K.L.N.

COLLEGE OF ENGINEERING
(An Autonomous Institution, Affiliated to Anna university, Chennai)
DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

Deep Learning based Attack Detection for

Cyber - Physical System Cybersecurity

GUIDE: TEAM MEMBERS:

Mr. ARUL SELVAN M REG. NO NAME

AP/CSE 910620104006 AKSHAYASUDHAN S
910620104301 ABHISHEK M R
910620104302 BHAIRAVAMANIKANDAN M

15/04/2024 KLNCE - Department of Computer Science and Engineering 1

 ABSTRACT
• To mitigate vulnerabilities in the CAN protocol's weak security structure, robust
IDS are crucial for protecting vehicles against malicious attacks
• To combat the drawbacks of RNN-based IDS methods, which can miss features
in the context vector due to reliance solely on past data, a solution is needed
• To enhance intrusion detection in-vehicle CAN bus systems, a novel multi-class
IDS is proposed. It utilizes a Transformer-based Attention Network (TAN),
eliminating the need for CNN and LSTM layers while efficiently classifying
attacks into multiple categories
• To validate its efficiency, experimental results confirm TAN's superiority in
detecting replay attacks without message labeling and its efficiency over
baselines. Moreover, TAN enhances performance via transfer learning with
limited data from diverse car models

15/04/2024 KLNCE - Department of Computer Science and Engineering 2

 INTRODUCTION
• To extract valuable insights from large datasets, entrepreneurs, researchers, and
individuals rely on data mining, also known as Knowledge Discovery in
Databases (KDD)
• To meet the increasing demand for application streaming, services now
leverage cloud servers, enabling high-end features with minimal user-side
installations and computing power
• To democratize high-end gaming experiences previously limited by hardware,
services like Google Stadia leverage cloud platforms, enabling access on any
internet-connected portable device
• To address risks posed by malicious activities in streaming services, such as
DDoS and Infiltration attacks, a reliable IDS is vital for real-time threat
detection and mitigation, ensuring cyber security
15/04/2024 KLNCE - Department of Computer Science and Engineering 3
 LITERATURE SURVEY
S.NO TITLE PUBLISHED AUTHOR DESCRIPTION
WEBSITE AND
YEAR
1. Network Traffic 2020 Ling Li, Ya • selection to find an optimal feature set
Classification Zhou, Han and log transformation to improve the
Using K-means Xiong, Cailin accuracy. The experimental results on
Clustering Hu, Xiafei We different datasets convey that the method
can obtain up to 80% overall accuracy
• RCNN, Fast RCNN, and Faster
RCNN,comparison with previous
methods for recognition

2. Deep Learning 2021 Aji Achmad • The system used Convolutional Neural
Convolutional Mustofa1 and Network to identify which of the tomato
Neural Networks Indra Budi2 diseases is present on the monitored
for Radio tomato plants. The F-RCNN trained
Identification, 2018 anomaly detection model produced a
confidence score of 80 %.
• It has more efficiency in term of
embedding capacity.

15/04/2024 KLNCE - Department of Computer Science and Engineering 4

S.NO TITLE PUBLISHED AUTHO DESCRIPTION
WEBSITE R
AND YEAR

3. End-to-end 2022 Sonali • By applying deep recurrent neural networks we learn to

radio traffic Gandhi, discriminate between several application layer traffic
sequence Asst.Prof. types on top of a constant envelope modulation
recognition Monali • Training dataset Time is High.
with recurrent Gandhi
neural
networks, 2018

4. MedBIoT: 2021 Alejandro • Machinelearning-basedintrusion detectionsystems aim to

Generation of Guerra- overcome network security limitations relyingheavilyon
an IoT Botnet Manzanar data quantity and quality. In the case of IoT networks
Dataset in a esa,Jorge these data are scarce and limited tosmall-sized networks.
Medium-sized Medina- • High vulnerable and prone to be compromised devices.
IoT Network Galindo, • High-performance effective intrusion detection system
Hayretdin (IDS).
Bahsi b
and Sven
Nomm

15/04/2024 KLNCE - Department of Computer Science and Engineering 5

 EXISTING SYSTEM

• In existing system, DDoS attacks and then, selected features are passed to
the classifiers, i.e. support vector machine, decision tree, naïve Bayes, and
multilayer perceptron to identify type of attack

