Proceedings of the 2023 International Conference on Machine Learning and Automation

DOI: 10.54254/2755-2721/44/20230067

IoT based security and privacy implementation in smart home

J. Rajasekhar1, T. Thanusha1, G. Naga Jyothi1 and K. Tejaswi1 and Laith

Abualigah2,3,4,5
1
Department of Electronics and Computer Engineering, Koneru Lakhmaiah Education
Foundation, Vaddeswaram, AP, 522502, India;
2
Computer Science Department, Prince Hussein Bin Abdullah Faculty for Information
Technology, Al al-Bayt University, Mafraq 25113, Jordan.
3
MEU Research Unit, Middle East University, Amman, Jordan.
4
Hourani Center for Applied Scientific Research, Al-Ahliyya Amman University,
Amman 19328, Jordan.

5
[email protected]

Abstract. Internet-of-Thing’s technology is being increasingly important in our daily lives. As

IoT technology evolved, IoT devices face a data protection hazard, particularly smart home IoT
gateway devices, which became evident. The demand for a low-cost, secure smart home gateway
device or router among smart home users. The problem is that as the internet of things (IoT)
becomes more ubiquitous, there is a growing need to simplify wireless network control
mechanisms. Because data collecting and the process includes processes such as monitoring,
judging, and controlling are all involved in IoT, the control mechanism is challenging to simplify.
Many internets of things technology offer memory and communication capabilities, and are
easily vulnerable to hacking, due to the mobile software available at the tip of one's fingers to
operate the linked gadgets to the web. In the Internet of Things, secure data transfer is always a
concern. To increase safety in IoT and wireless networks, the current study introduces a unique
RSA-based method, as well as the AES algorithm and the lightweight protocol message queue
telemetry transport (MQTT).

Keywords: IoT, MQTT, security, aes kdsb algorithm.

1. Introduction
Internet technology is becoming increasingly important in people's lives, benefiting individuals of all
ages, from children to the elderly. Different types of apps that mix internet technology will become an
increasingly significant component of enhancing people's lives as innovation and the Web advance. The
Internet of Things (IoT) is indeed a network of interconnected devices. Furthermore, those IoT systems
can be viewed and operated remotely via the web. It allows users to easily access gadgets linked to a
network. A smart house is a collection of several home devices that simplify fundamental home
functions and employ new protection vectors that can be monitored via the web.
The potential of a harmful network attack or criminal behaviour is growing more common as IoT
technology improves and advances. As IoT systems are linked via the web known as the Internet, critical
data will be transmitted via the Internet. As IoT technology improves, crime hackers may attempt to

© 2024 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).

202
 Proceedings of the 2023 International Conference on Machine Learning and Automation
DOI: 10.54254/2755-2721/44/20230067

hack data sent from an IoT gateway, IoT devices and an IoT network can access the internet by
exploiting flaws in gateway devices, Integrated denial of service, hacking and other novel techniques of
attack. IOT consists of a large number of little data blocks that are exchanged between networks from
components such as various types of sensors. There would be a few issues with IoT. Although the
Internet Protocol has been utilized for the majority of communication, TCP/IP or UDP/IP application
protocols currently require Internet access. When the Hypertext Transfer Protocol (HTTP) is used for
IoT connectivity, a huge number of small data blocks are sent, resulting in significant performance
deterioration. Furthermore, IP addressing varies depending on physical location, making network
control challenging. To solve these issues. MQTT is a simple protocol for sharing IoT network resources.
MQTT eliminates protocol overheads and allows for high-speed IoT connectivity.

2. Literature survey
This paper illustrates using the Raspberry Pi to secure wireless home automation. We use sensors for
security purposes such as door theft protection and gas detection [1]. In addition, the user can control
the appliances in the home via the internet via a mobile application from any location.[2] Gas leakage,
temperature and humidity, and the on/off of lights and fans in the house are all monitored and managed
in the proposed methodology. An alarm notification will be sent to our Gmail account if a gas leak is
detected in the residence.[3] Using PIR, LDR, and DHT11 sensors, this proposed system cost-effectively
manages security-related issues. This sensor recognizes the face, captures the process with the Pi camera,
and sends an email notification in a couple of moments.[4] An automated self-regulating mechanism is
recommended that monitors the ambient temperature automatically Using the fan speed control system,
the heater, and the keypad, adjust the fan speed depending on the current room temperature and
predetermined preference settings. the heater, and the keypad. For the central control system, a
Raspberry Pi is employed.[5] The air conditioner will detect and adjust the temperature by sensing the
number of people in the room using sensors and cameras. If the number of people in the room increases,
the temperature will automatically raise or decrease depending on the number of people in the room also
the data will be saved in the cloud.[6] In kitchen appliances, the gas sensor will detect the presence of
gas and send an alert to our Gmail account; it can also send an alert without Wi-Fi if the data is stored
in the cloud.[7] The authors have demonstrated a basic application of Raspberry Pi in smart things
control via the Internet (E-mail) in a Raspberry Pi-based interactive home automation system via Email,
in which the topic of the obtained e-mail is read by the proposed methodology and the process reacts to
the respective guidelines.[8] Home automation systems are built to automate operations such as remote
control of home appliances. Wireless Sensor and Actuators Networks (WSANs) are becoming
increasingly popular in-home automation

