Smart Placement of Routers in Agricultural Crop Areas

P. H. G. Coelho, J. F. M. Amaral, K. P. Guimarães, E. N. Rocha and T. S. Souza

State Univ. of Rio de Janeiro, FEN/DETEL, R. S. Francisco Xavier, 524/Sala 5001E, Maracanã, RJ, 20550-900, Brazil

thaynans.souza@gmail.com

Keywords: Genetic Algorithms, Fuzzy Systems, Precision Agriculture, Artificial Intelligence Applications.

Abstract: The utilization of technologies in agriculture, which is called precision agriculture, is progressively necessary

to optimize crop yields. The purpose of this paper is to present an optimized positioning of routers, seeking a

robust topology of the network, in order to cover the sensor monitoring devices spread in an agricultural crop

area, sending data such as temperature, soil humidity, incidence of luminosity, etc., which allows the farmer

to make better decisions regarding the cultivation of his/her land. For this, genetic algorithms will be used to

determine the location of routers in a network through evolutionary techniques associated with a fuzzy

aggregation method. Typical application cases are presented and discussed to illustrate the proposal.

1 INTRODUCTION

Family farming is responsible for producing about

70% of the food that is available for consumption by

the Brazilian population. It is made up of small rural

producers, traditional peoples and communities,

foresters, aqua culturists, extractivists and fishermen.

The sector stands out for the production of corn,

manioc root, vegetables, beans, sugarcane, rice,

swine, poultry, coffee, wheat, castor beans, fruit and

vegetables. In family farming, the management of the

property is shared by the family and the agricultural

production activity is the main source of income. In

addition, the family farmer has a particular

relationship with the land, his place of work and

residence. Productive diversity is also a striking

feature of this sector, as it often combines subsistence

production with production destined for the market.

Family farming contributes to the generation of

income and employment in the countryside and also

improves the level of sustainability of activities in the

agricultural sector.

Nowadays, the utilization of technologies in

agriculture, which is called precision agriculture, is

progressively necessary to optimize crop yields.

Challenges include effects of climate change to the

agricultural systems which can drastically reduce

agricultural productivity. In this direction, a recent

work in (Del Felipe et al., 2022) proposes a network

of wireless sensors applied to precision agriculture.

The sensor network collects data on environmental

variables such as soil moisture, temperature, and

ambient humidity. This data is sent wirelessly to a

head node responsible for uploading the information

to an Internet of Things (IoT) platform. Another

research paper (Dasig, 2020) shows recent advances

in Agriculture 4.0. Business models are discussed in

(Medici et al., 2021) to identify possible agribusiness

models for developed and developing countries,

particularly for the European context and sub-

Saharan Africa and South Asia-Pacific area. Smart

IoT based approach is proposed in (Sreeja et al.,

2020) which consists of three sensors, a Wi-Fi

module and a dc motor, to measure and display the

different parameters for the crop.

Technological advances have had a great impact

on Brazilian agriculture; however, family farming is

still taking slow steps in direction. Therefore, it is

vitally important to implement technological

technical support resources in family farming, relying

on government funding and making it possible to

increase production efficiency, which leads to an

increase in production and, consequently, its gains.

The purpose of this article is to optimize the

positioning of routers, seeking a robust topology of

the network, in order to cover the sensor monitoring

devices spread in an agricultural crop area, sending

data such as temperature, soil humidity, etc., which

allow the farmer to make better decisions regarding

the cultivation of his/her land. However, in practice,

the purely random positioning of router nodes can

result in poor communication performance with the

Coelho, P., Amaral, J., Guimarães, K., Rocha, E. and Souza, T.

Smart Placement of Routers in Agricultural Crop Areas.

