Skip to main content

Architecture

Arquitecture Diagram

1. Architecture Overview​

The system implements a hybrid layered architecture, combining elements of traditional layered architecture with a microservices approach for specific components. This approach aims to balance structural organization with the scalability and performance required for real-time operations.

2. Architecture Layers​

2.1. Presentation Layer​

  • Component: Mobile application
  • Technology: Kotlin for Android Automotive OS
  • Functionalities:
    • Visual interface and audible alerts
    • Navigation and monitoring
    • Data presentation to users
    • Integration with external APIs

2.2. Business Layer​

  • Technology: Python with Docker
  • Main Services:
    • Authentication Service: User authentication management
    • Delivery Service: Message delivery with persistence via Firebase
    • Statistics Handler: Event aggregation and prioritization
    • Proximity Filter: Relevant vehicle filtering based on proximity

2.3. Data Layer​

  • Main Database: PostgreSQL
    • Event Database: Historical event storage for statistics
    • Message Broker: Event flow management

2.4. Microservices Layer​

  • Architecture: Specialized microservices per event type
  • MVP Services:
    • Overtake Detection Engine: Overtaking detection
    • Speeding Detection Engine: Speeding detection
  • Expansion Strategy: Addition of more detection engines in the future

3. Data Flow and Processing​

3.1. Real-time Data Acquisition​

Sensors -> Hono (Connection Management) -> Ditto (Digital Twins Framework)

3.2. Event Processing​

Digital Twins -> Proximity Filter -> Detection Microservices -> Message

3.3. Notification Delivery​

Message Broker -> Aggregation/Prioritization -> Delivery Service -> Application

4. Infrastructure Components​

4.1. Digital Twins Framework (Ditto)​

  • Maintains real-time updated digital representations
  • Bidirectional communication via Hono
  • Foundation for event detection

4.2. Message Management​

  • Message Broker: Receives diverse events from microservices
  • Prioritization System: Classifies events by criticality

4.3. Persistence and Statistics​

  • Firebase: Persistence reinforcement for messages
  • PostgreSQL: Historical database for statistical analysis

5. Performance Considerations​

5.1. Response Time Optimization​

  • Microservices architecture to reduce congestion
  • Parallel processing of different event types
  • Proximity filtering to reduce computational load

5.2. Deployment Strategy​

  • Backend: Docker for containerization
  • Application: Android Studio for development and deployment
  • Environment: Android Automotive OS

6. Technologies and Tools​

LayerTechnologies
PresentationKotlin, Android Automotive OS, Android Studio
BusinessPython, Docker, Firebase
DataPostgreSQL
InfrastructureHono, Ditto, Message Broker

Tutors:

Group:

Institution: Telecommunications Institute of Aveiro (ITAv)