IoT Projects in Bangalore for ECE

An ESP32 node reads soil moisture (capacitive sensor), ambient temperature and humidity (DHT22), and light intensity (BH1750) every 5 minutes and publishes JSON payloads to an AWS IoT Core MQTT topic. A Lambda function processes incoming data, stores it in DynamoDB, and triggers an SNS alert when soil moisture drops below a configurable threshold. A relay-controlled solenoid valve opens automatically for a preset duration. A React dashboard hosted on S3 visualises 7-day sensor trends and allows remote valve control. The system reduces irrigation water usage by an estimated 35% compared to timer-based methods.

An ESP32 with an MPU6050 accelerometer and DS18B20 temperature sensor is mounted on a motor housing to capture vibration FFT spectra and bearing temperature at 1-minute intervals. Data is published over MQTT to a Mosquitto broker and forwarded to InfluxDB. A Grafana dashboard displays real-time and historical trends with threshold-based alerting via Telegram bot. A simple anomaly detection rule flags vibration amplitudes exceeding 2× the 30-day rolling mean. The project demonstrates industrial IoT architecture, time-series database integration, and predictive maintenance concepts.

An ESP32 controls four home appliances (lights, fan, AC, geyser) via 5V relay modules. Device state is synchronised with a Firebase Realtime Database, enabling control from a Flutter mobile app on Android and iOS. Google Assistant integration is achieved through IFTTT webhooks, allowing voice commands such as "turn off bedroom light." The firmware implements Wi-Fi reconnection with exponential back-off and stores the last known device state in NVS flash so appliances restore their state after a power cut. The mobile app shows real-time power consumption estimated from appliance ratings.

Five sensor nodes, each built around an Arduino Pro Mini with an MQ-135 CO₂/VOC sensor, MQ-7 CO sensor, and a PMS5003 particulate matter sensor, transmit readings every 10 minutes over LoRaWAN (RN2483 module) to a single-channel gateway. The gateway forwards packets to The Things Network, which routes data to ThingSpeak via MQTT. A MATLAB analysis script on ThingSpeak computes AQI values and generates hourly heatmaps. The project demonstrates multi-node LoRaWAN deployment, AQI calculation per CPCB norms, and cloud-based environmental analytics.

An ESP32 reads SpO₂ and heart rate from a MAX30102 pulse oximeter module, body temperature from an MLX90614 IR sensor, and ECG waveform from an AD8232 module. Readings are displayed on a 2.4-inch TFT and simultaneously published to a Node-RED dashboard over MQTT. If SpO₂ drops below 94% or heart rate exceeds 120 bpm, a Telegram alert is sent to the nurse's smartphone within 2 seconds. The project demonstrates multi-sensor integration, real-time threshold alerting, and medical IoT data visualisation.

An HC-SR04 ultrasonic sensor mounted inside a waste bin measures fill level and an ESP8266 transmits the percentage to a Firebase database every 30 minutes. A web dashboard displays a city map with bin locations colour-coded by fill level (green/yellow/red). A Python script running on a Raspberry Pi applies a nearest-neighbour heuristic to generate an optimised collection route for bins above 80% capacity and exports it as a Google Maps link. The project demonstrates IoT-enabled smart city infrastructure and basic route optimisation.

A DS18B20 waterproof temperature sensor connected to an Arduino Mega logs readings every 60 seconds to a microSD card with a DS3231 RTC timestamp. If temperature exceeds ±2°C of the setpoint (configurable via keypad), a SIM800L GSM module sends an SMS alert to the logistics manager. A 20×4 LCD displays current temperature, setpoint, and alert status. The logged CSV file can be downloaded at delivery and imported into Excel for compliance reporting. The project covers cold chain monitoring requirements per WHO PQS standards.

An ESP32 measures river water level using an HC-SR04 ultrasonic sensor mounted on a bridge pillar and transmits readings every 5 minutes to ThingSpeak. When the level crosses a warning threshold, a buzzer and red LED activate locally and a Telegram bot sends an alert to registered subscribers. Historical level data is visualised on a ThingSpeak chart with a 24-hour trend line. The system is solar-powered using a 10W panel and 7Ah LiPo battery with an TP4056 charge controller, demonstrating off-grid IoT deployment.

A PZEM-004T energy monitoring module measures voltage, current, power, frequency, and power factor of a household circuit and transmits readings over UART to an ESP32. The ESP32 publishes data to an MQTT broker every 10 seconds. A Node-RED dashboard displays real-time gauges and 24-hour consumption charts. Monthly energy cost is calculated using the BESCOM tariff slab structure and displayed as a running total. The project demonstrates non-invasive energy monitoring, MQTT integration, and dashboard design for residential IoT applications.