Important Questions and Answers
CP25C14-AI in IoT
Syllabus
1. Introduction to AI in IoT: Fundamentals of AI and IoT - IoT Architecture and AI
Integration - AI Models for IoT (Machine Learning, Deep Learning) - AI-Driven IoT
Applications (Smart Cities, Healthcare, Industrial IoT) - Ethical Considerations in AI-
Enabled IoT – Case Study: Smart Agriculture with AI-Enabled IoT Sensors.
Activities:
• Analytical study of AI integration within IoT ecosystems, focusing on architectural layers,
data flow, and intelligence distribution across device, edge, and cloud.
• Simulation-based generation and statistical analysis of IoT sensor data using computational
tools (e.g., Jupyter Notebook) to interpret trends, patterns, and variability.
2. Data Acquisition & Pre-processing in AI-Enabled IoT: Sensor Data Collection &
Processing - AI-Based Data Normalization, Filtering & Feature Engineering - Challenges in
Data Handling & Storage for AI Models - Ethical Considerations in AI-Driven Data
Processing – Case Study: Predictive Maintenance in Smart Factories.
Activities:
• Develop a conceptual framework for real-time sensor data preprocessing, detailing data
collection, normalization, filtering, and feature engineering stages.
• Analyze AI-enhanced IoT data filtering techniques, focusing on methodologies,
performance metrics, and practical challenges.
3. AI Models for IoT Data Processing & Decision-Making: Supervised & Unsupervised AI
Models in IoT - Classification, Clustering & Anomaly Detection - Deep Learning Models for
V.
IoT: CNNs, RNNs - Performance Metrics & Model Optimization – Case Study: AI-Driven
U
m
Traffic Monitoring System.
ap
Activities
at
• Analyze and justify the selection of appropriate AI models for real-time IoT data
hy
classification, considering problem context, data characteristics, and performance
,Pr
constraints.
of
• Conduct structured discussions or guided demonstrations on deep learning models (CNNs,
./
EC
N
ot
es
, RNNs) for IoT applications, focusing on architecture, working principles, and suitability for
time-series data.
4. Edge AI & Security for IoT Systems: Edge AI Deployment Strategies - Edge AI for Real-
Time Decision-Making - Deploying AI Models in Resource-Constrained IoT Devices - AI-
Powered Security & Intrusion Detection in IoT Networks - Ethical Implications & Privacy
Concerns – Case Study: AI- Based Intrusion Detection in Healthcare IoT Networks.
Activities:
• Conceptually analyze AI-powered anomaly detection frameworks for IoT systems,
focusing on model architecture, detection strategies, and performance evaluation criteria.
• Critically review and present recent research articles on security and privacy mechanisms
in IoT systems, with emphasis on AI-based intrusion detection approaches.
5. AI in IoT – Industry Trends: Industry Applications of AI in IoT - Smart Cities, Healthcare,
& Industrial IoT - Scaling AI for Large-Scale IoT Deployments - Sustainability & Scalability
Challenges in AI-Enabled IoT.
Activities:
• Undertake an analytical case study of AI-driven Smart Cities, examining sustainability,
scalability, governance frameworks, and ethical considerations.
6. AI in IoT – Ethics and Future Trends: Monetization Strategies & AI Business Models in
IoT - Green AI and IoT Regulations and Standardization - Edge AI advancements for real-
time IoT decision-making - Quantum AI & IoT interoperability for future applications - Case
Study: AI-Driven Smart Cities for Sustainable Urban Development
Activities:
• Prepare a conceptual design proposal for a real-time IoT-based decision-making system,
detailing system architecture, AI model rationale, deployment strategies, and regulatory
compliance aspects.
S.No. Description
Unit-I
Introduction to AI in IoT
Basic Concepts
S.No. Topic Important Basic Concepts
Artificial Intelligence (AI) enables machines to perform tasks such as
Fundamentals of
1 learning, reasoning, prediction and decision-making. Examples: expert
V.
