Q1)Explain Simplified IOT Architecture?

• Business Logic: Implements the rules and logic for how the IoT system Q3) Drivers Behind New Network Architectures
A simplified Internet of Things (IoT) architecture can be understood as a layered operates, such as setting thresholds for alerts, automating actions, or
approach that facilitates the connection of physical objects to the internet and optimizing processes. The evolution of network architectures is not merely a technological whim but a
enables data exchange and analysis. While more complex models exist, a • Cloud Platforms: Often, cloud-based platforms (e.g., AWS IoT, Azure IoT direct response to fundamental shifts in how we use and interact with digital
fundamental understanding can be achieved by considering four key layers: Hub, Google Cloud IoT Core) are used to host these applications, providing services and data. Several powerful drivers are compelling organizations and
1. Perception Layer (Sensing/Actuation Layer): This is the lowest layer and scalability, storage, and processing capabilities. service providers to rethink traditional network designs.
directly interacts with the physical world. It comprises the "things" themselves – 4. Business Layer (Management Layer): While sometimes integrated with the
1. Explosion of Data and Bandwidth Demands:
devices equipped with sensors and/or actuators. application layer, this layer focuses on the overall management, optimization, and
• Sensors: Collect data from the environment (e.g., temperature, humidity, strategic aspects of the IoT system from a business perspective. • Big Data & Analytics: The sheer volume of data generated by IoT devices,
light, motion, pressure, GPS coordinates). These can be simple or highly • Security & Privacy: Implementing measures to protect data, devices, and
social media, scientific research, and business operations demands
sophisticated. user privacy from unauthorized access and cyber threats. significantly higher bandwidth and more efficient data transfer mechanisms.
• Actuators: Receive commands and perform actions on the physical world • Device Management: Provisioning, updating, and monitoring the health and
• High-Definition Content: Streaming of 4K/8K video, virtual reality (VR),
(e.g., turning lights on/off, adjusting thermostats, opening/closing valves, status of connected devices. and augmented reality (AR) applications requires unprecedented bandwidth
controlling motors). This layer's primary role is to bridge the gap between the • Monetization & Business Models: Defining how the IoT solution generates
and low latency, pushing existing network capacities to their limits.
digital and physical realms, converting real-world phenomena into digital data revenue or provides business value (e.g., predictive maintenance, smart city • Cloud Computing: The pervasive adoption of cloud services (SaaS, PaaS,
and vice-versa. services, asset tracking). IaaS) means more data traversing wide area networks (WANs) to and from
2. Network Layer (Connectivity Layer): This layer is responsible for the reliable • Regulatory Compliance: Ensuring the IoT solution adheres to relevant laws
data centers, necessitating robust and scalable connections.
transmission of data collected by the perception layer to higher layers, and for and regulations. This layer aims to ensure the IoT system delivers on its
transmitting commands back to actuators. It encompasses various communication intended goals, providing a framework for operations, growth, and continuous 2. Rise of Cloud Computing and Hybrid/Multi-Cloud Environments:
technologies that connect the IoT devices. improvement.
• Wired Technologies: Ethernet, Fiber Optic. In essence, this simplified architecture illustrates a clear flow: physical data is • Distributed Applications: Applications are no longer confined to on-premise
• Wireless Technologies: Wi-Fi, Bluetooth, Zigbee, Z-Wave, LoRaWAN, NB- sensed, transmitted, processed into actionable insights, and then utilized for specific data centers but are distributed across multiple public clouds, private clouds,
IoT, Cellular (2G/3G/4G/5G). applications and business objectives. and edge locations. This necessitates network architectures that can
• Gateways: Often present at this layer, acting as intermediaries to translate
seamlessly connect and secure these disparate environments.
protocols between different device networks and the broader internet. They Q2) Compare and contrast IT and OT? • Application-Centricity: The focus has shifted from infrastructure to
aggregate data from multiple sensors and send it upstream. The choice of applications. Networks need to be agile and programmable to meet the
communication technology depends on factors like range, power specific performance and security requirements of various cloud-native and
consumption, data rate requirements, and cost. legacy applications.
3. Application Layer (Service Layer): This is where the actual value from IoT • Optimized Cloud Connectivity: Traditional WANs are often inefficient for
data is realized. It focuses on specific applications and services that utilize the cloud access, leading to the adoption of technologies like SD-WAN to provide
collected data for various purposes. direct, optimized, and secure paths to cloud resources.
• Data Processing & Analysis: Raw data from the network layer is processed,
3. Increasing Complexity and Management Challenges of Traditional
filtered, aggregated, and analyzed to extract meaningful insights. This often Networks:
involves big data analytics, machine learning, and artificial intelligence.
• User Interfaces: Provides dashboards, mobile applications, and web • Manual Configuration: Legacy networks often rely on manual
interfaces for users to monitor devices, visualize data, and control actuators. configuration, which is prone to human error, time-consuming, and hinders
agility.
1 2 3 4

• Lack of Visibility: Gaining end-to-end visibility across complex, multi- These drivers collectively push for network architectures that are more software- level detection, touch switches, object presence detection on conveyor 5. Environmental Sensors: These sensors monitor various environmental
vendor networks is challenging, making troubleshooting and performance driven, agile, secure, scalable, and automated, leading to innovations like SD- belts). parameters crucial for comfort, safety, and agricultural applications.
optimization difficult. WAN, SASE (Secure Access Service Edge), and edge computing. o Ultrasonic Sensor: Emits high-frequency sound waves and measures
• Scalability Limitations: Scaling traditional networks often involves the time it takes for the echo to return to determine distance (e.g., used in • Examples:
significant hardware upgrades and complex reconfigurations, which are not Q4) Different Types of Sensors with Examples car parking sensors, robot obstacle avoidance, liquid level o Humidity Sensor: Measures the amount of water vapor in the air (e.g.,

conducive to rapid business demands. measurement). used in HVAC systems, weather stations, smart homes for climate
Sensors are fundamental components of any IoT system or smart device, acting as control).