DOI: 10.5220/0011744800003467

In Proceedings of the 25th International Conference on Enterprise Information Systems (ICEIS 2023) - Volume 1, pages 99-106

ISBN: 978-989-758-648-4; ISSN: 2184-4992

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

sensor monitoring devices. In addition, the actual

deployment may need to consider restrictions and

geographic characteristics of the area in question,

making it necessary to search for different topologies

to distribute them. In fact, layout is a critical aspect in

mesh wireless networks. As such a problem is of the

NP hard type, one motivation to solve the mesh router

placement problem and look for the optimal solution

with adequate performance is to follow an approach

using evolutionary techniques that involve genetic

algorithms, including fuzzy aggregation. A wireless

mesh network (WMN) can be understood as a

communication network composed of radio links

planned in a mesh topology. There are two types of

nodes in WMNs: routers and clients. The group of

mesh routers, connecting to each other, forms the

backbone for the set of clients that aggregate the

mesh. Some mesh routers act as Internet gateways to

intermediate data traffic between the Internet and the

WMN. The low design cost and quick installation of

WMNs makes them a cost-effective choice for

establishing wireless connectivity for mobile users

anytime, anywhere. These characteristics can be

useful mainly in regions or developing countries,

whose decreasing costs of implantation and

maintenance of infrastructure can make possible

automations in several levels, optimizing levels of

production and income of small family farmers. The

good quality and operability of WMNs depends

heavily on placing mesh router nodes in the desired

area to achieve network connectivity, stability and

coverage. The placement of routers in a mesh network

is not a trivial problem. Typically, this is a problem

that can be solved using traditional evolutionary

techniques such as weighted-sum approach genetic

algorithms or Pareto-based techniques. Weighted-

sum evaluation for genetic algorithms leads to

difficult assignment of appropriate weights, while

Pareto techniques require the designer to select the

most suitable solution among the set of presented

solutions.

Several studies using intelligent computational

systems have been carried out by universities and

research centers around the world. In (Girgis et al.,

2014) genetic algorithm and simulated annealing is

used to search for a low-cost network with restrictions

and determine the minimum number of routers. In

(Rezaei et al., 2011) a genetic algorithm coupled with

circle packing techniques is proposed which consists

of positioning non-identical circles without

overlapping within another circle, maximizing

connectivity and coverage of the area network.

Router Nodes Placement Using Artificial Immune

Systems is used in (Coelho et al., 2017), and (Coelho

et al., 2015) for industrial applications. Recently,

coyote optimization algorithm was used to solve the

mesh router nodes placement in wireless mesh

networks (Mekhmoukh Taleb et al., 2022). The

authors claim good results in simulations carried out

for typical scenarios. In this work, we use genetic

algorithms to determine the location of routers in a

mesh network through evolutionary techniques

associated with a fuzzy aggregation method (Coelho

et al., 2019).

This paper is organized into four sections. The

second section deals with modeling the problem

followed by discussion of the case studies in section

three. Finally, section four closes the article with

conclusions.

2 HYBRID FUZZY- GENETIC

PROPOSED MODEL

The model used to solve the mesh router placement

issue was developed considering its multi-objective

aspect. In addition to the full coverage of the

acquisition points (sensor monitoring devices),

restriction zones were proposed in two scenarios,

making it difficult to install the routers. To meet these

requirements, a fuzzy-genetic approach was used,

which proposes a fusion of Genetic Algorithm and

Fuzzy Logic techniques (Coelho et al., 2019).

The genetic algorithm (GA) is inspired by

biological evolution, since it makes use of a selection

of individuals, using genetic operators and operates in

a random and oriented way, seeking an optimal

solution within a population. The main application of

genetic algorithms is to solve optimization problems

with very large or complex search areas, which makes

the use of traditional techniques impractical. In the

case of a search, a comparison is made between the

evolution of the species and the problem in question:

a population of individuals (possible solutions)

identified by chromosomes are evaluated and

associated with an aptitude and subjected to a process

of evolution, through selection. and reproduction, for

several generations. Fitness is the quality of the

individual result compared to the general fitness

transferred (Coelho et al., 2019).

The objective function or fitness is defined based

on the problem specification and is critical for

successful implementation. In general, the objective

function involves only a single criterion. However,

most real problems involve more than one objective

to be considered, so the objective function must use

methods to convert vector quantities into a scalar. The

general parameters of the GA influence its

performance and can be used to establish a stopping

ICEIS 2023 - 25th International Conference on Enterprise Information Systems

100

criterion to run the algorithm. Such parameters

include population size, maximum number of

generations and operator application rate. The choice

of parameters must meet established empirical

criteria or the specific characteristics of the problem.

