IoT Architecture and Frameworks Lecture Notes

Learning Objectives

By the end of this lesson, you will be able to:

• Explain the IoT architecture and frameworks
• Describe IoT interoperability as a design consideration
• Discuss industry-aligned use cases

IoT Device Architecture

Layers of Device Architecture

1. Base Layer

   • Consists of IoT devices (e.g., sensors)
   • Capable of sensing, computing, and connecting

2. IoT Gateway/Aggregation Layer

   • Aggregates data from various sensors
   • Forms the definition engine to set rules for data aggregation

3. Processing Engine/Event Processing Layer

   • Cloud-based layer with algorithms and data processing elements
   • Processes data obtained from sensor layer and displays it on dashboards

4. Application/API Management Layer

   • Interface between third-party applications and infrastructure
   • Managed by device managers and identity access managers for security

IoT Reference Architecture

Device Layer

• Main component with interconnected devices (e.g., Bluetooth, Zigbee, Raspberry Pi)
• Directly connected to communication layers

Communication Layer

• Includes REST protocols and application-level protocols
• Tightly coupled with device layer, generating significant data

Bus/Aggregation Layer

• Acts as a message broker
• Supports HTTP server and/or MQTT broker
• Aggregates communications via gateway and bridges different protocols

Event Processing and Analytics Layer

• Drives data transformation and event processing
• Stores data in a database
• Creates a web-based engine for external API interaction
• Enables dashboard creation for analytics

Identity Layer

• Cybersecurity capabilities including policy control and OAuth 2.0
• Identity services (XACML PDP, user directory such as LDAP)

IoT Reference Frameworks

• Common framework: ISO 30141
  • Provides vocabulary, reusable designs, and best practices
  • Contains secure application standards to benefit organizations

IoT Standardization and Design Considerations

Key IoT Standards

• M2M: Machine-to-machine service layer
• Contiki: Open-source OS for low-cost, low-power IoT microcontrollers
• Light OS: Unix-like OS for wireless sensor networks
• Random Phase Multiple Access: Proprietary standard for IoT connections
• Sigfox: Proprietary low power, low throughput for IoT and M2M communications

Interoperability Challenges

• Coexistence of multiple systems and formats
• Manufacturing designs hinder global agreements on standards
• Low-power devices face challenges in data exchange over lossy networks

Design Considerations for IoT Solutions

• Wireless capability, functionality, interoperability
• Secure storage, immediate boot capacity, device categorization
• Bandwidth, cryptographic control, power management
• Establish a dispute resolution mechanism for failures

Integration Stages of IoT Processes

1. Stage 1: Networked things (wireless sensors and actuators)
2. Stage 2: Sensor data aggregation systems and analog to digital conversion
3. Stage 3: Edge IT systems for data pre-processing
4. Stage 4: Data center and cloud for analysis, management, and storage

IoT Architecture Areas

1. Client-side IoT device layer
2. Server-side IoT gateway layer
3. IoT platform layer – pathways for client/operator connection

Centralized vs Decentralized Architectures

• Centralized Architecture: Managed from a single hub, associated with cloud services.
• Decentralized Architecture: Autonomous communication without a central hub, more suited for IoT applications.

Use Cases

Smart Farming IoT Design

• Requires precise architecture, efficiency, and product quality.
• Key components:
  1. Data Engine: Robust processing engine for data storage and output.
  2. Hardware: Durable and maintainable, self-fixing algorithms are advantageous.
  3. Mobile Access: Smartphone applications for offline/online access.
  4. Cloud Infrastructure: With an edge layer for effective smart farming.

Diabetes Management with IoT

• Use IoT to monitor blood sugar levels remotely.
• Hardware: Selecting suitable sensors for integration with glucometer.
• Software: Requires algorithms for service management.

Key Takeaways

• Ability to explain IoT reference architectures and frameworks.
• Understanding of IoT interoperability and its design considerations.
• Awareness of industry-aligned use cases.