CYBER DETECTION AND PREVENTION SYSTEM L T P E C

23IC22C INTERNET OF THINGS LTPC3003

INTRODUCTION
COURSE OUTCOMES 8
UponIntroduction
the successfulto Intrusion
completion– Need
of thisof Intrusion
course, Detection
the student will –
beClassification
able to: of Intrusion Detection
Systems - Components and– Source of vulnerabilities – Attacks against
CO1: analyze various design methodologies and enabling technologies for Internet of Things various Security
platform.
Objectives – Countermeasures of attacks.
CO2: apply the appropriate model for building things in IoT.
CO3:INTRUSION DETECTION
examine the standard AND by
IoT protocols PREVENTION
emphasizing theirTECHNOLOGIES
characteristics and interoperability. 10
CO4:Host Based
apply IDS – Network
the utilities of ESP32Based IDS – Information
for implementing Source
diverse IoT for IDS – Host and Network
applications.
CO5: analyze the applications
vulnerabilities of IoT in real-time
and countermeasures –, misusescenarios.
detection: pattern matching, rule based and state-
IoT DESIGN
based METHODOLOGIES
anomaly 9
detection: Statistical based, machine learning based, data mining-based hybrid
Internet of
detection.Things - Physical Design- Logical Design- IoT Enabling Technologies - IoT Levels and
Deployment
IDS ANDTemplates - Domain Specific IoTs - IoT and M2M - IoT System Management with NETCONF-
IPS ARCHITECTURE
YANG- IoT Platforms Design Methodology.
10
IoT MODELING 9
Tiered architecture: Single-tiered, multi-tiered, peer-to-peer – Sensor: Functions, deployment and
IoT Reference architecture - High-level M2M ETSI architecture - IETF architecture - OGC architecture -
securitymodels:
Reference – Agent: Functions,
Domain model Manager component:
- Information Functions,model
model - Functional deployment and security
- Communication – - Case
model
Information
Studies: Monitoringflowsoilin conditions,
IDS and IPS – Defending
Smart IDS/IPS – Case study on Commercial and open-
fridge solutions.
source
IoT PROTOCOLSIDS. 9
Protocol Standardization for IoT - MQTT - CoAP – AMQP - Zigbee - BLE - 6LowPAN - LoRaWAN- Z- 8
ALERT MANAGEMENT AND CORRELATION DATA FUSION
WaveAlert correlation – Preprocess – Correlation techniques – post process – Alert correlation
- NarrowBand-IoT.
IoTArchitecture
APPLICATIONS - - Cooperative discovery of intrusion chain – abstraction-based
9 intrusion detection –
Role of ESP32
Interest in IoT
based applications - and
communication ESP-IoT Development
cooperation Framework(IDF)-
– Agent Architecture and GPIO
based cooperation
Programming
WIRELESS - Interfacing
IDPS sensors - Creating a web server -Data Storage - Edge computing with ESP32 9 -
Applications: LED Blinking,
Threats against WLANs,Maintaining the distance
802.11 Wireless of Things,Attacks,
Infrastructure Bluetooth
WEPcommunication, Humidity
Attacks, Wireless Client
measurement
Attacks, Bluetooth Attacks, Cell phones, Other Hybrid Devices Attack Detection, Jail breaking -
REAL TIME IoT SYSTEM 9
Threat Briefing – Quantifying risk - Return on Investment (ROI)
Smart lighting - Intrusion detection system - Emergency response - Smart parking -Weather monitoring -
L: 45, TOTAL: 45 PERIODS
Forest fire detection - Smart grid - Inventory management - Smart payment - Smart irrigation - Wearable
REFERENCES
electronics.
1.Ali A. Ghorbani, Wei Lu, “Network Intrusion Detection and Prevention: Concepts
L: 45, TOTAL: and
45 PERIODS
Techniques”,
REFERENCE BOOKS First Edition, Springer US, 2019
1. 2. Al-Sakib
Arshdeep Khan
Bahga andPathan, “The State
Vijay Madisetti, of the Art
“Internet in Intrusion
of Things Prevention
– A Hands-on and Detection”,
Approach” First Orient
Second Edition,
Edition,Private
Blackswan CRC Press,Limited 2014
- New Delhi, 2019.
2. Simone Cirani, Gianluigi Ferrari, Marco Picone and Luca Veltri, “Internet of Things: Architectures,
Protocols and Standards” First Edition,Wiley, 2018.
3. David Hanes, Gonzalo Salgueiro,Patrick Grossetete, Robert Barton and Jerome Henry, “IoT Fundamentals:
Networking Technologies, Protocols, and Use Cases for the Internet of Things”, First Edition, Cisco Press,
2017.
4. Vedat Ozan Oner, “Developing IoT Projects with ESP32”, First Edition, Packt Publishing, 2021.
5. Olivier Hersent, David Boswarthick and Omar Elloumi, “The Internet of Things - Key applications and
Protocols”, Second Edition, Wiley, 2018.
6. Rajkumar Buyya and Amir Vahid Dastjerdi , "Internet of Things: Principles and Paradigms", Second
Edition, Elsevier, 2020.
7. Marco Zappatore, "Internet of Things- Architectures, Protocols and Standards" First Edition, Springer,
2018.
23CT22C MACHINE LEARNING
ALGORITHMS LTPC 3 00 4
CO1: apply the mathematical and statistical prospective of
machine learning. 9
Components of learning – Learning models – Types of
learning – supervised -unsupervised – reinforcement – Curse
of Dimensionality – Bias and Variance – Learning Curve –
Classification – Training versus testing – Error and Noise
CO2: build the Predictive and Probabilistic modeling
techniques for regression and classification.
9
Predictive Models: Univariate linear regression – Multivariate
linear regression – Logistic regression – Stochastic Gradient
Descent – Linear Multiclass Prediction – Probabilistic Models:
Density Estimation – Naïve Bayes Models – Statistical
Estimation – Conditional Models
CO3: apply the supervised and unsupervised learning
techniques and optimize the models.
9
Supervised Models: Neural Networks – Multilayer Networks
with Backpropagation Algorithm – Case Study: Face
Recognition – Kernel Tricks – Soft-SVM with Kernels.
Unsupervised Models:Measuring dissimilarity – Spectral
clustering –Hierarchical clustering – K- Means clustering –
Fuzzy C-Means Clustering – Dimensionality Reduction using
PCA
CO4: develop the reinforcement learning algorithms to
solve complex problems.
9
Passive reinforcement learning: Direct utility estimation –
Adaptive Dynamic Programming – Temporal-Difference
Learning – Active reinforcement learning: exploration –
learning an action- utility function – Generalization in
reinforcement learning – policy search – Applications in game
playing – applications in healthcare using AutoML (Google).
CO5: evaluate the machine learning models to extract and
optimize the features. 9
Analytical Learning: Discovering new features – Deductive
Learning – Knowledge level learning - Feature Learning :
Feature Selection – Greedy Selection Approaches – Feature
Manipulation and Normalization – Perceptron Learning