Department of Electronics and Communication Engineering

End Semester Examination November 2025

22EC701 – IOT Design
Year/Semester: IV/VII Answer Key Class: ECE
Part A – Answer All (10*2=20 Marks)
1. Identify the domain in which physical components operate in coordination with
computational logic. [CO1,K2]
The domain in which physical components (like sensors, actuators, and machines) operate
together with computational logic (such as controllers, processors, and software) is called a
Cyber-Physical System (CPS) domain.

2. State the category of the following elements (MQTT, COAP, and HTTP) which requires
for communication? [CO1,K1]
MQTT, CoAP, and HTTP all belong to the category of application-layer communication
protocols used for data exchange in IoT and networked systems.

3. A microcontroller executing a fixed task signals membership in which category of

computing units? [CO2,K2]
A microcontroller that executes a fixed, predefined task belongs to the category of embedded
computing units (embedded systems), because it is dedicated to performing a specific control
function within a larger system.

4. A high-speed synchronous protocol is preferred for display modules. Identify the

protocol type involved. [CO2,K2]
The high-speed synchronous protocol commonly used for display modules is SPI (Serial
Peripheral Interface). SPI supports fast, full-duplex data transfer with a clock signal, making
it suitable for driving LCD/TFT displays efficiently.

5. Provide the type of signal involved in microcontroller to reads 1 or 0 from a pin.

[CO3,K2]
The type of signal a microcontroller reads as 1 or 0 from a pin is a digital signal. A logic
HIGH represents ‘1’ and a logic LOW represents ‘0’.

6. Summarize the signals handling in DAC,controlling LED brightness. [CO3.K2]

In DAC (Digital-to-Analog Converter)–based LED control, the digital signal from the
microcontroller is converted into a continuous analog voltage. By varying this output voltage
level, the current through the LED changes, which smoothly adjusts the LED brightness
instead of switching it simply ON or OFF.

7. List out the characteristic features of WSN. [CO4,K1]

 Low power and low-cost sensor nodes deployed in large numbers.
 Wireless communication with self-configuring and self-healing capability, enabling
nodes to form ad-hoc networks and cooperate in sensing and data forwarding.
8. Justify the importance of machine learning in loT. [CO4.K2]
Machine learning is important in IoT because it helps devices analyze the large amount of
data they collect and automatically detect patterns or abnormalities. This allows IoT systems
to make smarter decisions, improve accuracy, enable prediction (like fault detection), and
reduce human intervention.

9. Compare RAM and Flash Memory. [CO5,K2]

RAM is a volatile memory used for temporary data storage while a device is running; data is
lost when power is turned off. Flash memory is non-volatile storage used to permanently store
programs and data; it retains data even without power.

10. In modern automated industries, what essential role do sensing elements play in
maintaining efficiency and operational accuracy? [CO5,K2]
In modern automated industries, sensing elements continuously monitor physical parameters
such as temperature, pressure, position, and flow. By converting these real-world conditions
into electrical signals for control systems, they ensure accurate process control, quick fault
detection, and high operational efficiency.

Part B – Answer All (5*16=80 Marks)

11)a) Compare lot systems and cyber-physical systems with suitable parameters. [CO1,K2]
 IoT systems and Cyber-Physical Systems (CPS) are closely related but differ in
scope, integration, and functionality. IoT mainly focuses on connecting devices to
the internet so they can collect, transmit, and sometimes process data using
protocols like MQTT or HTTP, with cloud platforms commonly handling analytics
and control.
 In contrast, CPS tightly integrates computational logic with physical processes so
that sensing, computation, control, and actuation occur in a closed feedback loop,
often in real time. IoT devices generally emphasize data communication and
remote monitoring, whereas CPS emphasizes system reliability, autonomy, and
coordination between hardware and software for critical operations such as
industrial automation, smart grids, medical systems, and autonomous vehicles.
 CPS systems normally require high precision, deterministic timing, safety, and
control theory principles, while IoT may tolerate delays since it is often used for
monitoring and data logging. In terms of architecture, IoT relies heavily on cloud-
based processing, whereas CPS can use edge or embedded controllers for faster
responses.
 Overall, while IoT connects the physical world to the digital world, CPS goes a
step further by deeply integrating computation and control with physical actions to
achieve intelligent, self-regulated behavior.