• Security Silos: Traditional security measures are often perimeter-focused, o Photoelectric Sensor: Uses a light emitter and a receiver to detect an
the "eyes and ears" that gather data about the physical world. They convert physical o Gas Sensor: Detects the presence and concentration of specific gases
struggling to cope with distributed applications and an increasingly mobile object when it interrupts or reflects a light beam (e.g., used in automatic
parameters into measurable electrical signals, which can then be processed and (e.g., carbon monoxide detectors, industrial gas leak detection, air
workforce. doors, production line counting, security systems).
analyzed. Sensors can be broadly categorized based on the physical quantity they quality monitoring).
4. Enhanced Security Requirements and Threat Landscape: measure. 3. Light Sensors (Photo Sensors): These sensors detect light intensity and convert o Pressure Sensor: Measures pressure (e.g., used in barometers for
it into an electrical signal. They are essential for ambient light control, object weather forecasting, tire pressure monitoring systems, medical devices
• Sophisticated Cyber Threats: The increasing volume and sophistication of 1. Temperature Sensors: These sensors measure heat or cold and convert it into an
detection, and imaging. for blood pressure).
cyberattacks (ransomware, DDoS, APTs) necessitate more dynamic, electrical signal. They are ubiquitous and critical for environmental monitoring,
o Air Quality Sensor: Detects various pollutants or particulate matter in
intelligent, and pervasive security capabilities built into the network itself. industrial processes, and comfort. • Examples: the air (e.g., smart home air purifiers, outdoor air quality monitoring
• Distributed Workforce: The shift to remote and hybrid work models means o Photodiode/Phototransistor: Converts light into current/voltage (e.g., stations).
• Examples:
users are accessing corporate resources from diverse, less controlled used in light meters, optical communication receivers, remote controls).
o Thermistor: Resistance changes significantly with temperature (e.g.,
environments, challenging traditional perimeter-based security. o Photoresistor (LDR - Light Dependent Resistor): Resistance changes This categorization provides a fundamental understanding of the diverse capabilities
used in digital thermometers, car engines for temperature monitoring).
• Zero Trust Security: The principle of "never trust, always verify" is driving with light intensity (e.g., used in street lights for automatic on/off of sensors, which are the cornerstone of data collection in the interconnected world.
o Thermocouple: Generates a voltage proportional to the temperature
network architectures towards micro-segmentation, identity-based access, and control, dark-activated switches).
difference between two dissimilar metals (e.g., used in industrial Q5) Classification of Actuators Based on Energy Type
continuous authentication regardless of location. o Ambient Light Sensor (ALS): Measures the amount of ambient light to
furnaces, kilns, exhaust gas temperature measurement).
• IoT Security: The proliferation of IoT devices, many with limited built-in adjust screen brightness on smartphones, laptops, etc., or for smart
o RTD (Resistance Temperature Detector): Resistance changes Actuators are essential components in automated systems, serving as the "muscles"
security, introduces new attack vectors that require network-level protection lighting control.
predictably with temperature (e.g., high-precision industrial temperature that translate control signals into physical motion or action. Their classification by
and isolation.
measurement, HVAC systems). 4. Motion and Presence Sensors: These sensors detect movement or the presence energy type highlights the fundamental way they achieve this conversion. This
5. Demand for Greater Agility, Automation, and Programmability: o Infrared (IR) Thermometer: Measures temperature from a distance of people/objects within a given area. categorization is crucial for selecting the appropriate actuator for a specific
based on emitted IR radiation (e.g., non-contact fever detection, application, considering factors like power requirements, precision, speed, and
• DevOps and CI/CD: Modern software development methodologies demand industrial process monitoring where direct contact is not possible). • Examples: environmental compatibility.
networks that can be provisioned, configured, and scaled rapidly to support o PIR (Passive Infrared) Sensor: Detects changes in infrared radiation
continuous integration and continuous deployment pipelines. 2. Proximity Sensors: These sensors detect the presence or absence of an object emitted by moving objects (e.g., used in motion-activated security lights, Here are the primary classifications of actuators based on the energy type they
• Network Automation: To cope with complexity and accelerate operations, without physical contact. They are widely used for automation, safety, and level burglar alarms, automatic doors). utilize:
there's a strong push for automating network tasks, from configuration detection. o Accelerometer: Measures acceleration and tilt in multiple axes (e.g.,
1. Electric Actuators: These actuators are powered by electricity and convert
management to troubleshooting. used in smartphones for screen rotation, activity trackers, vehicle
• Examples: electrical energy into mechanical motion. They are highly versatile, precise, and
• Programmability: Software-Defined Networking (SDN) and Network stability control).
o Inductive Sensor: Detects metallic objects by creating an widely adopted across various industries due to their clean operation and ease of
Function Virtualization (NFV) enable network control and functions to be o Gyroscope: Measures angular velocity or rotation (e.g., used in drones
electromagnetic field (e.g., used in assembly lines for part detection, integration with electronic control systems.
abstracted from hardware, allowing for programmatic control and dynamic for stability, virtual reality headsets, navigation systems).
counting, limit switches).
resource allocation. This allows networks to adapt on the fly to changing o Magnetometer (Digital Compass): Measures magnetic fields, used to
• Energy Source: Alternating Current (AC) or Direct Current (DC) electricity.
o Capacitive Sensor: Detects both metallic and non-metallic objects
business needs. determine direction (e.g., in smartphones for compass apps, navigation).