• The publicly available dataset as KDD Cup 99 are used for our
experimental study. From the results of the simulation, it is clear that
GOIDS with decision tree acquires high detection and accuracy with a low
false–positive rate

• For instance, the use of demonising as feature extractors may improve the
performance in the presence of high levels of noise

15/04/2024 KLNCE - Department of Computer Science and Engineering 6

 PROPOSED SYSTEM

• The proposed IDS model is aimed at detecting network intrusions by

classifying all the packet traffic in the network as benign or malicious classes

• DDOS attacks based on Cyber security Intrusion Detection System (KDD

Cup 99) dataset has been used to train and validate

• Classification Using CNN and LSTM Model predict

• Testing dataset for anomaly detection model finally classified attack or

normal

• It is more effective of performance analysis

15/04/2024 KLNCE - Department of Computer Science and Engineering 7

 FLOW DIAGRAM

15/04/2024 KLNCE - Department of Computer Science and Engineering 8

 MODULES

• Data selection and loading

• Data preprocessing
• Splitting dataset into train and test data
• Classification
• Prediction
• Result generation

15/04/2024 KLNCE - Department of Computer Science and Engineering 9

 DATA SELECTION

15/04/2024 KLNCE - Department of Computer Science and Engineering 10

 DATA PREPROCESSING

15/04/2024 KLNCE - Department of Computer Science and Engineering 11

 CNN ALGORITHM

15/04/2024 KLNCE - Department of Computer Science and Engineering 12

 LSTM ALGORITHM

15/04/2024 KLNCE - Department of Computer Science and Engineering 13

PREDICTION WEBPAGE

15/04/2024 KLNCE - Department of Computer Science and Engineering 14

 WEBPAGE

15/04/2024 KLNCE - Department of Computer Science and Engineering 15

 REFERENCES
1. Guo Y, Ten CW, Hu S, et al. Modeling distributed denial of service attack in advanced
metering infrastructure. 2015 IEEE power & energy society innovative smart grid
technologies conference (ISGT); 2015. p. 1–
2. A. A. Khan, M. H. Rehmani, and M. Reisslein, ‘‘Cognitive radio for smart grids: Survey of
architectures, spectrum sensing mechanisms, and networking protocols,’’ IEEE Commun
Surveys Tuts., vol. 18, no. 1, pp. 860–898, 1st Quart., 2016
3. Y. Lin, X. Zhu, Z. Zheng, Z. Dou, and R. Zhou, ‘‘The individual identification method of
wireless device based on dimensionality reduction,’’ J. Supercomput., vol. 75, no. 6, pp.
3010–3027, Jun. 2019
4. T. Liu, Y. Guan, and Y. Lin, ‘‘Research on modulation recognition with ensemble learning,’’
EURASIP J. Wireless Commun. Netw., vol. 2017, no. 1, p. 179, 2017.
5. Pathan A-SK, Azad S, Khan R, et al. Security mechanisms and data access protocols in
innovative wireless networks. London: Sage; 2018

15/04/2024 KLNCE - Department of Computer Science and Engineering 16

 REFERENCES
1. C. Shi, Z. Dou, Y. Lin, and W. Li, ‘‘Dynamic threshold-setting for RFpowered cognitive
radio networks in non-Gaussian noise,’’ EURASIP J. Wireless Commun. Netw., vol 2017,
no. 1, p. 192, Nov. 2017
2. Y. Tu, Y. Lin, J. Wang, and J.-U. Kim, ‘‘Semi-supervised learning with generative
adversarial networks on digital signal modulation classification,’’ Comput. Mater. Continua,
vol. 55, no. 2, pp. 243–254, 2018
3. Toledo AL, Wang X. Robust detection of MAC layer denial-of-service attacks in CSMA/CA
wireless networks. IEEE Trans Inf Forensics Secure. 2008;3(3):347–358
4. Yong-xiong Z, Liang-ming W, Lu-xia Y. A network attack discovery algorithm based on
unbalanced sampling vehicle evolution strategy for intrusion detection. Int J Comput Appl.
2017: 1–9
5. Zargar ST, Joshi J, Tipper D. A survey of defense mechanisms against distributed denial of
service (DDOS) flooding attacks. IEEE Commun Surv Tutorials. 2013;15(4):2046–2069

15/04/2024 KLNCE - Department of Computer Science and Engineering 17

 THANK YOU !

15/04/2024 KLNCE - Department of Computer Science and Engineering 18