Toll-Gate Monitoring & Traffic Analytics

A smart toll-gate prototype integrating IoT, Cloud Computing, Big Data, and a Next.js frontend to monitor vehicle access and analyze traffic in real time.

System architecture:

• Edge layer: ESP32 microcontrollers connected to RFID sensors for vehicle identification, ultrasonic sensors for distance/queue detection, and servo motors for gate control
• Transport: MQTT messaging over HiveMQ Cloud (TLS-secured) carries sensor events from edge nodes to the cloud
• Cloud backend: A Python subscriber running as a systemd service on AWS EC2 handles card validation, balance/tariff business logic, and persists data to DynamoDB and S3
• Analytics: A 4-node Hadoop MapReduce pipeline processes access logs for traffic volume, revenue, and peak-hour analysis
• Dashboard: Next.js web app providing real-time gate status, live RFID transactions, traffic volume charts, weekly heatmaps, queue-time estimates, and CRUD management for registered RFID cards

Key features:

• Dual-gate monitoring (entry & exit) with live status indicators
• Remote gate override from the dashboard
• Traffic congestion indicator (green/yellow/red)
• Access success vs. failure analytics
• System latency performance metrics
• CSV/PDF log export

My scope: ESP32 firmware (RFID, ultrasonic, servo), MQTT topic design over HiveMQ (TLS), Python subscriber with balance/tariff business logic on EC2, DynamoDB + S3 storage, and the 4-node Hadoop MapReduce pipeline (traffic, access, revenue). Also authored the IEEE-830 SKPL and an active-active HA design.

Tech stack: ESP32, MQTT, HiveMQ Cloud, Python, AWS (EC2, DynamoDB, S3), Hadoop MapReduce, Next.js, TypeScript, C++.