3. Proposed work
In the proposed system we are using different MCU to monitor and control the room conditions and all
this information is sent to the Raspberry pi and from there the data will be sent to the cloud Fig.1.shows
the block diagram of the kitchen room in the kitchen room we are using the gas sensor MQ-6 to find
any gas leakages in the room and notify the same to the cloud and automated to open the windows and
door of kitchen as well as a notification to the mail The room temperature was measured using a DHT11
sensor. The DHT11 is a widely used temperature and humidity sensor for prototypes that monitor a
specified area's ambient temperature and humidity. When the room temperature is raised, the
information is automatically transferred to the cloud, and the kitchen exhaust fan is turned on. The
exhaust fan will turn off automatically if the room temperature drops to a low level. The kitchen
appliances will be controlled and monitored by the cloud (such as lights and fan, fridge ON and OFF),
and all information or data will be stored in the cloud.

203
 Proceedings of the 2023 International Conference on Machine Learning and Automation
DOI: 10.54254/2755-2721/44/20230067

Figure 1. Block diagram of kitchen. Figure 2. Block diagram of Bed Room.

In the bedrooms and main hall, we use PIR, LDR, and DHT11 sensors as shown in Fig.2. The PIR
sensors detect infrared energy emitted by objects in their area of vision. Because the human body is the
most prevalent thing that a PIR sensor detects, these sensors are used in automatic light switches. As
indicated in Fig.2, a PIR sensor will be linked to the server. This sensor is used to control equipment
such as fans and air conditioners. If the number of people in the room increase, the AC temperature will
rise or the fan speed will increase. If the number of people is reduced, the temperature will be reduced
in accordance with the room temperature. The temperature will be controlled automatically when the
PIR sensor detects the movement of a person. An LDR is a component with a resistance that varies in
response to the amount of light it receives. When light falls on the LDR, the resistance reduces, while
in the dark, it increases. When an LDR is kept in the dark, it has a high resistance, but when it is
maintained in the light, it has a lower resistance. The lights in the room will be controlled by the LDR
sensor. If the light resistance is low, the lights will turn on automatically, conversely, if the resistance is
large, the lights will turn off automatically. When a person enters the room, the PIR sensor detects their
movement, and the light is turned on with the help of the LDR sensor and wisely, the lights will be
turned off when the person has left the room. The relays are used to control room appliances. All of the
sensors will be linked to the server, and the data will be saved in the cloud. In all other rooms, the same
sensors and processes will be used. The Raspberry Pi will be used to connect all of the servers, and the
information from the servers will be kept in the cloud.

Figure 3. Block Diagram Server.

Fig.3 This system will detect the presence of an intruder and quickly send an email notification to
the user. In addition, a photo of the intruder captured with the Pi camera will be sent in this email. A
Raspberry Pi is in charge of the entire setup. This system can be installed at the front door of your home
and you can monitor it by email from anywhere in the World. Here, we use cameras to automatically
open the door. They can gain entry to the house by storing the known person database in the Raspberry
pi SD card, the camera will take a picture of that person and send it to the Raspberry pi and from there
It will take a glance through the database. The door would be opened if a picture is found in the database.
If an unknown person attempts to open the door, the camera will take a picture of that person and search
in the database if it is not identified the image an alert warning will be sent to the user's email address.
The entire data will be preserved on the cloud. We store data in the cloud using Wi-Fi, however, once
the data is saved in the cloud, the appliances or sensors can work without Wi-Fi. The MCUs in Figure
1&2 are also connected to the server Raspberry Pi. From the sensor data will be sent to the cloud and

204
 Proceedings of the 2023 International Conference on Machine Learning and Automation
DOI: 10.54254/2755-2721/44/20230067

those MCU’s will be connected to the cloud and from there we can monitor and control different rooms
conditions and we can control them from any whare

3.1. MQTT protocol

Figure 4. MQTT Broker Job.

Fig.4 shows MQTT is a lightweight open messaging protocol that enables connectivity with limited
bandwidth to transfer sensor information easily. Machine-to-machine (M2M) communication is enabled
through the protocol, which follows a publish/subscribe communication structure.

3.2. AES kdsb based algorithm

AES is a block cypher algorithm that has key and block lengths of 128,192and 256 bits. Based on key
length, the methods are carried out in at most comparable rounds. For each round, AES has four sorts
of transformations, Shift rows, add round keys, shift bytes, and mix columns are the four main operating
modes. The non-linear exchange operation is specified as the sub bytes. The bytewise permutation is
represented by the shift rows; the four mixing operations are represented by the six columns, and the
XOR procedure of the state with the round keys is represented by the add round keys.
S-box design
It is referred to be the AES algorithm's soul. Because it provides strong security, the general public
is unaware of the essentials of the s-box. Furthermore, the substitution process is both efficient and quick.
Because of the static structure of the S-box, this basic data encryption was able to defeat the brute force
attack. To solve the shortcomings, the suggested work employs a key-dependent s-box method. The key
block and the user data block are the foundations of the schema. The data block is the algorithm's input.

