IoT Physical Design, Communication Models, and Technologies

Posted on Mar 20, 2025 in Technology

1. Physical Design of IoT

The physical design of IoT consists of IoT devices (things) that have unique identities and can perform sensing, actuation, and monitoring.

IoT Device Capabilities:

  • Collect Data: Sensors gather data from the environment.
  • Exchange Data: Devices communicate with other devices/applications via networks.
  • Process Data Locally or in the Cloud: Some devices perform on-device processing, while others send data to centralized servers.

IoT Protocols (Layers):

  • Link Layer – Defines how data is physically transmitted (Wi-Fi, Zigbee, Ethernet).
  • Network Layer – Manages IP addressing and packet routing (IPv4, IPv6, 6LoWPAN).
  • Transport Layer – Handles end-to-end communication (TCP, UDP).
  • Application Layer – Defines interaction between applications and devices (HTTP, MQTT, CoAP).

2. IoT Functional Blocks

  • Devices – Sensors, actuators, controllers.
  • Communication – Data transfer between IoT components.
  • Services – Monitoring, control, and data processing.
  • Management – System governance and configuration.
  • Security – Data encryption, authentication, and integrity.
  • Application – User interface for monitoring and control.

3. IoT Communication Models

Request-Response Model:

The client sends a request, and the server responds. Example: HTTP-based web communication.

Publish-Subscribe Model:

Data publishers send messages to a broker, and subscribers receive updates. Example: MQTT for IoT messaging.

Push-Pull Model:

Producers push data into queues, and consumers pull when needed. Example: Message queue-based IoT systems.

Exclusive Pair Model:

Persistent, full-duplex connection between client and server. Example: WebSockets for real-time IoT data transfer.

4. IoT Characteristics

  • Self-Adapting: Devices adjust based on environmental changes.
  • Self-Configuring: Auto-updates and configurations.
  • Interoperability: Supports multiple communication protocols.

5. IoT Enabling Technologies

  • Wireless Sensor Networks (WSN) – Sensors gather and transmit data.
  • Cloud Computing – IoT data storage and processing in the cloud.
  • Big Data Analytics – Extracts meaningful insights from IoT data.

6. Levels of IoT

The levels of IoT are categorized based on the complexity of deployment and cloud involvement:

  • Level 1: A single IoT device operates independently, performing sensing or actuation without cloud integration. Example: A smart light that turns on/off based on motion.
  • Level 2: A single IoT node processes data locally but uses cloud storage for data retention. Example: A smart irrigation system that stores soil moisture data in the cloud.
  • Level 3: A single IoT node exists, but both data storage and data analysis happen in the cloud. Example: A smart thermostat that collects temperature data and sends it to the cloud for analysis.
  • Level 4: Multiple IoT nodes perform local processing while storing data in the cloud. A remote application monitors and controls them. Example: A smart traffic monitoring system that processes local traffic data but sends reports to a central cloud.
  • Level 5: A wireless sensor network (WSN) consisting of multiple interconnected nodes sends data to the cloud, where storage and analysis take place. Example: Smart agriculture systems using multiple sensors to monitor environmental conditions.
  • Level 6: A complex system with multiple independent end nodes and a centralized cloud-based analytics platform. A central controller manages and optimizes system operations. Example: Industrial automation in a smart factory with multiple IoT-enabled devices.

7. IoT vs. M2M

M2M (Machine-to-Machine) refers to direct communication between devices using proprietary or non-IP-based protocols, typically in industrial applications. M2M systems store and process data locally without cloud involvement.

IoT (Internet of Things) involves device connectivity via IP-based networks (such as Wi-Fi or MQTT) with cloud-based storage and analytics. IoT devices provide more flexibility and scalability for applications like smart homes, healthcare, and industrial automation.

While M2M is generally hardware-focused with fixed communication paths, IoT integrates software-driven analytics, cloud computing, and remote access for real-time decision-making.

Module 2

1. Sensors and Actuators

Sensors:

Convert physical quantities into electronic signals.

Examples: Temperature sensor, humidity sensor, accelerometer.

Actuators:

Convert electrical signals into mechanical action.

Examples: Servo motors, stepper motors, solenoids.

Classification:

  • Active vs. Passive – Requires external power (Active) vs. self-powered (Passive).
  • Contact vs. Non-contact – Requires direct contact vs. remote sensing.
  • Absolute vs. Relative – Measures absolute value vs. relative difference.

2. Wireless Sensor Networks (WSN)

Consist of nodes that sense, process, and transmit data wirelessly.

Challenges:

  • Limited power, processing, and memory.
  • Large-scale deployment for accurate monitoring.