In order to carry out multi-purpose case studies on

genetic algorithms we can use a fuzzy aggregation

method.

The use of fuzzy systems makes it possible to

simultaneously evaluate all objectives, integrating

user preferences in relation to each objective and for

each situation. This feature offers a good advantage

over multi-objective Pareto optimization methods, as

it does not require user interference to choose the best

solution at the end of the process, since preferences

or specifications are entered before evolution in a

simpler and more interpretable way through fuzzy

logic, so the evolution process is guided in the

direction of pre-established preferences. Each

individual in the GA population represents a possible

solution to the problem. During the evaluation

process, individuals are submitted to the function or

model that describes the problem and the results

obtained in relation to each objective are used as

inputs to the fuzzy system. For each individual in the

population, the fuzzy aggregation method is applied,

producing a single physical fitness value. Figure 1

illustrates the evaluation model using the Fuzzy

Aggregation Method. The selection, crossover,

mutation, population size and maximum number of

generations rates are defined by the designer before

starting the algorithm. The fuzzy aggregator rules are

designed to meet the preferences needed to solve the

problem, considering each objective. Evolution ends

when a certain stopping criterion is reached. The most

frequent stopping criterion is specified by a certain

maximum number of generations. Another possibility

is to establish a fitness value to be reached or stop the

execution of the algorithm when there is no evolution

for a certain number of generations. After evaluating

all individuals of the current generation, the genetic

algorithm continues the evolution process. The fuzzy

aggregation system presents the usual functioning of

a fuzzy inference system. Each system input

corresponds to an objective and membership

functions are triangular or trapezoidal in shape. The

genetic algorithm with the incorporation of the fuzzy

aggregator used in this work follows the model

presented in Figure 2.

Figure 1: Fuzzy evaluation model with aggregator system.

Figure 2: Genetic algorithm and fuzzy aggregator.

3 CASE STUDIES

In this work, three scenarios were considered for case

studies whose objective was positioning low power

battery operated routers for a mesh network to be used

for data acquisition in an agricultural environment of

size 50 m x 50 m. In all of them, the Matlab Fuzzy

Toolbox was applied.

In the first study, the genetic algorithm was used

with the sole objective of positioning the routers so

that each monitoring point in the field was covered by

at least one router.

In the other two studies, it was considered that, in

the same field, there are areas with a higher or

hindering installation cost. Therefore, these areas

were considered not suitable for installing routers. In

this way, the genetic algorithm was used together

with a fuzzy aggregation method to perform a multi-

objective application where it is desired to position

routers so that each monitoring point is in contact

with at least one router installed in a low-cost area.

The specification of data acquisition points as well as

the interconnection between the routers is not part of

the scope of this work.

Smart Placement of Routers in Agricultural Crop Areas

101

3.1 First Case Study

The environment of this first case study is an

agricultural area of 2500 m

, where the distribution of

routers must be carried out in such a way that each

monitoring point is covered by at least one router.

Each soil condition monitoring device consists of a

low power SoC (System On a Chip) IoT

microcontroller, equipped with sensors, battery

operated, capable of sending data within only 13

meters. Sensors such as temperature, soil moisture,

incidence of luminosity and relative humidity. air and

soil pH can be connected to the monitoring device.

The organization adopted for the routers is

homogeneous (all routers have the same

characteristics) and irregular. In order to achieve

these goals a traditional single-target genetic

algorithm is used. The 16 monitoring points are

positioned in the area, as shown in Figure 3. It is a

single-objective problem, with a strictly objective

assessment: to maximize the number of points

covered.

The values of the parameters of the genetic

algorithm are presented in Table 1. Such values led

the algorithm to find a satisfactory solution to this

simple case study.

Figure 4 shows the fitness curve of the best

individuals and the average population.

Figure 5 shows the location of the monitored

points and the positioning of the routers for the best

individual reached by the GA. The green square in

Figure 5 represents the area of the land in question

(50m x 50 m), the smaller blue circles are the

monitoring points and the large blue circles represent

the area that each of the routers is covering, the

routers being represented by the " x" in red.

Figure 3: Location of the 16 monitoring points, Case study 1.

Table 1: GA Parameters for case study 1.