Or
11)b) Examine different types of lol systems with relevant use-case mapping. [CO1,K2]
 Different types of IoT systems can be classified based on their application
domains and functional roles, each mapped to practical use-cases. Consumer
 IoT includes smart homes, wearables, and personal health devices where
sensors and mobile apps help users monitor fitness, control lighting, or manage
appliances remotely. Industrial IoT (IIoT) involves smart factories, predictive
maintenance, and process automation, where machines, robots, and sensors
share real-time data to improve productivity and reduce downtime.
 Healthcare IoT supports remote patient monitoring, smart medical devices, and
hospital automation, allowing continuous health tracking and early diagnosis.
Smart City IoT includes smart traffic systems, waste management,
environmental monitoring, and smart lighting to enhance urban services and
sustainability.
 Agricultural IoT uses smart irrigation, soil monitoring, and livestock tracking to
increase crop yield and resource efficiency. Commercial IoT covers retail
automation, smart supply chains, inventory tracking, and connected logistics
for better customer service and operational transparency.
 Finally, Environmental IoT monitors air quality, weather, and disaster-prone
regions for early warning systems. Together, these categories show how IoT
transforms everyday environments into intelligent, data-driven systems that
improve efficiency, safety, and decision-making.

12)a) Evaluate embedded hardware choices for high-performance IoT designs.[CO2 K3]
 Evaluating embedded hardware choices for high-performance IoT designs
involves understanding how different components support speed, reliability,
power efficiency, and connectivity needs. Microcontrollers (like ARM Cortex-
M) are suitable for low-power sensing and control, while more powerful
processors or SoCs (like Cortex-A or custom AI chips) are preferred for edge
analytics, image processing, or machine learning.
 Memory size and type are important because real-time data handling and
buffering require sufficient RAM, and firmware or application storage depends
on Flash capacity. Hardware must also support suitable communication
interfaces such as Wi-Fi, LTE, BLE, LoRa, or Ethernet based on data rate,
range, and energy constraints.
 Power management features such as sleep modes, low-power states, and
efficient regulators are critical for battery-operated IoT devices. Sensor and
peripheral compatibility, including ADCs, timers, GPIOs, and security modules
(like TPM or secure boot), ensures system robustness and data protection.
Finally, hardware should be scalable, cost-effective, thermally stable, and
supported by strong development tools and community resources to ensure
faster prototyping and reliable deployment in real-world IoT environments.

Or
12)b) With suitable examples, compare RS485 and RS232 for industrial loT
applications. [CO2,K3]
 RS232 and RS485 are both serial communication standards used in industrial
IoT, but they differ significantly in capability and application suitability. RS232
supports point-to-point communication over short distances (typically up to ~15
meters) and is commonly used for simple device connections such as between a
PC and a sensor module, barcode scanner, or PLC console port.
 It uses single-ended signaling, making it more vulnerable to electrical noise and
 less reliable in harsh factory environments. In contrast, RS485 supports multi-
drop communication, allowing up to 32 devices (or more with repeaters) on the
same bus over long distances up to ~1200 meters, making it ideal for
distributed industrial IoT networks like building automation, Modbus-based
control systems, and smart energy meters.
 RS485 uses differential signaling, which provides excellent noise immunity and
stable data transfer in electrically noisy environments with motors and heavy
machinery.
 While RS232 is simpler and cheaper for short-range, single-device links,
RS485 is preferred in industrial IoT for scalable, robust, long-distance, and
multi-node communication, especially where reliable field-level networking is
required.

13)a) Demonstrate the HTTP-based data exchange method in a small IoT setup.
[CO3,K3]
 In a small IoT setup, HTTP-based data exchange can be demonstrated using a
sensor node, a microcontroller with Wi-Fi (such as ESP8266/ESP32), and a
web server or cloud service. The sensor continuously measures a parameter like
temperature or humidity and the microcontroller formats this reading into an
HTTP request.
 Using the HTTP POST method, the device sends the data as a message body
(often in JSON format) to a server URL where it is stored or processed, while
the HTTP GET method can be used by a web dashboard or mobile app to
retrieve and display the latest readings.
 Each communication follows the client–server model where the IoT device acts
as the client and the server responds with a status code (such as 200 OK)
confirming successful delivery. This approach allows easy integration with web
technologies, supports RESTful APIs, and enables logging, analytics, and alerts
through simple web services.
 For example, a greenhouse monitoring system can periodically send sensor
values to a cloud API using HTTP POST, and the farmer can view real-time
data on a webpage via HTTP GET, demonstrating reliable, human-readable,
and platform-independent communication in IoT.