(liquids, plastics) by sensing changes in capacitance (e.g., used for liquid
5 6 7 8
 • Working Principle: Electric motors (DC, AC, stepper, servo) are the most 3. Pneumatic Actuators: Pneumatic actuators utilize compressed air or gas as their • Advantages: High stiffness and positional accuracy (especially screw types), 3. Memory Constraints:
common components, generating rotational or linear motion. Solenoids energy source. They are characterized by their speed, simplicity, and cleanliness, often self-locking (preventing back-driving), robust for certain loads.
convert electrical energy directly into linear push/pull force. Piezoelectric making them suitable for environments where contamination must be avoided, such • Disadvantages: Can be complex in design and assembly, subject to wear and • Limited Storage: Smart objects have very limited onboard memory for
actuators use the deformation of ceramic materials under an electric field for as food processing or pharmaceuticals. tear on mechanical parts, can be noisy, and often have limited range of motion storing collected data, operating system code, and application programs. This
very fine, precise movements. compared to fluid-based systems. necessitates frequent data transmission or aggressive data
• Examples: • Energy Source: Compressed air (or inert gases like nitrogen). aggregation/compression, which in itself consumes energy.
o DC Motors: Used in robotics, automotive power windows, and small • Working Principle: Compressed air is directed into a cylinder or motor, Q6) Limitations of Smart Objects in Wireless Sensor Networks (WSNs) • Firmware Footprint: The size of the firmware (operating system and
appliances. where its pressure expands and pushes a piston or rotates a vane to create application code) must be kept minimal, restricting the features and
mechanical motion. Smart objects, when integrated into Wireless Sensor Networks (WSNs), promise complexity that can be embedded.
o AC Motors: Common in industrial machinery, pumps, and fans.
• Examples: enhanced capabilities and richer data. However, their deployment and effective
o Stepper Motors: Provide precise angular control for printers, CNC
o Pneumatic Cylinders: Common in factory automation for pushing, utilization within WSNs come with several inherent limitations, primarily stemming 4. Communication Limitations:
machines, and robotic arms.
pulling, clamping, and opening/closing valves. from their physical characteristics, computational constraints, and the nature of
• Advantages: High precision and accuracy, clean operation (no fluid leaks), • Limited Bandwidth: WSN communication protocols (e.g., Zigbee,
o Pneumatic Motors: Used in air tools (drills, grinders), dental drills, and wireless communication.
relatively quiet, easy to control digitally, can be very energy efficient. LoRaWAN, Bluetooth Low Energy) are designed for low power and typically
• Disadvantages: Can be complex for very high force applications, sensitive to some robotic grippers.
1. Energy Constraints (Limited Power Budget): offer limited bandwidth. This restricts the amount and rate of data that smart
electrical noise, power loss results in no movement unless mechanically • Advantages: Clean operation (no oil), fast response times, relatively simple
objects can transmit.
locked. and inexpensive, inherently safe in explosive environments (no electrical • Battery Dependence: Most smart objects in WSNs are battery-powered, and • Short Range: Many WSN technologies are short-range, requiring multi-hop
sparks), easy to implement and maintain. replacing or recharging batteries is often impractical, especially in remote, routing for broader coverage, which adds latency and consumes more energy
2. Hydraulic Actuators: Hydraulic actuators leverage the power of pressurized, • Disadvantages: Generally lower force capabilities compared to hydraulics, hazardous, or large-scale deployments. for intermediate nodes.
incompressible fluid (typically oil) to generate significant forces and torques. They precision can be less than electric actuators due to air compressibility, requires • Power-Hungry Operations: Sensing, data processing, and particularly • Reliability Issues: Wireless communication is inherently prone to
are favored in heavy-duty applications where substantial power and robust an air compressor system (which can be noisy), energy efficiency can be wireless communication (transmitting and receiving) are energy-intensive interference, signal fading, and environmental obstacles, leading to packet loss
performance are required. lower due to air compression losses. operations. Even "smart" features like complex algorithms or advanced and unreliable data delivery.
encryption consume significant power, drastically shortening battery life. • Scalability Challenges: As the number of smart objects in a WSN increases,
• Energy Source: Pressurized hydraulic fluid (oil). 4. Mechanical Actuators: While often driven by an electric motor, mechanical
• Energy Harvesting Limitations: While energy harvesting (solar, vibration, managing network congestion, routing efficiency, and maintaining
• Working Principle: A pump creates high pressure in the fluid, which then actuators are classified by the fundamental mechanical principles they use to
RF) can extend life, it's often intermittent, insufficient for high-power connectivity becomes significantly more challenging.
acts on a piston within a cylinder (for linear motion) or vanes/gears within a convert rotational motion into linear motion, or to modify speed and torque. They
demands, and adds cost and complexity to the smart object.
motor (for rotary motion) to produce force. rely on the physical arrangement of components.
