From: An ensemble deep learning based IDS for IoT using Lambda architecture
Layers | Purpose | Security Issues |
---|---|---|
Perception | The perception layer represents the physical IoT devices like health monitors, security systems, smart home devices, autonomous vehicles and robotics-related devices | Eavesdropping Fake/malicious nodes Node capturing/Jamming Replay attacks Timing related attacks Social engineering attacks |
Transport | This layer is responsible for sending the collected data to either the cloud or edge device for further processing. It relies on internet gateways for moving data from the perception layer for onward processing | Denial of service (DoS) attacks Main-in-the-middle (MiTM) attacks Storage related attacks Exploitation attacks RPL attacks |
Processing | Once the data reaches the cloud or edge device the server transforms this data into information. Modern IoT architectures use ML and AI for data processing and analysis | Malware-related attacks Exhaustion attacks Resource depletion attacks |
Application | At this layer the Network administrators manage IoT device orchestration, create rule sets, and set service-level agreements for their IoT architecture | Cross site scripting (XSS) attacks Malicious code attacks Network disturbance Data loss Code Injection Buffer overflow attacks Phishing attacks Side-channel attacks |
Business | At this layer, information is transformed into business intelligence that drives decision-making by executives and stakeholders | Business logic-related attacks Zero-day attacks |