Or
13)b) Implement the communication approach aligned with sensor node constraints,
involving either wired or wireless interfacing. [CO3.K3]
 Implementing a communication approach aligned with sensor node constraints
requires careful consideration of power consumption, data rate, range, and
network topology. For low-power sensor nodes, wireless interfaces like LoRa,
Zigbee, or BLE are preferred, as they support long battery life and can handle
distributed deployments without physical cabling; for example, LoRa-based soil
moisture sensors can transmit data over several kilometers to a central gateway
in smart agriculture.
 In contrast, wired interfaces like RS485 or Ethernet are suitable when high
reliability, low latency, and noise immunity are critical, such as in industrial
 automation or factory-floor monitoring, where sensor nodes connect over multi-
drop buses to PLCs.
 The sensor node firmware manages the communication protocol, ensuring data
is periodically transmitted without exceeding energy constraints, and includes
error detection or acknowledgment mechanisms to maintain reliability.
 Additionally, addressing methods, buffering, and lightweight message formats
(like JSON or CBOR) are chosen to match the limited memory and processing
capabilities of the node. By selecting wired or wireless methods based on
deployment environment, distance, power availability, and network scale, the
communication system efficiently supports real-time monitoring, control, and
integration into larger IoT ecosystems.

14a) Compare traditional programming and machine learning approaches and justify
which one is suitable for easy access. [CO4,K2]
 Traditional programming and machine learning (ML) approaches differ
fundamentally in how they handle data and decision-making. In traditional
programming, developers write explicit rules and algorithms to process inputs
and produce outputs, making it predictable, transparent, and easy to debug; for
example, a temperature-controlled fan is programmed to switch on if the
temperature exceeds a set threshold.
 However, this approach struggles with complex, variable, or unstructured data.
In machine learning, the system learns patterns and relationships from historical
data without requiring explicit rules, enabling it to make predictions, detect
anomalies, or adapt to changing conditions, such as predicting energy usage
trends or detecting faults in industrial IoT devices.
 ML is more flexible and scalable for tasks with large, dynamic datasets, but it
requires computational resources, model training, and careful handling of data
quality. For easy access and simple applications, traditional programming is
often more suitable because it is straightforward to implement, understand, and
maintain without needing specialized knowledge in data modeling or training,
whereas ML is better suited for advanced, data-driven intelligence tasks where
adaptability and prediction are critical.

Or
14b) Analyze and explain the reason in detail about the deep learning [ suitability for
high-volume loT data]. [CO4,K2]
 Deep learning (DL) is highly suitable for high-volume IoT data because it can
automatically extract complex features and patterns from large, high-
dimensional datasets without manual intervention.
 IoT deployments, such as smart cities, industrial automation, and connected
vehicles, generate massive amounts of data from sensors, cameras, and devices,
which often include unstructured formats like images, audio, or time-series
signals.
 Traditional analytics or simple machine learning may struggle to handle this
scale and complexity, but deep learning models, especially convolutional neural
networks (CNNs) for image/video data or recurrent neural networks (RNNs)
and transformers for time-series and sequential data, can efficiently process and
 learn from these inputs.
 DL models can perform tasks like anomaly detection, predictive maintenance,
traffic prediction, and real-time decision-making with higher accuracy and
robustness. Additionally, the use of edge computing and cloud acceleration
allows DL to process data near the source or in distributed systems, reducing
latency and bandwidth requirements.
 Therefore, deep learning’s ability to scale, adapt, and uncover hidden
correlations in vast IoT datasets makes it particularly suitable for applications
requiring intelligence from high-volume, heterogeneous data streams

15a) .Evaluate actuator integration methods in an lot control system. [CO5,K4]

 Evaluating actuator integration methods in an IoT control system involves
analyzing how actuators are connected, controlled, and coordinated with
sensors, controllers, and communication networks to achieve precise and
reliable system operation.
 Actuators such as motors, solenoids, valves, and relays can be integrated using
direct interfacing with microcontrollers, through driver circuits or motor
controllers, or via networked interfaces like Modbus, CAN, or wireless
protocols.
 Direct interfacing is simple and cost-effective for small-scale systems but may
limit scalability and precision, whereas driver circuits provide proper
voltage/current handling, isolation, and smooth control, improving actuator
lifespan and system reliability.
 Networked integration allows remote control, real-time monitoring, and
synchronization of multiple actuators across distributed locations, essential in
industrial automation, smart grids, or robotic applications.
 The integration method must consider response time, power requirements,
communication latency, and feedback mechanisms from sensors for closed-loop
control, ensuring that actuators respond accurately to commands under varying
operating conditions.
 Additionally, safety features like overcurrent protection, fail-safe modes, and
diagnostic reporting enhance system robustness. By evaluating these factors,
designers can select actuator integration strategies that optimize efficiency,
reliability, and flexibility, ensuring seamless operation of IoT control systems in
complex industrial or smart environments.