5. Security Vulnerabilities:
• Examples: 2. Limited Computational Resources:
o Hydraulic Cylinders: Used in construction equipment (excavators, • Energy Source: Mechanical force (typically from an electric motor or manual
• Resource Constraints: The limited processing power and memory of smart
bulldozers), industrial presses, lifts, and aircraft landing gear. input). • Processor Speed & Memory: Smart objects, to remain small, low-cost, and objects often preclude the implementation of robust cryptographic algorithms
o Hydraulic Motors: Found in heavy machinery, winches, and some • Working Principle: Involves physical linkages, gears, screws, cams, and power-efficient, typically feature low-power microcontrollers with limited and complex security protocols.
power steering systems. levers to transfer and transform motion. processing power and memory (RAM/Flash). • Physical Tampering: Smart objects deployed in accessible environments are
• Advantages: Capable of generating extremely high forces and torques, robust • Examples: • Complex Algorithm Execution: This restricts the complexity of algorithms vulnerable to physical tampering, which can compromise their data or
and durable, precise control over force and speed, can hold position firmly. o Screw Jacks (Lead Screws/Ball Screws): Convert rotary motion from a
that can run on the device for local data processing, filtering, or sophisticated functionality.
• Disadvantages: Require a complex power unit (pump, reservoir, filter, motor into precise linear movement, used in linear stages, adjustable decision-making. Advanced machine learning or intricate data fusion often • Vulnerable Communication: Low-power wireless protocols can be
valves), can be messy due to potential leaks, less energy-efficient for light platforms, and heavy lifting. needs to be offloaded to gateways or cloud. susceptible to eavesdropping, jamming, and spoofing if not adequately
loads, typically slower than pneumatic systems, and can be noisy. o Gear Systems: Used to change speed and torque ratios in almost all
• Software Updates & Management: Performing over-the-air (OTA) software secured, leading to data breaches or denial-of-service attacks.
machinery with rotating parts (e.g., vehicle transmissions, industrial updates on devices with limited memory can be challenging and resource-
reducers). intensive, potentially leading to bricked devices or slow updates.
9 10 11 12

• Lack of Patching: Due to remote deployment and resource constraints, o Format: Datagram Size | Datagram Tag | Datagram 1. Publish-Subscribe Model: Unlike traditional client-server models where a
patching security vulnerabilities on smart objects can be difficult and slow, Offset client directly requests data from a server, MQTT uses a publish-subscribe
leaving them exposed. o Datagram Size: The original unfragmented IPv6 packet size. architecture.
o Datagram Tag: A unique identifier for all fragments belonging to the o Publishers: Clients that send data (messages) on a specific "topic."
6. Deployment and Maintenance Challenges: o Subscribers: Clients that express interest in receiving messages on one
same original IPv6 packet.
o Datagram Offset: Indicates the position of the current fragment's or more "topics."
• Remote Locations: Smart objects are often deployed in hard-to-reach or
o Broker: The central hub that manages message distribution. Publishers
dangerous locations, making physical access for maintenance, battery payload within the original IPv6 packet.
• Header Compression (HC) Header (Mandatory for IPv6/UDP): send messages to the broker, and the broker forwards these messages to
replacement, or troubleshooting very difficult and costly.
o This is the most critical part, significantly reducing the size of the IPv6
all subscribed clients. This decouples publishers and subscribers,
• Scalability of Management: Managing hundreds or thousands of individual
and UDP headers. Instead of sending the full 40-byte IPv6 header and 8- meaning they don't need to know about each other's existence.
smart objects (firmware updates, configuration changes, health monitoring)
byte UDP header, 6LoWPAN uses stateless and stateful compression 2. Topics: Messages are organized into hierarchical, UTF-8 strings called topics.
poses significant logistical and technical challenges. A more detailed view of the 6LoWPAN header stack in an actual frame would
techniques. For example, home/living_room/temperature or
• Environmental Durability: Smart objects need to withstand harsh look like this (from bottom to top, as it's built for transmission):
o It compresses fields that can be inferred from the link-layer (e.g., building/floor1/sensor_data. Subscribers specify which topics
environmental conditions (temperature extremes, humidity, dust, vibrations)
to ensure long-term operation, requiring robust and often more expensive 1. IEEE 802.15.4 MAC Header: source/destination prefixes from 802.15.4 addresses, common IPv6 they want to receive messages from. Wildcards (+ for single level, # for
packaging. fields like version, traffic class, flow label) or from context shared multiple levels) can be used for flexible subscriptions.
• This is the fundamental link-layer header, providing addressing between nodes. 3. Quality of Service (QoS): MQTT offers three levels of QoS to guarantee
Q7) A) With the neat diagram, explain 6LOWPAN header stacks? message delivery:
(source/destination MAC addresses, PAN ID), frame control, and sequence o The compressed header points to the remaining uncompressed fields.
numbers for the wireless medium. Its maximum payload is limited. o QoS 0 (At Most Once): Messages are sent without any
6LoWPAN (IPv6 over Low-Power Wireless Personal Area Networks) is an 3. Compressed IPv6 Header and UDP Header (if present): acknowledgment. Delivery is not guaranteed, but it's the fastest option.
adaptation layer that enables IPv6 packets to be efficiently transmitted over low- 2. 6LoWPAN Encapsulation Header(s): This is the adaptation layer that performs Suitable for non-critical data.
power, low-bandwidth wireless links, such as those defined by IEEE 802.15.4. The the magic. It begins with a Dispatch field, which is crucial as it indicates the type • After the 6LoWPAN compression, the IPv6 header is significantly reduced o QoS 1 (At Least Once): Messages are delivered at least once. The
core challenge 6LoWPAN addresses is the mismatch between IPv6's minimum of 6LoWPAN header(s) that follow and whether any compression has been applied. (often to a few bytes), followed by the compressed UDP header (if UDP is publisher re-sends the message until it receives an acknowledgment
MTU (Maximum Transmission Unit) of 1280 bytes and IEEE 802.15.4's small Depending on the dispatch value, various headers can be present: used as the transport layer). from the broker. Duplicates are possible.
