Pi 5 Anomaly Aware MQTT Broker — Build a Self-Filtering Sensor Network

Every part needed, pre-tested for compatibility, with an AI build companion trained on this exact project. Shipped from Bengaluru in 3-5 days.

Transform a Raspberry Pi 5 into an MQTT broker that doesn’t just pass data—it judges it. With three ESP32 sensor nodes streaming temperature, humidity, and vibration metrics, this kit trains you to deploy an Isolation Forest machine learning model at the edge. Anomalous payloads are automatically quarantined before they ever reach your dashboard, mimicking the integrity layers found in industrial SCADA systems. Whether you’re preparing for the Smart India Hackathon, a B.Tech mini-project, or a demo at an ATL Tinkering Lab, you’ll leave with a prototype that colleges and judges remember.

What You'll Build

A fully containerized Mosquitto broker on Raspberry Pi 5, receiving real-time sensor telemetry from three ESP32 boards over WiFi. Inside the broker pipeline, a Python service scores each incoming message using a trained Isolation Forest model. Normal data flows to a clean MQTT topic; anything flagged as an outlier gets rerouted to a quarantine topic for later inspection. The result: a dashboard that displays only trustworthy environmental readings, while suspicious spikes—like a sudden temperature jump or vibration burst—are silently isolated.

What You'll Learn

• Configure and secure a Mosquitto MQTT broker on Raspberry Pi 5
• Program ESP32 boards to read DHT22 and ADXL345 sensors over I2C and MQTT
• Train an Isolation Forest anomaly detection model on synthetic and real sensor patterns
• Embed ML inference into an MQTT broker pipeline for inline payload filtering

Kit Contents

Why Buy This Kit Instead of Sourcing Parts Separately

Who This Kit Is For

CBSE Class 11–12 students exploring AI-integrated IoT for their practical records; B.Tech ECE/EEE undergraduates needing a ready-made, hackathon-grade project for Smart India Hackathon or institutional tech fests; and ATL Tinkering Labs or coding clubs at IIT, NIT, VIT, or BITS looking to demonstrate edge‑AI concepts. If you’ve ever wanted to show a panel that your sensor network doesn’t just collect data—it self-polices—this kit was made for you.

Built and Backed by Compoden

Every Compoden kit ships with an AI build companion trained on this exact project — accessible via a QR code on the box, with WhatsApp and email backup. We've spent 10 years building projects for makers, schools, and institutions across India. If a part fails because of a manufacturing defect, replace it free within 7 days.