4. Experimental results

MQTT PROTOCOL VS HTTP

MQTT HTML

10000
8000
BYTES/1MIN

6000
4000
2000
0
0 50 100 150 200 250 300 350
TIME(S)

Figure 5. HTTP vs MQTT protocol.

According to the results provided in Fig.5 amount of data transferred per minute using the lightweight
protocol, MQTT is around 1500 bytes, compared to HTPP is 7500 bytes (the highest value). The ratio
for the number of bytes exchanged per minute is 1:5, implying the MQTT protocol sends one-byte data

205
 Proceedings of the 2023 International Conference on Machine Learning and Automation
DOI: 10.54254/2755-2721/44/20230067

against five bytes when using HTTP. As a result, the lightweight protocol saves four bytes per minute
in data transmission. One byte of data transported per minute is believed to equal one unit of electricity
consumption. The lightweight protocol consumes one unit of power, whereas HTTP consumes five units.
As a result, employing MQTT saves four units of power every minute of data transport. The percentage
of money saved by implementing the lightweight protocol.

Figure 6. Image sent to the mail.

Fig.6 shows the photographs of the persons who are waiting at the door delivered to the mail from
the Raspberry Pi, as well as all of the sensor data collected from the node MCUs, which will be sent to
the Raspberry Pi and subsequently to the cloud. We use node MCUs as a controller with PIR and LDR
sensors to detect human movements and use relays to control the appliances. We use an LPG gas sensor
and a DHT11 sensor to measure temperature, humidity, and gas in the kitchen and rooms. And this
information is transferred to the cloud. Figure 7 shows the data being transferred to the cloud and the
devices being controlled, and we can check the state of the device and the room's condition from
anywhere.

Figure 7. Data sending to the cloud.

5. Conclusion
This paper has suggested a system that uses a unique RSA-based algorithm as well as the AES KBSB
method. MQTT, a lightweight protocol, is used to send data between the devices in an energy-efficient
manner. Data is transmitted between devices using the publish-subscribe communication architecture.
The proposed system when tested has shown some prodigious results, the persons have been detected
who are registered in the cloud to unlock the door and the images have been sent to the user’s mail-in
many instances. Further, this system can be extended by building a software application, where the
system directly takes an action when a person intrudes.

References
[1] Sathishkumar V E, Changsun Shin, Youngyun Cho, “Efficient energy consumption prediction
model for a data analytic-enabled industry building in a smart city”, Building Research &
Information, Vol. 49. no. 1, pp. 127-143, 2021.
[2] Sathishkumar V E, Youngyun Cho, “A rule-based model for Seoul Bike sharing demand
prediction using Weather data”, European Journal of Remote Sensing, Vol. 52, no. 1, pp. 166-
183, 2020.

206
 Proceedings of the 2023 International Conference on Machine Learning and Automation
DOI: 10.54254/2755-2721/44/20230067

[3] Sathishkumar V E, Jangwoo Park, Youngyun Cho, “Seoul Bike Trip duration prediction using
data mining techniques”, IET Intelligent Transport Systems, Vol. 14, no. 11, pp. 1465-1474,
2020.
[4] Sathishkumar V E, Jangwoo Park, Youngyun Cho, “Using data mining techniques for bike
sharing demand prediction in Metropolitan city”, Computer Communications, Vol. 153, pp.
353-366, 2020.
[5] Sathishkumar V E, Yongyun Cho, “Season wise bike sharing demand analysis using random
forest algorithm”, Computational Intelligence, pp. 1-26, 2020.
[6] Sathishkumar V E, Myeongbae Lee, Jonghyun Lim, Yubin Kim, Changsun Shin, Jangwoo Park,
Yongyun Cho, “An Energy Consumption Prediction Model for Smart Factory using Data
Mining Algorithms” KIPS Transactions on Software and Data Engineering, Vol. 9, no. 5, pp.
153-160, 2020.
[7] Sathishkumar V E, Jonghyun Lim, Myeongbae Lee, Yongyun Cho, Jangwoo Park, Changsun
Shin, and Yongyun Cho, “Industry Energy Consumption Prediction Using Data Mining
Techniques”, International Journal of Energy Information and Communications, Vol. 11, no.
1, pp. 7-14, 2020.
[8] Ronggang Zhang, Sathishkumar V E, R. Dinesh Jackson Samuel, “Fuzzy Efficient Energy Smart
Home Management System for Renewable Energy Resources”, Sustainability, Vol. 12, no. 8,
pp. 1-15, 2020.

207