maximum frame size (typically 127 bytes, leaving even less for payload after MAC
o QoS 2 (Exactly Once): Messages are delivered exactly once. This
overhead). To overcome this, 6LoWPAN employs header compression and • Dispatch Header: The first byte of a 6LoWPAN packet. It identifies the type 4. IPv6/UDP Payload:
involves a four-way handshake between the publisher, broker, and
fragmentation. of the 6LoWPAN header(s) that follow (e.g., IPv6 header compression, mesh
• This is the actual data from the application layer. subscriber, ensuring no data loss or duplication. This is the slowest but
header, fragmentation header). This allows for efficient parsing and variable
The 6LoWPAN adaptation layer sits between the network layer (IPv6) and the data most reliable option.
header structures. Q8) Explain with example MQTT protocol?
link layer (e.g., IEEE 802.15.4 MAC). Its headers are prepended to the IPv6 packet, 4. Retained Messages: When a publisher sends a message with the "retain" flag
• Mesh Header (Optional):
effectively "stacking" on top of the MAC header and encapsulating the compressed set, the broker stores the last message published on that topic. Any new
o Used for link-layer forwarding over multiple 802.15.4 hops within the The MQTT (Message Queuing Telemetry Transport) protocol is a lightweight,
IPv6/UDP payload. subscriber to that topic will immediately receive this retained message. This is
6LoWPAN. It contains hop limit and source/destination addresses for publish-subscribe messaging protocol designed for constrained devices and low- useful for getting the last known state of a device.
intra-LoWPAN routing. bandwidth, high-latency, or unreliable networks. It's an ideal choice for Internet of
Here's a conceptual representation of the 6LoWPAN header stack and its typical 5. Last Will and Testament (LWT): When a client connects to the broker, it
o Format: Mesh Header | Remaining Hops | Forwarding Things (IoT) applications where efficiency and reliability are paramount.
components: can specify an LWT message and topic. If the client disconnects unexpectedly
Address
Key Concepts: (e.g., due to power loss), the broker will automatically publish this LWT
• Fragmentation Header (Optional): message on the specified topic. This allows other clients to be notified of a
o If an IPv6 packet is larger than the 802.15.4 frame payload capacity,
device's abnormal disconnection.
6LoWPAN fragments it. This header is added to each fragment.
13 14 15 16
Example Scenario: Smart Home Temperature Monitoring Q9) List and explain the key advantages of internet protocols? o Advantage: This allows for efficient use of network bandwidth, as For example, a smart factory can schedule maintenance for a specific machine
multiple communication streams can share the same physical links. It part only when ML predicts its failure, rather than on a fixed schedule.
Imagine a smart home with multiple temperature sensors and a display unit. also makes the network more resilient to failures, as packets can be • Anomaly Detection: IoT devices are susceptible to various malfunctions,
The Internet Protocol (IP) suite, often referred to as TCP/IP, forms the foundational rerouted around congested or failed paths. security breaches, or unusual operational patterns. ML models can learn
• Broker: A central MQTT broker running on a Raspberry Pi or a cloud
backbone of the internet and most modern networks. Its widespread adoption stems 5. Robustness and Fault Tolerance: "normal" behavior from continuous data streams. Any significant deviation
service.
o Explanation: Due to packet switching and decentralized routing, the IP from this norm is flagged as an anomaly. This is crucial for:
• Publishers (Temperature Sensors): from several key advantages that make it highly effective and versatile:
network is inherently robust. If one path or router fails, packets can often o Security: Detecting unusual network traffic patterns or unauthorized
o A sensor in the living room publishes temperature readings to the topic
1. Interoperability and Standardization: find alternative routes to their destination. access attempts on IoT devices.
home/living_room/temperature.
o Explanation: IP is an open standard, meaning its specifications are o Advantage: This makes the internet resilient to individual component o Quality Control: Identifying defects in manufacturing processes by
o A sensor in the bedroom publishes to
publicly available and non-proprietary. This allows different failures, ensuring continuous availability of services even in the face of analyzing visual data (e.g., using computer vision) or sensor readings.
home/bedroom/temperature.
manufacturers to build network devices (routers, switches, computers, disruptions. o Operational Monitoring: Flagging unusual energy consumption in a
o Each sensor connects to the broker, takes a reading (e.g., every minute),
smartphones) that can seamlessly communicate with each other, 6. Cost-Effectiveness: smart building or erratic sensor readings from an environmental
and publishes it to its respective topic with QoS 1. o Explanation: Being an open standard, IP allows for competitive monitoring system.
• Subscribers (Display Unit): regardless of their underlying hardware or operating system.
o Advantage: This standardization fosters a vast ecosystem of compatible development and manufacturing of networking equipment. The • Resource Optimization: ML plays a vital role in optimizing resource usage
o A display unit in the kitchen wants to show the temperatures of both the
devices and services, enabling global connectivity and preventing distributed nature of the network also reduces the need for expensive, across various IoT deployments.
living room and bedroom. It subscribes to home/+/temperature. centralized infrastructure. o Smart Grids: Analyzing real-time energy demand and supply data to
vendor lock-in.
