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

net/publication/338739242

Smart Architecture For Retailing System Using Loe Technique Cloud Computing

Article  in  International Journal of Scientific & Technology Research · January 2020

CITATION READS
1 443

2 authors:

Kaleeswari Chinnakkaruppan K.Kuppusamy Krishnamoorthy

Alagappa University 5 PUBLICATIONS   5 CITATIONS   
7 PUBLICATIONS   25 CITATIONS   
SEE PROFILE
SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Data Security View project

All content following this page was uploaded by Kaleeswari Chinnakkaruppan on 19 February 2020.

The user has requested enhancement of the downloaded file.

 INTERNATIONAL JOURNAL OF SCIENTIFIC & TECHNOLOGY RESEARCH VOLUME 8, ISSUE 11, NOVEMBER 2019 ISSN 2277-8616

Smart Architecture For Retailing System Using

Loe Technique Cloud Computing
C. Kaleeswari, Dr. K. Kuppusamy

Abstract: Internet of Things (IoT) is a recent technology, a vast range of smart applications are deployed in various platforms to utilize the quality of day
– to – day life. Community based Smart Retail is one of the important IoT application. Cloud acts like a server in this research work that offers easy to
access the data via the network connectivity with on – demand. In a recent era, to maintain the detailed list of goods in inventory management is so
complicated. The stock owner and the clients can faced the problems of incrementing sales, cost reduction, goods details, offers, searching a product to
spend a long time in a store. To solve these kinds of complexities, we propose a new scheme for Smart retailing System. Really, unification of
Location of Everything Techniques and Cloud Computing can make a smart service application for IoT. With the help of this paper admin can easily
maintain the inventory control and also clients can easily see the goods details like cost, offers, notification of the new product, location of the product,
product availability, etc. To use the cloud storage, the users can access the data and store the data in a secure environment.

Keywords : Internet of Things, Cloud Computing, Smart retail, LoE (Location of Everything), K-means Clustering.
————————————————————

1. INTRODUCTION range of software’s from simple websites from complicated

Applications of IoT and Cloud are increasing up-to-date. IoT applications [5]. Its applications are fatly deployed in new
can reach the real-time application areas where the humans or business, reduce the risks in innovative business, Global scale
others not able to reach, improves the need of sensors. Cloud and reach. In retails have some challenges like network
is the most secure paradigm for storage. Some of the widely extensibility, storage process of data, confidentiality, diversity,
used IoT areas are Smart Community, Smart Transport, and clients and employees attitude towards using almost using IoT
Smart grid and so on [1]. IoT is a generic network that includes technology. This paper proposes a new scheme for Smart
actuators, simulators, sensors, Smart devices and internet Retail for the management of inventory control in Cloud
connectivity that are used to collect these things and maintain assisted with IoT. One of the major risks is security in Cloud-
the data stored in the cloud for the security purpose [2]. The linked IoT. Motivated by an incrimination of security threats,
deployment of sensors is in any application area of IoT to data loss, IoT manufactures errors, Cloud Service Providers
acquire the data from the environment and forwards the (CSP) and researchers are spending more time with design
collected information to the Cloud. Humans can use the secure systems and develop the recent technology applied in
collected data for further processes. In a recent era of Smart real-time applications like Smart Retailing.
Community, Humans modern life and their associated services
such as retailing is structured, unstructured and semi RESEARCH CONTRIBUTION:
structured data strongly by usage of available resources from The main objective of this paper is to design a Location –
various sources [3]. Indeed, Location of Everything is the Based real time Retailing System to give a capability of IoT
major particle in IoT service where the activity is used to find based modern application service for clients and stock owners
the location of any object in the Smart Retail. According to the by using various components. This research work also
statistical analysis of integration of Cloud and IoT connected includes a localization that is used to discover the current
devices suddenly increasing around the world. Along with the location of the product in a store. This paper organised as
rapid development of these type of systems , Security-attacks follows: Section II presents related work. The Location –
will also increased and pretend a most serious security threats Based Retailing System Architecture has been described in
and privacy mining then even before [4]. Moreover, as the IoT Section III. Proposed research work Experimental Evaluation
has the wide range of usage in various real-time applications, and Comparative analysis discussed in Section IV. Finally,
intruders can endanger public and private security. conclusion is presented in Section V.
_________________________________
2. RELATED WORKS
 C.Kaleeswari is working as a project fellow in Department of Mohammad Wazid et al. [1] designed a new scheme for
Computational Logistics in Alagappa University, Karaikudi, India. HIoTNs, called UAKMP (User Authenticated Key Management
E-mail: [email protected] Protocol). Three type of authentication performed in this
 Dr.K. Kuppusamy is working as Professor and Head of the research work. They are: Smart Card, Password and
Department of Computational Logistics, Alagappa University,
Karaikudi, India. E-mail: [email protected]
Biometrics. The authors used the AVISPA tool to check the
UAKMP is secure. A. Al-Fuqaha et al. [2] Presented a
horizontal overview of relevant protocols are in IoT and also
this paper provides the IoT challenges, emerging technologies
like big data, cloud and fog computing. Javad Rezazadeh et al
[3], developed a smart application for shopping system. In this
system includes four components such as Data gathering
component, data percolating component, data extraction
Cloud environment gives product developers to construct a component and LoE (Location of Everything) component.
This research work provides the better results compared with
other localization mechanisms. Wei Zhou et al [4], proposed
the basic concept of IoT features and security and secrecy
650
IJSTR©2019
www.ijstr.org
 INTERNATIONAL JOURNAL OF SCIENTIFIC & TECHNOLOGY RESEARCH VOLUME 8, ISSUE 11, NOVEMBER 2019 ISSN 2277-8616