Communication Patterns:

  • Event-Driven: Data is transmitted only when a specific event occurs.
  • Periodic: Data is transmitted at regular intervals.

Applications: Smart cities, environmental monitoring, and industrial automation.

3. Network Function Virtualization (NFV) Use Case

Virtualizes network functions, reducing dependence on hardware routers and switches.

Example Use Case: IoT-based smart city networks using virtual firewalls and routers.

Advantages:

  • Cost-effective – No need for specialized hardware.
  • Scalability – Supports dynamic network changes.
  • Energy-efficient – Reduces power consumption.

4. Conventional Networking Issues & SDN

Issues in Conventional Networking Architectures:

Complex Network Management:

  • Traditional networks rely on manual configurations and vendor-specific hardware.
  • Hard to modify settings for thousands of IoT devices dynamically.

High Costs of Network Equipment:

  • Requires expensive, specialized hardware (routers, firewalls, switches).
  • Frequent hardware upgrades increase costs.

Limited Scalability:

  • Conventional networks are not designed for large-scale IoT deployments.
  • Cannot handle millions of IoT devices efficiently.

Lack of Automation:

  • Cannot dynamically adapt to changing traffic loads.
  • Requires manual intervention for updates and troubleshooting.

Security Challenges:

  • Network-wide security policies are difficult to enforce.
  • Vulnerable to DDoS attacks, unauthorized access, and data breaches.

How SDN Solves These Issues:

Centralized Network Control:

  • SDN separates the control plane (decision making) from the data plane (packet forwarding).
  • A centralized SDN controller manages the entire network dynamically.

Programmable Network Management:

  • Uses Open APIs to enable automated network management.
  • Example: If network congestion occurs, the SDN controller automatically reroutes traffic.

Reduced Hardware Dependency:

  • SDN allows software-based routing and switching using generic hardware.
  • Reduces the need for expensive proprietary routers and firewalls.

Improved Scalability:

  • Can handle millions of IoT devices by intelligently managing traffic flows.
  • Works efficiently in cloud data centers.

Enhanced Security:

  • Centralized security policies can be implemented across the network.
  • Dynamically detects and mitigates DDoS attacks and network anomalies.

Wireless Sensor Network (WSN) Working

What is a Wireless Sensor Network (WSN)?

A Wireless Sensor Network (WSN) is a network of sensor nodes that collect, process, and transmit data wirelessly for monitoring and control.

Components of WSN:

  • Sensor Nodes:

    • Devices with sensors, microcontrollers, and wireless modules.
    • Example: Temperature, humidity, motion sensors.

  • Sink (Gateway):

    • Collects data from multiple sensor nodes.
    • Acts as a bridge between sensors and the internet.
    • Example: Raspberry Pi, IoT gateway routers.

  • Communication Network:

    • Transmits sensor data using Zigbee, Wi-Fi, LoRa, Bluetooth, or 5G.

  • Cloud Server:

    • Stores and processes sensor data.
    • Runs AI/ML models for predictions and analytics.

  • User Interface (Mobile App/Web Dashboard):

    • Displays sensor data.
    • Sends commands to actuators.

How WSN Works?

  • Sensing:
    • Sensors collect real-time environmental data (e.g., temperature, pressure).
  • Data Transmission:
    • Sensor nodes send data to the sink (gateway) wirelessly.
  • Data Processing:
    • The gateway filters and processes the collected data.
  • Cloud Storage & Analysis:
    • Data is sent to the cloud for storage and AI-based analytics.
  • User Access & Control:
    • Users monitor the data via mobile apps, dashboards, or alerts.
    • Actuators can be triggered remotely (e.g., turning on an alarm if motion is detected).

5. Smart Device Block Diagram

Components of a Smart Device:

  • Processing Unit (Microcontroller/Processor)
    • Controls the smart device functions.
    • Processes data from sensors and makes decisions.
    • Examples: ESP32, Raspberry Pi, STM32, Arduino.
  • Sensors & Actuators
    • Sensors collect environmental data (e.g., temperature, humidity, motion).
    • Actuators perform actions (e.g., turning on a motor, adjusting brightness).
  • Communication Unit (Wireless/Wired)
    • Enables Wi-Fi, Bluetooth, Zigbee, LoRa, 4G, 5G connectivity.
    • Example: Wi-Fi module ESP8266, Bluetooth module HC-05.
  • Power Source
    • Supplies energy to the device.
    • Battery-powered, solar-powered, or AC-powered.
  • User Interface (Display, Mobile App, Web Dashboard)
    • Allows users to monitor and control the device.
    • Example: LCD screens, mobile applications, voice assistants.