Parameters Values

Generations 200

Search Re

ion X=[-25, 25]; Y=[-25, 25]

ulation 300

Precision 50 cm

itness Number of covered

oints

Selection Geometric Normalization 8%

Crossover rate 80%

Mutation rate 1%

Figure 4: Fitness curve for case study 1.

Figure 5: Routers’ positioning for case study 1.

It can be seen that the routers were positioned

satisfying the established criteria, since each

monitoring point is being covered by a router. This

first study did not take into account differences in the

cost of installing the routers in relation to areas with

more difficult access. Considering that cost is

something that is important to be reduced in most

projects, mesh networks would be no different.

Therefore, the proposal of the second case study is to

ICEIS 2023 - 25th International Conference on Enterprise Information Systems

102

configure the network taking into account the

installation costs.

3.2 Second Case Study

As in the first case study, the environment is an

agricultural area of 2500 m

, where the spatial

distribution of the routers must be carried out. The

monitoring device has a range of 13 m and the

organization of the routers is homogeneous (all

routers have the same characteristic) and irregular.

For this scenario it is necessary that:

- each monitoring point is covered by at least one

router;

- the routers are not positioned in places where the

installation cost is high.

So, we have a multi-objective problem: covering the

area and reducing costs.

To achieve these goals, a genetic algorithm is used

in conjunction with a fuzzy aggregation scheme.

The developed fuzzy system is of the Mamdani

type, characterized by being simpler and more

interpretable than TSK-type systems, and all rules

have the same degree of importance, that is, weights

equal to one.

The fuzzy aggregation system has two inputs:

“NCS” (Number of Covered Sensor monitoring

devices) and "Cost". Both inputs have 3 membership

functions. The output is the Fitness that receives an

evaluation between 0 and 10, and also has 3

membership functions. The system uses the default

Matlab configuration (And method = min ; Or method

= max; Implication = min; Aggregation = max).

The Defuzzification method is the mean of

maxima (MoM).

Figure 6 shows the fuzzy aggregation system used

in this case study.

Figure 6: Fuzzy aggregation system for case study 2.

Figure 7 shows the "NCS" (Number of Covered

Sensor monitoring devices) input membership

functions.

Figure 7: NCS membership functions for case study 2.

Figure 8: Cost membership functions for case study 2.

Figure 9: Fitness membership functions for case study 2.

Smart Placement of Routers in Agricultural Crop Areas

103

Similarly, Figure 8 illustrates the “Cost”

membership functions. Finally, Figure 9 shows the

membership functions for the output (Fitness) of the

fuzzy system.

The Fuzzy Aggregator rules, as shown in Figure

10, are as follows:

1. If (NCS is low) and (Cost is low)

then (Fitness is poor)

2. If (NCS is low) and (Cost is medium)

then (Fitness is poor)

3. If (NCS is low) and (Cost is high)

then (Fitness is poor)

4. If (NCS is medium) and (Cost is low)

then (Fitness is poor)

5. If (NCS is medium) and (Cost is medium)

then (Fitness is poor)

6. If (NCS is medium) and (Cost is high)

then (Fitness is poor)

7. If (NCS is high) and (Cost is low)

then (Fitness is good)

8. If (NCS is high) and (Cost is medium)

then (Fitness is medium)

9. If (NCS is high) and (Cost is high)

then (Fitness is poor)

Figure 10: Fuzzy aggregator rules for case study 2.

It should be noted that the system is quite

interpretable from the 9 rules presented. In other

words, it is important to note that the 9 rules can be

reduced to just 5. Indeed, it is noted that when NCS is

low or medium, the Cost does not matter. No matter

the Cost, Fitness will be poor (2 rules). It only makes

sense to evaluate the Cost when NCS is high (3 rules).

Some tests were performed changing the values of

the parameters of the genetic algorithm and it was

observed that the same parameters of case study 1

also led to good results in case study 2.

Figure 11 shows the area, the location of

monitoring devices, and the regions with restrictions

on the installation of routers, where warehouses

restrict the placement of routers.

Figure 11: Routers’ positioning for case study 2.

