GTU 3160716 IOT and applications Winter 2025 Solved Paper & Solutions

Role of "Things" in IoT

In IoT, "Things" refer to identifiable physical or virtual objects embedded with electronics, sensors, actuators, and communication hardware that interact with the physical environment.

Key Roles:

1. Sensing & Perception: Capturing real-world physical phenomena (temperature, motion, light, pressure) and converting them into digital data.
2. Actuation: Executing physical actions based on control commands (turning on a motor, closing a valve, switching a relay).
3. Identity & Connectivity: Possessing a unique identifier (IP, MAC address, RFID tag) enabling communication over local and wide-area networks.

Significance of AI/ML in IoT (AIoT)

While IoT devices collect massive volumes of raw sensor data (Big Data), AI/ML provides the intelligence to extract actionable insights:

1. Predictive Maintenance: Analyzes vibration, acoustic, and thermal sensor trends in industrial machines to forecast equipment failures before downtime occurs.
2. Anomaly Detection: Identifies operational outliers (e.g., sudden pressure surges in smart pipelines or anomalous network traffic on smart home devices).
3. Real-time Edge Intelligence: ML models deployed on edge nodes (TinyML) allow local decision-making (e.g., facial recognition or emergency braking) with microsecond latency without depending on cloud connectivity.
4. Adaptive Resource Optimization: Dynamically optimizes energy consumption in smart buildings and smart grids based on historical usage patterns.

IoT Architecture Stack

Concrete Working Example: Smart Cold-Storage Monitoring

1. Perception Layer: A DHT22 temperature sensor connected to an ESP32 microcontroller measures warehouse temperature every 5 seconds.
2. Network Layer: The ESP32 formats the measurement into a lightweight JSON payload and transmits it over Wi-Fi using the MQTT protocol to the broker at coldstorage/unit1/temp.
3. Middleware Layer: A cloud broker (AWS IoT Core) receives the payload, logs it to a time-series database (DynamoDB), and evaluates an alert rule.
4. Application Layer: If temperature exceeds $4^\circ\text{C}$, the rule engine triggers an immediate SMS alert to the facility manager and updates the mobile dashboard.

Categories of Sensors in IoT

1. By Power Source:

• Active Sensors: Require external electrical excitation to operate (e.g., Ultrasonic, Radar, Hall Effect).
• Passive Sensors: Directly generate an electrical signal in response to stimulus without external power (e.g., Thermocouples, Photovoltaic cells, Piezoelectric sensors).

2. By Output Signal:

• Analog Sensors: Produce continuous voltage/current (e.g., LM35, LDR).
• Digital Sensors: Produce discrete binary values or transmit data over serial protocols (I2C/SPI) (e.g., DHT11, BME280).

3. By Measurand (Property): Thermal (thermistor), Acoustic (microphone), Optical (photodiode), Environmental (gas, humidity, PM2.5), Motion/Kinematic (accelerometer, gyroscope).

Interfacing Ultrasonic Sensor (HC-SR04) with a Microcontroller

Interfacing Procedure

1. Connections: Connect sensor VCC to $5\text{V}$, GND to Ground, TRIG to a digital output pin, and ECHO to a digital input pin.
2. Trigger Pulse: The microcontroller pulls TRIG HIGH for $10\,\mu\text{s}$ and returns it to LOW.
3. Sonic Burst: The sensor transmits eight $40\text{ kHz}$ ultrasonic pulses and drives ECHO HIGH.
4. Echo Measurement: ECHO remains HIGH until the reflected sound wave returns. The microcontroller measures this pulse duration ($t$ in microseconds) using a hardware timer.
5. Distance Computation:

Advantages of IoT in Healthcare (IoMT)

1. Continuous Remote Patient Monitoring: Tracks vital signs 24/7 without hospitalization.
2. Early Emergency Detection: Automated triggers for arrhythmias, falls, or sudden oxygen drops.
3. Reduced Healthcare Costs: Decreases readmission rates and bed occupancy in hospitals.
4. Improved Medication Adherence: Smart pillboxes and ingestible sensors track compliance.

Detailed Role of IoT in Healthcare

1. Remote Vital Sign Telemetry:

• Wearable biosensors continuously record ECG, pulse oximetry ($ext{SpO}_2$), blood glucose, and core temperature.
• Devices transmit telemetry via BLE to a patient's smartphone gateway, forwarding records to Electronic Health Record (EHR) platforms.

2. Smart Hospital Asset Tracking:

• RFID tags and Bluetooth beacons track defibrillators, wheelchairs, and oxygen cylinders in real time, preventing equipment misplacement.

3. Infection Control & Hygiene Monitoring:

• IoT soap dispensers and smart badges enforce hand-hygiene compliance for healthcare workers entering ICUs.

4. Chronic Disease Management:

• Continuous Glucose Monitors (CGM) communicate directly with automated insulin pumps to form closed-loop artificial pancreas systems.

Short Note: Zigbee

Zigbee is a low-power, low-data-rate wireless mesh networking standard designed for personal area networks (WPAN) based on the IEEE 802.15.4 physical and MAC layer.

Key Specifications:

1. Operating Frequency: $2.4\text{ GHz}$ ISM band worldwide (also $868\text{ MHz}$ EU, $915\text{ MHz}$ US).
2. Data Rate & Range: $250\text{ kbps}$; typical indoor range $10\text{--}75\text{ meters}$.
3. Mesh Topology: Supports self-healing, multi-hop mesh routing connecting up to 65,000 nodes.
4. Device Roles:

• Zigbee Coordinator (ZC): One per network; forms the network and stores security keys.
• Zigbee Router (ZR): Intermediate node that routes data packets between nodes.
• Zigbee End Device (ZED): Low-power battery node (sensor) that sleeps periodically.