AI
U
systems, speech recognition and image processing.
m
Internet of Things (IoT) connects physical devices through the internet to
ap
Fundamentals of
2 collect and exchange data. Main components: sensors, actuators, connectivity,
at
IoT
hy
cloud and user applications.
,
Four layers: (1) Perception layer – sensors and actuators, (2) Network layer –
Pr
3 IoT Architecture
data communication, (3) Processing layer – cloud/edge computing, (4)
of
./
EC
N
ot
es
, Application layer – user services.
AI and IoT IoT sensors collect real-time data, while AI analyses the data to identify
4
Integration patterns, predict events and make intelligent decisions.
A branch of AI that learns patterns from IoT data. Types: supervised learning,
Machine Learning
5 unsupervised learning and reinforcement learning. Applications: fault
(ML)
detection and prediction.
Uses artificial neural networks with multiple layers to analyse complex IoT
Deep Learning
6 data. Applications: image recognition, speech processing and predictive
(DL)
maintenance.
Smart City AI-enabled IoT is used for smart traffic management, intelligent street
7
Applications lighting, waste management, pollution monitoring and energy management.
Healthcare Wearable sensors monitor heart rate, temperature and blood pressure. AI
8
Applications analyses patient data for disease prediction and remote patient monitoring.
Connects industrial machines and sensors. AI supports predictive
Industrial IoT
9 maintenance, quality inspection, process automation and production
(IIoT)
optimization.
Ethical Important issues include data privacy, security, transparency, algorithmic bias,
10
Considerations accountability and informed user consent.
AI-enabled IoT sensors monitor soil moisture, temperature, humidity and crop
11 Smart Agriculture conditions. AI recommends irrigation, detects diseases and improves crop
yield.
Benefits of AI in Real-time monitoring, automation, reduced operating cost, improved
12
IoT efficiency, accurate prediction and better decision-making.
Important Equations (Copy to Word)
Equation
Concept
Mean of sensor data Mean = (x1 + x2 + ... + xn) / n
Linear Regression y = mx + c
Prediction Error (MSE) MSE = (1/n) × Σ(yi - ŷi)²
Accuracy Accuracy = (Correct Predictions / Total Predictions) × 100
Soil Moisture Percentage Moisture (%) = ((ADC Value - Dry Value) / (Wet Value - Dry Value)) × 100
V.
U
Neural Network y = f(Wx + b)
m
IoT Data Transmission Data = Number of Sensors × Sampling Rate × Data Size
ap
Important Exam Points
at
hy
,
AI: Makes machines intelligent.
Pr
IoT: Connects physical devices to the internet.
of
./
EC
N
ot
es
, AI + IoT: Enables intelligent monitoring and automation.
ML: Learns patterns from sensor data.
DL: Uses multilayer neural networks.
IIoT: AI applications in industries.
Smart Agriculture: Reduces water usage and improves crop productivity through intelligent
irrigation.
Ethics: Ensures privacy, security, fairness and transparency.
Case Study – Smart Agriculture with AI-Enabled IoT Sensors:
Soil moisture, temperature and humidity sensors collect field data. The data is transmitted to an IoT
gateway or cloud platform. AI analyses the collected data and determines irrigation requirements. An
automatic water pump is switched ON or OFF based on soil moisture conditions. This reduces water
wastage, labour cost and improves crop productivity.
Multiple Choice QA
1. What is the primary purpose of integrating Artificial Intelligence (AI) with the Internet of Things
(IoT)?
A) To increase hardware cost
B) To enable intelligent data analysis and automated decision-making
C) To reduce internet connectivity
D) To eliminate sensors
Answer: B
Reason: AI analyzes the data collected by IoT sensors and makes intelligent decisions. This
integration supports automation, prediction, and real-time monitoring.
2. Which of the following represents the correct basic IoT architecture?
A) Application – Network – Perception
B) Perception – Network – Application
V.