The + wildcard allows it to receive messages from o Advantage: This contributes to lower costs for network deployment and optimize power distribution, reduce waste, and incorporate renewable
2. Scalability:
home/living_room/temperature and access, making internet connectivity widely accessible and affordable energy sources more effectively.
o Explanation: The IP architecture is designed to scale from small local
home/bedroom/temperature. networks (LANs) to the global internet. It achieves this through a globally. o Smart Homes/Buildings: Adjusting lighting, HVAC (heating,
o When a new temperature reading is published by a sensor, the broker ventilation, and air conditioning), and other systems based on occupancy
hierarchical addressing scheme (IP addresses) and routing mechanisms
receives it and forwards it to the display unit, which then updates its that efficiently direct traffic across vast networks. Q10) What are the common applications of machine learning in IoT? patterns, external weather, and user preferences to minimize energy
display. o Advantage: This allows the internet to accommodate billions of devices consumption.
• Example of Retained Message: The synergy between Machine Learning (ML) and the Internet of Things (IoT) is
and a rapidly growing volume of data, adapting to increasing demand o Agriculture: Optimizing irrigation and fertilization based on soil
o If the living room sensor publishes its temperature with the retain flag,
transforming various industries by enabling devices to not just collect data, but to
without fundamental architectural changes. moisture, weather forecasts, and crop health data collected by IoT
and then the display unit (or any new subscriber) comes online, it will interpret it, learn from it, and act intelligently. The massive amounts of data
3. Flexibility and Heterogeneity: sensors.
immediately receive the last published temperature for the living room, generated by IoT devices provide the perfect fuel for ML algorithms to uncover
o Explanation: IP operates independently of the underlying physical • Personalization and Contextual Awareness: In consumer IoT, ML enhances
even if no new reading has occurred yet. This provides immediate state hidden patterns, make predictions, and automate complex tasks.
network technologies (e.g., Ethernet, Wi-Fi, fiber optics, cellular). It user experience by making devices more adaptive and personalized.
information. provides a common abstraction layer above diverse hardware. Here are some common applications: o Smart Homes: Learning user routines to automate tasks like turning on
• Example of Last Will and Testament (LWT): o Advantage: This "network of networks" concept allows different types lights before arrival, adjusting thermostats, or playing preferred music.
o The living room sensor could configure an LWT message "Living Room • Predictive Maintenance: This is perhaps one of the most impactful o Wearable Devices: Analyzing biometric data to provide personalized
of networks to interconnect and communicate, providing immense
Sensor Offline" on the topic home/status/living_room. If the flexibility in network design and deployment. A device on a Wi-Fi applications. ML algorithms analyze real-time sensor data (vibration, health insights, fitness recommendations, or early warnings for potential
sensor suddenly loses power, the broker will publish this message, network can seamlessly communicate with a device on a wired Ethernet temperature, pressure, acoustic signatures) from industrial machinery, health issues.
allowing other parts of the smart home system (e.g., a notification network. vehicles, or even home appliances. By identifying subtle anomalies and trends • Computer Vision and Image Recognition: IoT devices equipped with
service) to know about the sensor's ungraceful shutdown. 4. Packet Switching: that precede failures, ML can predict when a component is likely to break cameras (e.g., smart cameras, drones, industrial inspection systems) generate
o Explanation: IP breaks data into small, manageable units called down. This allows for proactive maintenance, reducing unplanned downtime, vast amounts of image and video data. ML, particularly deep learning, can
packets. Each packet contains source and destination IP addresses and is extending equipment lifespan, and significantly lowering operational costs. process this data for:
routed independently across the network.
17 18 19 20

o Surveillance: Object detection, facial recognition, and suspicious the NameNode to signal its availability and periodically reports its o Purpose: This provides the common utilities and libraries that support o Battery Life: Security measures, especially encryption, consume power.
activity detection for security applications. data blocks. the other Hadoop modules. It includes the necessary file systems, RPC For battery-powered IoT devices, this can be a trade-off against desired
o Retail Analytics: Understanding customer flow, shelf inventory, and ▪ Secondary NameNode (Helper): It's not a standby for the (Remote Procedure Call) mechanisms, and other core components battery life, leading to less secure implementations.
popular products. NameNode, but rather a helper. It periodically merges the required by HDFS, YARN, and MapReduce. 2. Lack of Standardization and Fragmentation:
o Traffic Management: Analyzing traffic patterns to optimize signal NameNode's edit logs with its file system image to prevent the edit o Diverse Ecosystem: The IoT landscape is highly fragmented with a vast
timings and reduce congestion. logs from becoming too large, thus reducing NameNode startup Diagram: array of devices, communication protocols, operating systems, and
• Predictive Analytics for Business Insights: Beyond operational time. manufacturers. This lack of a unified standard makes it difficult to
improvements, ML on IoT data can provide valuable business intelligence. 2. Yet Another Resource Negotiator (YARN): establish consistent security best practices and apply universal security
Retailers can predict demand for products based on inventory levels and o Purpose: YARN is Hadoop's resource management and job scheduling solutions.
customer footfall from IoT sensors. Logistics companies can optimize framework. It decouples the resource management capabilities from the o Interoperability Challenges: Securing interactions between devices
delivery routes based on real-time traffic and weather data. data processing capabilities, allowing various data processing engines from different vendors can be complex due to varying security
(like MapReduce, Spark, Hive, etc.) to run on Hadoop. implementations and proprietary protocols.
o Components: 3. Weak Authentication and Authorization:
11) Explain the elements of Hadoop with a neat diagram? ▪ ResourceManager (Master): The global master that manages the o Default/Hardcoded Credentials: Many IoT devices ship with weak,
allocation of cluster resources (CPU, memory) to various easily guessable, or even hardcoded default usernames and passwords
Hadoop is an open-source software framework for storing and processing large
applications running on Hadoop. It schedules jobs and monitors the (e.g., "admin/admin," "root/password"). Users often fail to change these,
datasets (Big Data) in a distributed computing environment. It is designed to scale
health of the NodeManagers. creating easy entry points for attackers.
up from single servers to thousands of machines, each offering local computation
▪ NodeManager (Slaves): Runs on each slave node and is o Lack of Multi-Factor Authentication (MFA): Most IoT devices do not
and storage. Its core strength lies in its ability to handle failures gracefully and
responsible for managing resources on that specific node. It reports support MFA, which significantly increases the risk of unauthorized
process data in parallel.
resource usage to the ResourceManager and launches and monitors access if credentials are compromised.