impacts of eight IoT new features. Dr. M. Sughasiny [5],

suggested a model cooperation key processing for providing User
data process and seclusion in in Cloud – linked IoT using
Elliptical curve cryptography (ECC) and Hash Map. The author Retail App Wi-Fi with Arduino
LoE
describes the services of security and privacy via the key – PIR sensor Raspberry pi
Gateway
based encryption and decryption in real – time applications. Authentication
Santhosh H.Kalange [7], offered a smart retailing based on
IoT. The author using the RFID technology with arduino Admin Section
Product Details
programming for the inventory management. One drawback of
this research work is all user’s android phones are haven’t to Cloud Day-to-Day
Updating
read the capacity of RFID tags. B.G.J. Ten Bok [6], provides Offers
how IoT permits retailers to discover their applications in
innovative ways by seeing at the key drivers of smart Remarks
environment in the retail business. The experimental results
given are in format of Micro – Segmentation. It offers the real –
time analytics, scalability, privacy and safety. The experts Figure 1: System Architecture
group investigates this research work that plays a minor role
only. Ming – Chang lee [8], introduced the PMA (Privacy LoE can show a leading aspect of Location – Based System
Mining Approach) to extract the percolate from smart home in Iota services. This component gives product details in a
applications by managing a sequence of subtractions and store, location of goods, employees, clients and any physical
statistical analytics on sensor data collections produced by objects that are stored in the cloud and the admin can track
smart homes. Outcome of this research work is to decrease and the user can extract all products based on the LoE
the global sensor topology for smart homes and provide the component. Cloud collects the user’s data as well as admin
data privacy and secure mining. Majid Al – Ruithe [9], has data. The client’s data (location of the product) is collected
presented the vast conversations of the recent events, roles, from their Android phone and sensors with IoT connected
liabilities and services are the key supports of cloud assisted in objects. Along with the admin retail based appropriate data like
the IoT. It also gives a generic architecture for data stock details, goods, offer notifications, remarks, etc.) Are
governance, secrecy and produces a way for future stored in the cloud and reviewed by the admin system up-to-
enhancement process. Dylan Hicks[10], implemented a date. Wi-Fi is used for internet connectivity both the clients’
practicable study that leverages the IoT technology to store android phone and admin system. ST758 PIR Microwave
products “smart”. An open – Source Android application is Motion Sensor Switch is fixed all indoors in the retail store.
developed and used by the clients in this research work. This sensor is fit up into lights to sensing the location of the
product. There are two things to find the locations of the goods
3. PROPOSED SYSTEM such as LoE and the Sensor. These are all used for much
better than retailing process.
3.1 A SMART ARCHITECTURE FOR RETAIL SYSTEM
Arduino
Smart Retailing applications and their quality are most ST758 Sensor
Preprocessin GA with
PIR Data
important for both Stock owners and Clients. The proposed Raspberry Pi
Gateway
scheme is based on smart IoT devices such as Indoor
Inductive Elevator Door sensor ST758 PIR Microwave Motion
Sensor Switch Small Motion Sensor, Android phone with retail User Authentication Access the Clou
app and the cloud infrastructure. In this scheme, each client data d
has their own profile that can easily maintain it with any time at
anywhere. The system offers product details that are available Figure 2 : Block Diagram of SRS
at the stock. So, the customers can select their wanted list of
items using their username and password to login for order the 3.2 HARDWARE REQUIREMENTS
products. Smart Retail System (SRS) added one new feature
of remarks section. If any complaints, suggestions used to 3.2.1. PIR Sensor:
upload in this page. It will be processed day-to-day by the
admin section. The Client’s information’s, admin details,
inventory management details are securely stored in the
Cloud. The leading factors of the proposed system activity are
Location- Based System, Data Privacy Component, IoT
components that are presented in this research work.