Or
15b) An assessment of communication protocols required for peripheral module
interaction can be developed? [COS,K4]
 An assessment of communication protocols for peripheral module interaction in
IoT systems involves analyzing the data transfer requirements, reliability, speed,
distance, and power constraints of the connected devices.
 Protocols such as I2C, SPI, UART, CAN, and Modbus are commonly used for
wired peripheral communication, where I2C suits short-distance, multi-device
connections with moderate speed, SPI provides high-speed point-to-point
communication, and UART enables simple asynchronous data exchange.
 For networked or distributed peripherals, wireless protocols like BLE, Zigbee,
LoRa, and Wi-Fi are assessed based on range, data throughput, energy
 consumption, and network topology. Each protocol must be evaluated for
latency, error detection, scalability, and compatibility with sensors and
actuators, ensuring accurate and timely data exchange in real-time control
systems.
 Additionally, factors like interrupt handling, bus arbitration, multi-master
support, and security mechanisms play a critical role in selecting the right
protocol. By systematically comparing these parameters against the application
requirements, a robust communication strategy can be developed that ensures
efficient, reliable, and synchronized operation of peripheral modules,
supporting both small-scale IoT devices and large industrial deployments.

Part C – Answer All (1*15=15 Marks)

16.a) A smart transportation system uses sensors, cloud connectivity, and automated
control. Evaluate how these components interact to form an loT-enabled CPS
ecosystem. [CO1,K4]
 In a smart transportation system, sensors, cloud connectivity, and automated
control interact to form a fully IoT-enabled Cyber-Physical System (CPS)
ecosystem by continuously linking the physical world with computational
intelligence. Sensors such as cameras, LIDAR, GPS modules, and traffic flow
detectors collect real-time data on vehicle positions, road conditions,
congestion, and environmental parameters.
 This data is transmitted via wired or wireless networks to cloud platforms,
where it is aggregated, processed, and analyzed using algorithms, machine
learning, or predictive models to generate actionable insights. Automated
control systems, including traffic signals, adaptive speed limits, autonomous
vehicles, and route optimization modules, receive these insights and execute
real-time decisions, adjusting traffic flow, rerouting vehicles, or controlling
signals to improve safety and efficiency.
 Feedback loops are crucial, as sensor updates continuously inform the
computational layer, allowing the system to adapt dynamically to changing
conditions. The tight integration of sensing, computation, communication, and
actuation ensures high reliability, low latency, and coordinated control, which
are key characteristics of CPS.
 Furthermore, cloud connectivity enables remote monitoring, historical data
analysis, and system scalability, while security protocols maintain data integrity
and protect against cyber threats. Collectively, these components create a
responsive, intelligent, and self-regulating transportation ecosystem,
demonstrating how IoT-enabled CPS can enhance mobility, reduce congestion,
and support smart city initiatives.

Or
b) A city plans to deploy loT for waste, traffic, and water management. Propose an
integrated multi-level IoT architecture suitable for this requirement and justify the
design choices.[COÍ,K4]
 For a city planning to deploy IoT for waste, traffic, and water management, an
integrated multi-level IoT architecture can be proposed with three primary
 layers: perception, network, and application/analytics.
 The perception layer includes diverse sensors and actuators such as smart bins
with fill-level sensors, traffic cameras and vehicle counters, and water flow and
quality sensors; these collect real-time data from the urban environment.
 The network layer employs a combination of wireless protocols (LoRaWAN for
long-range low-power devices like waste sensors, BLE/Zigbee for localized
networks, and 4G/5G or fiber for high-data traffic sources like cameras) and
wired protocols where reliability is critical, ensuring secure and low-latency
communication.
 The edge computing layer preprocesses data locally to reduce bandwidth,
enable rapid response (e.g., traffic signal adjustments or leak detection), and
filter irrelevant data before transmission to the cloud. The cloud/application
layer integrates advanced analytics, machine learning, and visualization
dashboards for city administrators to monitor trends, predict demand, optimize
resource allocation, and provide citizen services.
 Redundancy, cybersecurity, and scalability are incorporated to handle device
failures, cyber threats, and future urban expansion.
 This multi-level architecture ensures efficient data collection, reliable
communication, real-time processing, and actionable insights, enabling a smart
city ecosystem where waste collection is optimized, traffic congestion
minimized, and water resources monitored intelligently

Faculty HoD/ECE DEAN