Elements of Hadoop: application containers. 4. Insecure Communication and Data Encryption:
▪ ApplicationMaster: One ApplicationMaster is launched for each o Unencrypted Data Transmission: A significant number of IoT devices
Q12) What are the common challenges in IoT security?
Hadoop primarily consists of four core components: application (e.g., a MapReduce job). It coordinates the transmit sensitive data (personal information, sensor readings) over
application's execution by negotiating resources with the IoT devices introduce a unique set of security challenges due to their diverse nature, unencrypted channels, making it vulnerable to eavesdropping and
1. Hadoop Distributed File System (HDFS): ResourceManager and monitoring the application's progress. interception by attackers.
limited resources, widespread deployment, and often insufficient attention to
o Purpose: This is Hadoop's primary storage layer. HDFS is designed to 3. MapReduce: o Weak Encryption Algorithms: Even when encryption is used, it might
security during development. These challenges can lead to serious consequences,
store very large files across a cluster of commodity hardware. It achieves o Purpose: This is Hadoop's original distributed data processing rely on outdated or weak algorithms that are easily breakable.
including data breaches, privacy violations, physical damage, and service
fault tolerance by replicating data blocks across multiple machines. framework. It's a programming model and an execution engine for 5. Lack of Regular Updates and Patch Management:
disruptions.
o Components: processing large datasets in parallel across a cluster. o Limited Firmware Update Mechanisms: Many IoT devices lack a
▪ NameNode (Master): The central authority that manages the file o Phases: Here are some common challenges: secure and reliable mechanism for over-the-air (OTA) firmware updates.
system namespace (directories, files, blocks). It stores metadata ▪ Map Phase: Takes input data, splits it into smaller chunks, and This makes it difficult for manufacturers to push security patches for
about the files (e.g., file names, directories, permissions, and the processes each chunk independently. It produces intermediate key- 1. Inherent Device Constraints: newly discovered vulnerabilities.
mapping of data blocks to DataNodes). There is only one active value pairs. o Limited Processing Power & Memory: Many IoT devices are small, o "Set and Forget" Mentality: Consumers often treat IoT devices as
NameNode. ▪ Reduce Phase: Takes the intermediate key-value pairs from the low-cost, and have minimal computing resources. This often restricts appliances and don't regularly check for or apply updates, leaving them
▪ DataNode (Slaves): These are the workhorses of HDFS. They Map phase, groups them by key, and performs aggregation or their ability to implement robust security features like strong encryption, exposed to known exploits.
store the actual data blocks and perform read/write operations as summarization operations to produce the final output. complex authentication protocols, or frequent security updates. o End-of-Life Support: Manufacturers may cease support for older IoT
instructed by the NameNode. Each DataNode sends "heartbeats" to 4. Hadoop Common: devices, leaving them permanently vulnerable.
21 22 23 24
 6. Physical Vulnerabilities: o Setting the pinMode() for various pins (e.g., whether a pin will be an } # Set up the pin as output
o Physical Tampering: IoT devices deployed in accessible environments INPUT or an OUTPUT). GPIO.setup(LED_PIN, GPIO.OUT)
void loop() {
(e.g., smart meters, public sensors) are susceptible to physical attacks. o Initializing serial communication using Serial.begin(). // Turn the LED on (HIGH voltage).
Attackers might try to extract data, inject malware, or alter device o Setting up libraries or external components. // digitalWrite sets the voltage level of a digital pin. HIGH
behavior through physical access. try:
• Any code within this function will run completely before the loop() means 5V (or 3.3V depending on board).
o Lack of Physical Security Features: Many devices lack tamper-
function begins. digitalWrite(ledPin, HIGH); print("LED blinking - Press CTRL+C to stop")
detection mechanisms or secure hardware components that prevent // Wait for 1000 milliseconds (1 second). while True:
unauthorized physical access. 2. loop() function: // delay() pauses the program for a specified number of
milliseconds. # Turn LED on
7. Data Privacy Concerns:
delay(1000); GPIO.output(LED_PIN, GPIO.HIGH)
o Extensive Data Collection: IoT devices collect vast amounts of • This function, as its name suggests, contains the main program logic and runs
sensitive personal and behavioral data (e.g., health data from wearables, repeatedly, forever, as long as the Arduino board is powered. // Turn the LED off (LOW voltage).
print("LED ON")
location data from smart vehicles, home activity from smart speakers). • After the setup() function finishes, the loop() function starts executing // LOW means 0V. sleep(1) # Wait for 1 second
o Insufficient Privacy by Design: Many IoT products are developed with and continues to do so cyclically. digitalWrite(ledPin, LOW);
functionality prioritized over privacy, leading to potential misuse or • This is where you place the code that performs the core actions of your // Wait for another 1000 milliseconds.