Figure 3 : Indoor Inductive Elevator Door sensors ST758 PIR

This item is a new power saving switch. It associates
microwave sensor cast with high – frequency electro –
magnetic wave, IC (Integrated Chip). It involves in security,
ease, self moving ability, power saving and real – time
functions. The vast investigation field id composing of

651
IJSTR©2019
www.ijstr.org
 INTERNATIONAL JOURNAL OF SCIENTIFIC & TECHNOLOGY RESEARCH VOLUME 8, ISSUE 11, NOVEMBER 2019 ISSN 2277-8616

detectors. It works by accepting human needs a location of Preprocessing:

something. It has an easy installation process. Detection is Preprocessing is one of the data extraction process that
applied via the doors, panes of glass or thin walls and lights. includes changing crude data into a justifiable organize. Real-
Fetch the sensor with the lamp and energy source according world data is regularly inadequate, conflicting, and/or missing
to the wire – connection. Finally, on the power switch and in certain behaviours or patterns, and is likely to contain
check whether it is sensing or not. numerous mistakes. Data pre-processing may be a
demonstrated strategy of settling such issues.
3.2.2 Arduino: Pre-processing Steps
Arduino is an easily available electronics product used with Step 1: Import the libraries
integration of hardware and software. This circuit board Step 2: Import the data-set
connected with some smart devices able to execute Step 3: explore the missing values
something innovative in online. It has ease of access handling Step 4: See the explicit Values
by the user experience and also it has been applied in a lot of Step 5: Cacophonous the data-set into coaching and watch
various projects and real – time application. It provides and show the dataset.
scalability for advanced users. It executes on MAC OS, Step half dozen: Feature Scaling
windows and Linux. It offers some advantages: low – cost, low
– power, wireless transformation, long battery life time
capacity and efficiency. Sensors are used to gathers the data
from the platform continuously shifted to another
transformation.

Figure 5 : Pre-processed Dataset

Figure 4 : Arduino UNO Board 4. EXPERIMENTAL EVALUATION

3.2.3Raspberry Pi Gateway: 4.1 Genetic Algorithm (Evolutionary Approach):

The Evolutionary algorithm is an approach for clarifying both
restricted and depraved optimization problems that is
established on instinctive selection, the operation that rides
organic evolution. This algorithm frequently revises a
population of separate outcomes. At every step, it choose
individuals at arbitrary from the recent population to be parents
and utilize them to deliver the siblings for the later genesis.
Over consecutive generations, the populace "derives" toward
Figure 5 : Industrial IoT Gateway an excellent result. Apply this mechanism to clarify different
types of optimization problems that are not applicable for
It controls the bidirectional data bus among various protocols standard optimization algorithms, inclusive problems in which
and networks. Another program of gateway is to transmit the objective function is unstable, homogeneous, problematic,
variety of internet protocols and network. It connects the or hugely nonlinear.
physical devices and sensors. It pre - processes the data The genetic algorithm uses three major categories of
collected from the sensors for the next transmission stage. It regulations at each and every step to build the later
also gives level of security for network and transmitting data in generation from the recent populace:
the cloud environment.  Selection constraints choose the separations, called
Root, that devote to the populace at the later
3.3 COMPONENTS OF THE PROPOSED WORK generation.
 Crossover constraints associate two root nodes to
3.3.1 Data Mining Components: develop leaf nodes for the later generation.
Data mining could be a method utilized by firms to show  Mutation constraints apply frequent modifications to
information into helpful information. By mistreatment code to separate root nodes to develop child nodes.
seem for patterns in giant batches of information, businesses In this Paper is to overcome the drawbacks of K–
will learn a lot of concerning their customers to develop means Clustering for implementing some basic border
simpler promoting methods, increase sales and reduce prices. restraints on the k values and the Genetic Algorithm is applied
Data processing depends on effective information assortment, to diminish the cross-validation error. Genetic Algorithm is also
storage, and laptop process. Extract the data from the sensors used for filtering and processing the data.
that are classified using the Naïve Bayes Classification GA Fitness Calculation:
Algorithm.
652
IJSTR©2019
www.ijstr.org
 INTERNATIONAL JOURNAL OF SCIENTIFIC & TECHNOLOGY RESEARCH VOLUME 8, ISSUE 11, NOVEMBER 2019 ISSN 2277-8616

Binary of 110111 is the genotype and 55 is the phenotype Table. 1 : displays the Error Comparison and Improvement
depiction. After depiction every chromosome the authority path Evaluation in terms of various statistical measures of the
to distribute to search the space, later is to forecast the fitness proposed and existing system in the literature.
value of each separation. Consider that the fitness function
utilized in the example below: Error APT WCL Orbital SRS %Improvement
f(x)=4x+4
Where x is the chromosome value Average 1.88 2.18 1.42 0.91 38.7%

Then the value of the fitness in preceding chromosome is: Variance 1.43 2.06 0.49 0.23 81.3%
f (55)=4(55)+4=229
The process of determining the value of the fitness is a Max 4.11 4.55 2.96 1.72 37.5%
chromosome is called fitness calculation.
GA is used in this research work for reduce the cross- Min 0.20 0.35 0.23 0.18 37.2%
validation error to get an optimal results in clustering.

