1.3.1 Functional Requirements of an IoT Architecture

An architecture is the "Blue Print" of the system. A university-grade architecture must satisfy:

• Modularity: Being able to swap out the physical link (e.g., switching from Wi-Fi to Ethernet) without rewriting the application logic.
• Scalability: The architecture must handle 10 sensors in a lab or 10 million in a city with the same fundamental structure.
• Interoperability: Supporting heterogeneous devices from different vendors.
• Reliability: Handling link failures, packet loss, and node outages gracefully.

1.3.2 The 5-Layer Integrated Model (Middleware Focus)

The 5-layer model is the primary focus for technical system design and academic examination:

1.3.3 The Fog and Edge Computing Architecture (Distributing the Load)

Modern IoT systems avoid sending all data to the cloud due to bandwidth costs and latency.

• Edge Computing: Data processing happens ON the sensor node itself. (e.g., an ESP32 performing an FFT on vibration data and only sending the "Peak Frequency" to the cloud).
• Fog Computing: Data processing happens on a local gateway (e.g., a Raspberry Pi) that serves a local cluster of sensors. This provides "Local Survivability"—if the internet goes down, the building still stays smart.

1.3.4 Security Architecture Vertical Plane

Security is not a single layer; it is a Vertical Cross-section that must protect every part of the system:

• Hardware Security: Secure Boot, JTAG locking, and Hardware Security Modules (HSM).
• Network Security: Link-layer encryption (AES-128 in Zigbee) and Network-layer security (IPsec/DTLS).
• Cloud Security: IAM (Identity and Access Management), Encryption at rest, and Virtual Private Clouds (VPC).