Figure 12 shows the curve of the best individual and

the average fitness of the population. From the plots

it can be seen that the best individual reached the

maximum fitness around the 30th generation and the

average followed this evolution.

Figure 12: Fitness curve for case study 2.

The green square in Figure 11 represents the coverage

area (50 m x 50 m), the smaller blue circles are the

monitoring points, the larger blue circles represent the

area that each of the routers is covering. The red x are

the routers and the red squares are the regions of the

area where the cost of installing routers is high. In the

same figure it can be seen that the routers were

positioned according to the established criteria, since

each monitoring point is to be covered by a router and

no router was positioned in the area where the

installation cost is high. It can also be observed that,

with the restriction of positioning the routers in areas

ICEIS 2023 - 25th International Conference on Enterprise Information Systems

104

of higher cost, it was necessary to use one more router

to cover the area, totaling five routers. Such

additional cost of routers should obviously outweigh

installing a router rather than higher cost. The results

obtained in the simulations fulfilled their objective of

determining the positioning of routers in the network.

3.3 Third Case Study

In this case, restrictions closer to real scenarios were

proposed. Based on the same basic setup as the

previous ones, the restrictions in this one are a stream

that crosses the land (where, of course, you don't want

to place sensors, nor is it possible to install routers)

and a 10 m x 15 m tool shed, which also restricts the

installation of the routers. Sensors were repositioned

out of the waterway.

For comparison purposes, the same parameters as

in the previous case were kept in the Fuzzy

aggregator. It can be seen that the optimized

positioning suggested in this case is different from the

previous one, which was in fact expected, given the

changes in the restriction regions and sensor

positioning. It can also be observed that it takes a few

more generations to reach the best individual, given a

greater difficulty in the scenario. Figure 13 shows the

final location of the routers including the area, the

location of monitoring points, and the regions with

restrictions on the installation of the routers, where a

stream and a shed restrict the placement of the

routers. Figure 14 shows the curve of the best

individual and the fitness average of the population.

Figure 13: Routers’ positioning for case study 3.

From the graphs it can be seen that the best individual

reached maximum fitness around the 40th generation

and the average followed this evolution.

In both case studies, the presented fuzzy

aggregator system simultaneously evaluates the two

objectives (maximize NCS - Number of Covered

Sensor monitoring devices and minimize Cost),

integrating designer’s preferences. It offers an

interesting advantage over multi-objective Pareto

optimization methods, since specifications are

entered in a simpler and more interpretable way

through fuzzy logic, and so the evolution is guided to

pre-established preferences.

Figure 14: Fitness curve for case study 3.

4 CONCLUSIONS

In this work, case studies were presented with

applications in mesh networks, whose objective is to

optimize the positioning of routers in a small rural

property to optimize the production of family

agriculture using sensors with automation for data

acquisition. In the scenario of the first case study,

with only one objective, a genetic algorithm

presented satisfactory solutions. In the other case

studies, a hybrid fuzzy-genetic evolutionary

technique was applied to a multi-objective system, in

which the cost objective was included in the routing

issue.

The fuzzy aggregator system makes it possible to

simultaneously evaluate all objectives, integrating

user preferences in relation to each objective and for

each situation. This proposed system offers a good

advantage over multi-objective Pareto optimization

methods, as it does not require user interference to

choose the best solution at the end of the process,

since preferences or specifications are entered before

evolution in a simpler and more interpretable way

through the use of fuzzy logic, so the evolution

Smart Placement of Routers in Agricultural Crop Areas

105

process is guided in the direction of pre-established

preferences.

For future work on precision agriculture, we plan

to run experiments with weighted-sum GA and Pareto

traditional methods for comparison’s sake. It is also

expected the inclusion of new targets for fuzzy

aggregation and possibly the design of a variable size

chromosome in the GA modeling so that evolution

can also determine the number of routers suitable for

field coverage. In addition, comparisons with other

algorithms such as Particle Swarm Optimization

(PSO)(Lin, 2013), Whale Optimization Algorithm

(WOA) (Mirjalili et al., 2016), Bat Algorithm

(BA)(Lin et al., 2014), African Vulture Optimization

Algorithm (AVOA) (Abdollahzadeh et al., 2021) etc.

are foreseen in future works.

