19. Water: A Review," The Journal of Zoology, vol. 3, no. 1, pp. 1-5, 2022.
25. Sun, H., & He, Y. (2017, December). Research and Application of 47. G. M. e. Silva, D. F. Campos and J. A. T. Brasil, "Advances in
Water Quality Evaluation of a Certain Section of Yangtze River Technological Research for Online and Situ Water Quality Monitoring—
Based on Fuzzy Neural Network. In 2017 International A Review," Department of Hydraulics Engineering and Sanitation, pp. 1-
Conference on Industrial Informatics-Computing Technology, 28, 2022.
Intelligent Technology, Industrial Information Integration 48. K. Murphy, B. Heery, and D. Zhang, "A low-cost autonomous optical
(ICIICII) (pp. 301-304). IEEE. sensor for water quality monitoring," ResearchGate, pp. 520-527, 2014.
26. Tace, Y., Tabaa, M., Elfilali, S., Leghris, C., Bensag, H., & 49. D. S. Simbeye and S. F. Yang, "Water Quality Monitoring and Control
Renault, E. (2022). Smart irrigation system based on IoT and for Aquaculture Based on Wireless Sensor Networks," Journal of
machine learning. Energy Reports, 8, 1025-1036. Networks, pp. 840-849, 2014.
27. Wang, J., Raza, A., Hu, Y., Buttar, N. A., Shoaib, M., Saber, K., ... 50. K. Saravanan, E. Anusuya and R. Kumar, "Real-time water quality
& Ray, R. L. (2022). Development of monthly reference monitoring using Internet of Things in SCADA," Environment
evapotranspiration machine learning models and mapping of Monitoring Assessment, p. 190, 2018.
Pakistan—A comparative study. Water, 14(10), 1666. 51. H. R. S. Fowzia Akhter*, M. E. E. Alahi and S. C. Mukhopadhyay, "An
28. Rokade, A., Singh, M., Arora, S. K., & Nizeyimana, E. (2022). IOT- IoT-enabled portable sensing system with MWCNTs/PDMS sensor for
Based Medical Informatics Farming System with Predictive Data nitrate detection in water," Measurement: Journal of the International
Analytics Using Supervised Machine Learning Measurement Confederation, pp. 1-10, 2021.
Algorithms. Computational and Mathematical Methods in 52. Adnan, M., Wang, Q., Sohu, N., Du, S., He, H., Peng, Z., Liu, Z., Zhang,
Medicine, 2022. X. and Bai, C., 2023. DFT Investigation of the Structural, Electronic, and
29. Nikoo, M. R., & Mahjouri, N. (2013). Water quality zoning using Optical Properties of AsTi (B i)-Phase ZnO under Pressure for
probabilistic support vector machines and self-organizing Optoelectronic Applications. Materials, 16(21), p.6981.
maps. Water resources management, 27, 2577-2594. 53. Amur, Z.H., Hooi, Y.K., Soomro, G.M., Bhanbhro, H., Karyem, S. and
30. Heddam, S. (2017). Generalized regression neural network based Sohu, N., 2023. Unlocking the Potential of Keyword Extraction: The
approach as a new tool for predicting Total Dissolved Gas (TDG) Need for Access to High-Quality Datasets. Applied Sciences, 13(12),
downstream of spillways of dams: a case study of Columbia river p.7228.
basin dams, USA. Environmental Processes, 4(1), 235-253. 54. Sohu, N., Zardari, N.A., Rahu, M.A., Mirani, A.A. and Phulpoto, N.H.,
31. Childs, P. R., Greenwood, J. R., & Long, C. A. (2000). Review of 2019. Spectrum Sensing in ISM Band Using Cognitive Radio. Quaid-E-
temperature measurement. Review of scientific Awam University Research Journal of Engineering, Science &
instruments, 71(8), 2959-2978. Technology, Nawabshah. 17(01), pp.21-27.
32. Wu, Z., Wang, J., Bian, C., Tong, J., & Xia, S. (2020). A MEMS-
based multi-parameter integrated chip and its portable system for
water quality detection. Micromachines, 11(1), 63.
33. Alam, A. U., Clyne, D., & Deen, M. J. (2021). A low-cost multi-
parameter water quality monitoring system. Sensors, 21(11),
3775.
34. Simić, M., Stojanović, G. M., Manjakkal, L., & Zaraska, K. (2016,
November). Multi-sensor system for remote environmental (air
and water) quality monitoring. In 2016 24th telecommunications
forum (TELFOR) (pp. 1-4). IEEE.
35. Lynch, J. P., Huang, H., Sohn, H., & Wang, K. W. (2019). Sensors
and Smart Structures Technologies for Civil, Mechanical, and
Aerospace Systems 2019. In Proc. of SPIE Vol (Vol. 10970, pp.
1097001-1).
36. S. Dutta, D. Sarma and P. Nath, "Ground and river water quality
monitoring using a smartphone-based pH sensor," AIP
Advances, vol. 5, no. 5, 2015.
37. A. U. Alam, D. Clyne and M. Deen, "A Low-Cost Multi-Parameter
Water Quality Monitoring System," Sensors, pp. 1-4, 2021.
38. M. A. Hossain, J. Canning, S. Ast, P. J. Rutledge, and A.
Jamalipour, "Early warning smartphone diagnostics for water
security and analysis using real-time pH mapping," Photonic
Sensors, pp. 289-297, 2015.
39. V. D. Silva, R. C. d. Freitas and B. Janegitz, "Microfluidic paper-
based device integrated with a smartphone for point-of-use
colorimetric monitoring of water quality index," Measurement,
p. 164, 2020.
40. A. A. Azman, M. Rahman, and M. Ali, "A low-cost nephelometric
turbidity sensor for continual domestic water quality

Page | 19

View publication stats