delay(1000); # Turn LED off
unauthorized access to user data. project, such as reading sensor values, controlling actuators (like LEDs or
o Regulatory Compliance: Navigating complex global data privacy motors), and sending data. GPIO.output(LED_PIN, GPIO.LOW)
// After these two delays, the loop() function will finish its
regulations (like GDPR, CCPA) for diverse IoT data streams adds current cycle print("LED OFF")
another layer of challenge. Example: Blinking an LED // and immediately start again from the beginning, causing the sleep(1) # Wait for 1 second
8. DDoS and Botnet Threats: LED to blink continuously.
Let's illustrate this with the classic "Blink" example, which makes an LED }
o Vulnerable Devices as Botnet Nodes: The sheer number of insecure
connected to digital pin 13 blink on and off. except KeyboardInterrupt:
IoT devices makes them prime targets for botnet attacks. Attackers can
Arduino This cycle (HIGH, delay, LOW, delay) repeats indefinitely due to the loop() print("\nProgram stopped by user")
compromise these devices and use them to launch large-scale // This is a single-line comment. Anything after // on the same
Distributed Denial of Service (DDoS) attacks against other targets, as function, resulting in the continuous blinking of the LED. This simple example
line is ignored by the compiler.
seen with the Mirai botnet. /* clearly demonstrates the fundamental structure and execution flow of an Arduino
finally:
This is a multi-line comment. program.
13) Explain with neat example a basic structure of arduino It can span multiple lines. Q14) Write a python program on RaspberryPi to blink LED? # Clean up GPIO settings
programming? */ import RPi.GPIO as GPIO GPIO.cleanup()
from time import sleep print("GPIO cleaned up")
Arduino programming, based on C++, follows a straightforward and predictable // Define a constant for the LED pin number. Using const int is
structure. Every Arduino sketch (the term for a program) is fundamentally good practice
// as it makes the code more readable and easier to modify. 15) Explain smart city security architecture?
composed of two essential functions: setup() and loop(). const int ledPin = 13;
# Set up the GPIO mode
GPIO.setmode(GPIO.BCM) # Use Broadcom pin numbering Smart City Security Architecture
1. setup() function: void setup() {
// Initialize the digital pin as an output. A Smart City Security Architecture refers to the comprehensive framework
• This function is executed only once when the Arduino board powers up or is // This tells the Arduino that we will be sending signals out # Define the LED pin designed to protect the vast and interconnected digital infrastructure of a smart city.
reset. of this pin LED_PIN = 17 # Change this to whichever GPIO pin you're using5 Its primary goal is to ensure the confidentiality, integrity, and availability (CIA)
• It's primarily used for initialization tasks. This includes: // to control something (in this case, an LED). of data and services, safeguarding citizens, critical infrastructure, and government
pinMode(ledPin, OUTPUT);
operations from cyber threats, physical attacks, and system failures.
25 26 27 28

This architecture typically adopts a layered approach, addressing security at every The architecture must be scalable, adaptable, and interoperable, allowing diverse
level: systems to communicate securely while maintaining a unified security posture
across the entire smart city ecosystem.
1. Physical Layer Security: Focuses on protecting the physical components like
IoT sensors, cameras, smart meters, control units, and network hardware from
tampering, theft, or damage. This includes secure enclosures, tamper
detection, and physical access controls.
2. Network Layer Security: Secures the communication channels that connect
devices and systems. This involves using robust protocols (e.g., VPNs, IPSec),
strong encryption (e.g., TLS/SSL), intrusion detection/prevention systems 1) Explain Simplified IOT Architecture?
(IDS/IPS), firewalls, and network segmentation to prevent unauthorized
access, data interception, and denial-of-service (DoS) attacks. 2) Compare and contrast IT and OT?
3. Data Layer Security: Deals with the protection of collected, stored, and
processed data. Measures include data encryption at rest and in transit, data 3) Explain Drivers Behind New Network Architectures?
anonymization/pseudonymization, access control mechanisms (role-based
access control - RBAC), data integrity checks, and secure data storage 4) Explain different types of sensors with example?
solutions (e.g., blockchain for immutable records). Privacy-by-design 5) Classify actuators based on energy type?
principles are crucial here.
4. Application Layer Security: Secures the software applications and services 6) List out the limitation of smart objects in WSNs?
that provide smart city functionalities (e.g., smart traffic management, public
safety apps, utility management). This involves secure coding practices, 7) With the neat diagram, explain 6LOWPAN header stacks?
regular vulnerability assessments, application firewalls, robust authentication
(MFA), and authorization mechanisms. 8) Explain with example MQTT protocol?

Key Principles and Measures: 9) List and explain the key advantages of internet protocols?

• Cyber Resilience: Ability to withstand, recover from, and adapt to adverse 10) What are the common applications of machine learning in IOT?
cyber events. 11) Explain the elements of Hadoop with a neat diagram?
• Identity and Access Management (IAM): Strict control over who can access
what resources. 12) What are the common challenges in IOT security?
• Threat Intelligence & Monitoring: Continuous monitoring for anomalies,
real-time threat detection, and proactive response. 13) Explain with neat example a basic structure of arduino
• Incident Response: Defined procedures for handling security breaches programming?
effectively.
• Regulatory Compliance: Adherence to data protection laws (e.g., GDPR) 14) Write a python program on RaspberryPi to blink LED?
and industry standards. 15) Explain smart city security architecture?
• Supply Chain Security: Ensuring the security of third-party hardware and
software components.
29 30