ACKNOWLEDGEMENTS

This study was financed in part by the Coordenação

de Aperfeiçoamento de Pessoal de Nível Superior –

Brasil (CAPES) – Finance Code 001, and FAPERJ.

REFERENCES

Del Felipe, M.R., Vázquez, M.L. and Bermello, J.L.P.,

2022.Wireless sensor network applied to precision

agriculture: A technical case study at the technical

university of Manabí. in Communication, smart

technologies and innovation for society. Smart

innovation, systems and technologies, vol 252, Á.

Rocha, P.C. López-López, J.P. Salgado-Guerrero, Eds.

Singapore: Springer.

Dasig, D.D., 2020. Implementing IoT and wireless sensor

networks for precision agriculture, in Internet of things

and analytics for agriculture, vol 2. Studies in big data,

vol 67, P. Pattnaik, R. Kumar, and S. Pal, Eds.

Singapore: Springer.

Medici, M., Carli, G., Tagliaventi, M. R., and Canavari,

M., 2021. Evolutionary scenarios for agricultural

business models, in Bio-economy and agri-production,

D. Bochtis, C. Achillas, G. Banias, and Maria Lampridi,

Eds. Academic Press, 2021, pp. 43-63.

Sreeja, B.P. , Manoj Kumar, S. , Sherubha, P., Sasirekha,

S.P. , 2019. Crop monitoring using wireless sensor

networks, in Materials today, Proceedings.

Girgis, M. R. , Mahmoud, T. M., Abdullatif, A. B. , and

Rabie, A. M., 2014. Solving the wireless mesh network

design problem using genetic algorithm and simulated

annealing optimization method, International Journal

of Computer Applications, vol. 96, no. 11, pp. 1-10.

Rezaei, M., Sarram, M. A., Derhami, V., and Sarvestani, H.

M., 2011. Novel Placement Mesh Router Approach for

Wireless Mesh Network, Proceedings of the

International Conference on Wireless Networks.

Coelho, P. H. G., do Amaral, J. L. M., do Amaral, J. F. M.,

de Arruda Barreira, L F., and Barros, A. V., 2017.

Router nodes placement using artificial immune

systems for wireless sensor industrial networks, in

Lecture Notes in Business Information Processing, vol.

291, Springer International Publishing, pp. 155-172.

Coelho, P. H. G., do Amaral, J. L. M., do Amaral, J. F. M.,

de Arruda Barreira, L F., and Barros, A. V., 2015.

Applying artificial immune systems for deploying

router nodes in wireless networks in the presence of

obstacles, in Lecture Notes in Business Information

Processing, vol. 227, Springer International Publishing,

pp. 167-183.

Mekhmoukh Taleb, S. , Meraihi, Y. , Gabis, A. B., Mirjalili

S., Zaguia A., and Ramdane-Cherif, A. ,2022. Solving

the mesh router nodes placement in wireless mesh

networks using coyote optimization algorithm, in IEEE

Access, vol. 10, 2022, pp. 52744-52759.

Coelho, P., do Amaral, J. M., Guimarães, K., and Bentes,

M., 2019. Layout of routers in mesh networks with

evolutionary techniques, in Proceedings of the 21st

International Conference on Enterprise Information

Systems – vol. 1, pp. 438-445.

Lin, C.-C., 2013. Dynamic router node placement in

wireless mesh networks: A PSO approach with

constriction coefficient and its convergence analysis, in

Inf. Sci., vol. 232, pp. 294-308.

Mirjalili, S. and Lewis, A., 2016. The whale optimization

algorithm, Adv. Eng. Softw., vol. 95, pp. 51-67.

Lin C.-C., Li, Y.-S., and Deng, D.-J., 2014. A bat-inspired

algorithm for router node placement with weighted

clients in wireless mesh networks, in Proc. 9th Int.

Conf. Commun. Netw. China, pp. 139-143.

Abdollahzadeh, B., Gharehchopogh, F. S., and Mirjalili, S.,

2021. African vultures optimization algorithm: A new

nature-inspired Metaheuristic algorithm for global

optimization problems, Comput. Ind. Eng., vol. 158.

ICEIS 2023 - 25th International Conference on Enterprise Information Systems

106