I483 Course Project - IoT Sensor Data Processing Pipeline

Overview

This repository contains the complete implementation for the I483 course assignments, covering IoT sensor data acquisition, MQTT/Kafka messaging, and stream analytics using Apache Flink. The project demonstrates a comprehensive IoT data processing pipeline from sensor data collection to real-time analytics and visualization.

Project Structure

├── kadai1_esp-idf/          # Assignment 1: ESP-IDF sensor implementation
├── kadai1_micropython/      # Assignment 1: MicroPython multi-sensor system
├── kadai2_golang_processor/ # Assignment 2: MQTT/Kafka data processing (Go)
├── kadai3_flink_kotlin/     # Assignment 3: Stream analytics (Flink + Kotlin)
├── other_messagepump_kotlin_refactoring/ # Additional message pump implementation

Assignments

Assignment 1: Sensor Data Acquisition

• ESP-IDF: BH1750 light sensor (15s intervals)
• MicroPython: Multi-sensor system (SCD41, RPR-0521rs, BH1750, DPS310)
• Features: Async management, MQTT publishing, high precision

Assignment 2: MQTT/Kafka Processing (Go)

• MQTT/Kafka: Data publishing, subscription, stream processing
• Analytics: Rolling average (5-min), CO2 threshold (700ppm), LED control
• Topics: i483/sensors/[ID]/[SENSOR]/[TYPE]

Assignment 3: Stream Analytics (Kotlin + Flink)

• Flink: Real-time processing (min/max/avg, 5-min windows)
• Visualization: Grafana dashboards, Apache IoTDB
• CEP: Occupancy detection, multi-sensor fusion
• Topics: i483-sensors-[ID]-analytics-[SENSOR]-[METRIC]-[TYPE]

Technical Stack

• Hardware: ESP32, SCD41, BH1750, RPR-0521rs, DPS310 sensors
• Languages: C (ESP-IDF), Python (MicroPython), Go, Kotlin
• Messaging: MQTT, Apache Kafka
• Analytics: Apache Flink, Grafana, Apache IoTDB
• Deployment: Docker, Docker Compose

Author

XU Pengfei (2510082) - JAIST I483 Course Project 2025

• Special Thanks to my AI Assistant(s): GPT（4o、4.1）, Claude(3.7、4), Gemini-2.5 😄