5. EXPERIMENTAL RESULTS 6. CONCLUSION

In this section visualize the Fitness calculation using Genectic A Smart Retailing System is suggested in this paper. This
algorithm results representation of fitness evaluation using scheme automatically controls the retail which is very capable
Python 3.7 for 32-bit OS. to stockowner in industry statistical analysis of data
representation. LoE is the most believable technique for
developing IoT applications and services. Cloud is a storage
platform for the collected data by sensor is so securely. SRS
model estimates the location of the mobile objects. The
experimental results have been examined and evaluate the
performance of the SRS scheme. The designed scheme could
achieve 38.7% higher precision than APT, WCL and Orbital
methods.

ACKNOWLEDGEMENT
This research work has been written with the financial support
of Rashtriya Uchchatar Shiksha Abhiyan (RUSA – Phase 2.0)
Figure 6 : Python implementation for GA grant sanctioned vide Letter No. F.24-51/2014-U, Policy
(TNMulti-Gen), Dept. of Edn. Govt. of India, Dt. 09.10.2018.
In this section also presents an average of localization error
comparison of Existing methodologies with table
representation. Sensor positioning in indoors and lights
REFERENCES
[1] Mohammed Wazid, Ashok kumar Das, Neeraj kumar,
accuracy experiments are reported in this Diagram. Our
Mauro conti, MinhoJo.(2017). Design of Secure User
proposed scheme could successfully extract the accuracy of
Authenticated Key Management Protocol for Generic IoT
user’s information, admin details and goods details in
Networks. IEEE Internet of Things Journal.
inventory management. The Fig. 5 shows the proposed
[2] A. Al-Fuqaha, M. Guizani, M. Mohammadi, M. Aledhari,
system standard deviation of localization error compared with
and M. Ayyash. (2015). “Internet of Things: A survey on
other mechanisms exists in the literature. This figure is plotted
enabling Technologies, protocols, and
by using some statistical analysis of other existing localization
applications.In IEEE communications surveys tutorials,
techniques.
Vol. 17, no. 4, pp. 2347-2376, Fourthquarter.
[3] Javad Rezazadeh, Kumbesan Sandrasegaran, Xiaoying
Error Comparision kong,. A Location-Based Smart Shopping System with
IoT Technology.
[4] Wei Zhou, Yuqing Zhang, and Peng Liu. (2018, January
31). The Effect of IoT New Features on
[5] Security and Privacy: New Threats, Existing solutions,
and Challenges yet to be solved.
[6] Dr. M. Sughasiny. (2017, May). Enhanced Security
Framework for Cloud-Linked IoT. International Journal of
Advances in Scientific Research and Engineering. Vol.
03, Issue 4, ISSN: 2454-8006.
[7] B.G. J. ten Bok. Innovating the Retail Industry; an IoT
approach . (2016, July). In 7th IBA Bachelor thesis
Conference.
[8] Santosh H. Kalange, Dipti A. Kadam, Asmita B.Mokal,
Figure 7 : Average Localization Error Comparisons Avinash A. Patil. (2017, November). International
Research Journal of Engineering and Technology.
Volume: 04, Issue: 11, e-ISSN: 2395-0056, p-ISSN:
2395-0072.
653
IJSTR©2019
www.ijstr.org
View publication stats

INTERNATIONAL JOURNAL OF SCIENTIFIC & TECHNOLOGY RESEARCH VOLUME 8, ISSUE 11, NOVEMBER 2019 ISSN 2277-8616

[9] Ming-Chang Lee, Jia-Chun Lin, Olaf Owe. (2018, [11] Dylan Hicks, Kevin Mannix, Hannah M. Bowles, Byron J.
September). Privacy Mining from IoT-based Smart Gao. (2013, October). SmartMart: IoT-based In-store
Homes. 13th International Conference on Broad-Band Mapping for Mobile Devices.
Wireless Computing, Communication and Applications.
[10] Majid Ai-Ruithe, Siyakha Mthunzi, Elhadj Benkhelifa.
(2016). Data Governance for Security in IoT & Cloud
Converged Environments. In IEEE.

654
IJSTR©2019
www.ijstr.org