C) Network – Application – Storage
U
D) Storage – Perception – Network
m
ap
at
Answer: B
hy
,
Reason: IoT architecture commonly consists of three layers:
Pr
of
./
EC
N
ot
es
CP25C14-AI in IoT
Syllabus
1. Introduction to AI in IoT: Fundamentals of AI and IoT - IoT Architecture and AI
Integration - AI Models for IoT (Machine Learning, Deep Learning) - AI-Driven IoT
Applications (Smart Cities, Healthcare, Industrial IoT) - Ethical Considerations in AI-
Enabled IoT – Case Study: Smart Agriculture with AI-Enabled IoT Sensors.
Activities:
• Analytical study of AI integration within IoT ecosystems, focusing on architectural layers,
data flow, and intelligence distribution across device, edge, and cloud.
• Simulation-based generation and statistical analysis of IoT sensor data using computational
tools (e.g., Jupyter Notebook) to interpret trends, patterns, and variability.
2. Data Acquisition & Pre-processing in AI-Enabled IoT: Sensor Data Collection &
Processing - AI-Based Data Normalization, Filtering & Feature Engineering - Challenges in
Data Handling & Storage for AI Models - Ethical Considerations in AI-Driven Data
Processing – Case Study: Predictive Maintenance in Smart Factories.
Activities:
• Develop a conceptual framework for real-time sensor data preprocessing, detailing data
collection, normalization, filtering, and feature engineering stages.
• Analyze AI-enhanced IoT data filtering techniques, focusing on methodologies,
performance metrics, and practical challenges.
3. AI Models for IoT Data Processing & Decision-Making: Supervised & Unsupervised AI
Models in IoT - Classification, Clustering & Anomaly Detection - Deep Learning Models for
V.
IoT: CNNs, RNNs - Performance Metrics & Model Optimization – Case Study: AI-Driven
U
m
Traffic Monitoring System.
ap
Activities
at
• Analyze and justify the selection of appropriate AI models for real-time IoT data
hy
classification, considering problem context, data characteristics, and performance
,Pr
constraints.
of
• Conduct structured discussions or guided demonstrations on deep learning models (CNNs,
./
EC
N
ot
es
, RNNs) for IoT applications, focusing on architecture, working principles, and suitability for
time-series data.
4. Edge AI & Security for IoT Systems: Edge AI Deployment Strategies - Edge AI for Real-
Time Decision-Making - Deploying AI Models in Resource-Constrained IoT Devices - AI-
Powered Security & Intrusion Detection in IoT Networks - Ethical Implications & Privacy
Concerns – Case Study: AI- Based Intrusion Detection in Healthcare IoT Networks.
Activities:
• Conceptually analyze AI-powered anomaly detection frameworks for IoT systems,
focusing on model architecture, detection strategies, and performance evaluation criteria.
• Critically review and present recent research articles on security and privacy mechanisms
in IoT systems, with emphasis on AI-based intrusion detection approaches.
5. AI in IoT – Industry Trends: Industry Applications of AI in IoT - Smart Cities, Healthcare,
& Industrial IoT - Scaling AI for Large-Scale IoT Deployments - Sustainability & Scalability
Challenges in AI-Enabled IoT.
Activities:
• Undertake an analytical case study of AI-driven Smart Cities, examining sustainability,
scalability, governance frameworks, and ethical considerations.
6. AI in IoT – Ethics and Future Trends: Monetization Strategies & AI Business Models in
IoT - Green AI and IoT Regulations and Standardization - Edge AI advancements for real-
time IoT decision-making - Quantum AI & IoT interoperability for future applications - Case
Study: AI-Driven Smart Cities for Sustainable Urban Development
Activities:
• Prepare a conceptual design proposal for a real-time IoT-based decision-making system,
detailing system architecture, AI model rationale, deployment strategies, and regulatory
compliance aspects.
S.No. Description
Unit-I
Introduction to AI in IoT
Basic Concepts
S.No. Topic Important Basic Concepts
Artificial Intelligence (AI) enables machines to perform tasks such as
Fundamentals of
1 learning, reasoning, prediction and decision-making. Examples: expert
V.