Bluetooth Low Energy (BLE) Stack Architecture

Explanation of Components

1. Physical Layer (PHY): Operates in the $2.4\text{ GHz}$ ISM band across 40 RF channels (3 advertising channels, 37 data channels) using GFSK modulation.
2. Link Layer (LL): Controls RF packet transmission and manages 5 operational states: Standby, Advertising, Scanning, Initiating, and Connected.
3. HCI (Host Controller Interface): Standardized communication boundary allowing the Host and Controller to run on separate physical chips via UART or SPI.
4. L2CAP: Multiplexes data channels, handles packet fragmentation/reassembly, and negotiates Quality of Service.
5. ATT (Attribute Protocol): Client/server protocol where the server exposes state as "Attributes" consisting of a 16-bit Handle, UUID, and Value.
6. GATT (Generic Attribute Profile): Organizes attributes into hierarchical Profiles, Services, and Characteristics.
7. GAP (Generic Access Profile): Defines how BLE devices broadcast, discover each other, and establish connections (Broadcaster, Observer, Peripheral, Central).

Cloud Deployment Models

1. Public Cloud: Infrastructure owned and managed by third-party cloud providers (AWS, Azure, Google Cloud), delivered over the public internet with multi-tenant resource sharing.
2. Private Cloud: Cloud infrastructure provisioned for exclusive use by a single organization, offering maximum data governance and security behind an enterprise firewall.
3. Hybrid Cloud: Combines private and public cloud environments, bound together by standardized technology allowing data and application portability.
4. Community Cloud: Infrastructure shared collaboratively by multiple organizations possessing shared regulatory, security, or mission goals (e.g., healthcare or government agencies).

What is Fog Computing?

Fog Computing is an architectural paradigm that extends cloud computing storage, compute, and networking capabilities closer to the data-producing IoT devices at the local network edge.

Key Characteristics of Fog Computing

1. Low Latency: Data processing occurs on local gateway nodes within milliseconds, avoiding transatlantic cloud roundtrips.
2. Geographical Distribution & Edge Localization: Distributed nodes deployed across factories, intersections, and campuses.
3. Bandwidth Conservation: Aggregates and filters redundant high-frequency sensor streams locally, transmitting only summaries or anomalies to the central cloud.
4. Offline Resilience: Edge operations and local automated safety shutdowns continue functioning even when internet connectivity fails.

Case Study: Adafruit IO Cloud Platform

1. Overview

Adafruit IO is a cloud service designed by Adafruit specifically for makers, students, and IoT developers to stream, log, and visualize sensor data and control hardware remotely.

2. Core Architectural Components

• Feeds: Core data channels holding sensor readings (e.g., temperature, humidity, relay_status) along with timestamps and metadata.
• Dashboards: Customizable web interfaces containing drag-and-drop widgets (gauges, line charts, toggles, sliders, color pickers, maps).
• Triggers (Reactive Automation): Event-based rules triggering actions (e.g., sending an email or webhook when a feed value exceeds a limit).
• Communication APIs:
• REST API: Standard HTTP endpoints for polling and posting data.
• MQTT Broker: Real-time bi-directional publish/subscribe broker accessible on io.adafruit.com on port 1883 (or 8883 for TLS).

3. Implementation Workflow Example

1. An ESP32 measures ambient room luminosity using an LDR sensor.
2. The device authenticates using an AIO_USERNAME and AIO_KEY.
3. Over MQTT, it publishes light values to topic username/feeds/room-light.
4. A dashboard gauge updates instantaneously; if luminosity drops below 50 lux, a trigger sends an alert to switch on the porch light.

How IoT Improves Warehouse Management

1. Automated Inventory Tracking: Passive and active RFID tags and BLE beacons on pallets eliminate manual barcode scanning and prevent misplaced stock.
2. Environmental Monitoring: Wireless sensors continuously record temperature, humidity, and volatile organic compounds (VOCs) to preserve perishable goods.
3. Autonomous Guided Vehicles (AGVs): Fleet of automated forklifts communicate via Wi-Fi/5G to optimize route navigation and material transport.
4. Predictive Equipment Health: Vibration sensors on conveyor belts identify motor bearing wear before breakdown occurs.

What is Arduino?

Arduino is an open-source electronics platform based on easy-to-use hardware microcontrollers and an integrated software development environment (IDE).

Key Features of Arduino Architecture (e.g., Arduino UNO / ATmega328P)

1. 8-bit AVR RISC CPU: Operates at $16\text{ MHz}$ clock speed with single-cycle execution of most instructions.
2. Memory Hierarchy:

• Flash Memory: $32\text{ KB}$ for program code storage (0.5 KB used by bootloader).
• SRAM: $2\text{ KB}$ for volatile runtime variables and stack.
• EEPROM: $1\text{ KB}$ for non-volatile parameter persistence across power cycles.

3. I/O Capabilities:

• 14 Digital I/O Pins (6 support hardware Pulse-Width Modulation - PWM).
• 6 Analog Input Pins connected to an internal 10-bit Successive Approximation ADC.

4. Communication Peripherals: Hardware USART (Serial), SPI, and I2C (Two-Wire Interface - TWI) support.

Arduino Traffic Light Controller Implementation

Circuit Connections

• Red LED Anode: Digital Pin 10 (via $220\,\Omega$ current limiting resistor)
• Yellow/Orange LED Anode: Digital Pin 9 (via $220\,\Omega$ resistor)
• Green LED Anode: Digital Pin 8 (via $220\,\Omega$ resistor)
• Cathodes (All LEDs): Common Ground (GND)