AI
U
systems, speech recognition and image processing.
m
Internet of Things (IoT) connects physical devices through the internet to
ap
Fundamentals of
2 collect and exchange data. Main components: sensors, actuators, connectivity,
at
IoT
hy
cloud and user applications.
,
Four layers: (1) Perception layer – sensors and actuators, (2) Network layer –
Pr
3 IoT Architecture
data communication, (3) Processing layer – cloud/edge computing, (4)
of
./
EC
N
ot
es
, Application layer – user services.
AI and IoT IoT sensors collect real-time data, while AI analyses the data to identify
4
Integration patterns, predict events and make intelligent decisions.
A branch of AI that learns patterns from IoT data. Types: supervised learning,
Machine Learning
5 unsupervised learning and reinforcement learning. Applications: fault
(ML)
detection and prediction.
Uses artificial neural networks with multiple layers to analyse complex IoT
Deep Learning
6 data. Applications: image recognition, speech processing and predictive
(DL)
maintenance.
Smart City AI-enabled IoT is used for smart traffic management, intelligent street
7
Applications lighting, waste management, pollution monitoring and energy management.
Healthcare Wearable sensors monitor heart rate, temperature and blood pressure. AI
8
Applications analyses patient data for disease prediction and remote patient monitoring.
Connects industrial machines and sensors. AI supports predictive
Industrial IoT
9 maintenance, quality inspection, process automation and production
(IIoT)
optimization.
Ethical Important issues include data privacy, security, transparency, algorithmic bias,
10
Considerations accountability and informed user consent.
AI-enabled IoT sensors monitor soil moisture, temperature, humidity and crop
11 Smart Agriculture conditions. AI recommends irrigation, detects diseases and improves crop
yield.
Benefits of AI in Real-time monitoring, automation, reduced operating cost, improved
12
IoT efficiency, accurate prediction and better decision-making.
Important Equations (Copy to Word)
Equation
Concept
Mean of sensor data Mean = (x1 + x2 + ... + xn) / n
Linear Regression y = mx + c
Prediction Error (MSE) MSE = (1/n) × Σ(yi - ŷi)²
Accuracy Accuracy = (Correct Predictions / Total Predictions) × 100
Soil Moisture Percentage Moisture (%) = ((ADC Value - Dry Value) / (Wet Value - Dry Value)) × 100
V.
U
Neural Network y = f(Wx + b)
m
IoT Data Transmission Data = Number of Sensors × Sampling Rate × Data Size
ap
Important Exam Points
at
hy
,
AI: Makes machines intelligent.
Pr
IoT: Connects physical devices to the internet.
of
./
EC
N
ot
es
, AI + IoT: Enables intelligent monitoring and automation.
ML: Learns patterns from sensor data.
DL: Uses multilayer neural networks.
IIoT: AI applications in industries.
Smart Agriculture: Reduces water usage and improves crop productivity through intelligent
irrigation.
Ethics: Ensures privacy, security, fairness and transparency.
Case Study – Smart Agriculture with AI-Enabled IoT Sensors:
Soil moisture, temperature and humidity sensors collect field data. The data is transmitted to an IoT
gateway or cloud platform. AI analyses the collected data and determines irrigation requirements. An
automatic water pump is switched ON or OFF based on soil moisture conditions. This reduces water
wastage, labour cost and improves crop productivity.
Multiple Choice QA
1. What is the primary purpose of integrating Artificial Intelligence (AI) with the Internet of Things
(IoT)?
A) To increase hardware cost
B) To enable intelligent data analysis and automated decision-making
C) To reduce internet connectivity
D) To eliminate sensors
Answer: B
Reason: AI analyzes the data collected by IoT sensors and makes intelligent decisions. This
integration supports automation, prediction, and real-time monitoring.
2. Which of the following represents the correct basic IoT architecture?
A) Application – Network – Perception
B) Perception – Network – Application
V.
C) Network – Application – Storage
U
D) Storage – Perception – Network
m
ap
at
Answer: B
hy
,
Reason: IoT architecture commonly consists of three layers:
Pr
of
./
EC
N
ot
es
