Project 03 — IoT / Embedded Systems

DorperNet

An end-to-end IoT smart livestock monitoring system built for D'Impian Agro Farm — automating weighing, RFID animal tagging, LoRa wireless communication, and cloud-based real-time data management across a working farm environment.

RoleIoT System Developer
DurationJul – Oct 2023
ClientD'Impian Agro Farm
ESP32LoRaRFID Load CellsIoTCloud Google Sheets APIPython

System Overview

What DorperNet Does

A fully integrated farm automation system — from hardware sensors on the animal to data visualized in the cloud. Three subsystems working together for real-time livestock management.

01

RFID Animal Tagging

Each Dorper sheep is tagged with a unique RFID identifier. The reader logs individual animal identity at the weighing station automatically, enabling per-animal tracking over time.

02

Automated Weighing

Load cell sensors connected to microcontrollers capture weight readings. Data is processed and averaged for accuracy, then paired with the RFID tag for each measurement session.

03

LoRa + Cloud Sync

Node-to-Node LoRa transmission sends data wirelessly across the farm to a LoRa gateway. A Python backend uploads to Google Sheets for centralized, real-time livestock records accessible anywhere.

Hardware Stack

Components & Setup

The system runs on off-the-shelf embedded hardware deployed in a real farm environment — designed for durability, low power, and long-range communication.

Sensing Node

  • MCUESP32 — main processing unit, WiFi + BT capable
  • Load CellHX711 amplifier + load cell for weight measurement
  • RFIDRFID reader module for animal tag identification
  • LoRa RadioSX1276 / compatible LoRa module — long-range TX
  • PowerField-ready regulated supply for the sensing node and peripheral modules

Gateway & Backend

  • GatewayLoRa gateway node receiving from field devices
  • BackendPython script parsing and forwarding data
  • CloudGoogle Sheets via API for centralized data storage
  • MonitoringReal-time dashboard accessible remotely
  • RangeValidated for long-range farm communication between the weighing node and gateway

Documentation

Photos & Field Deployment

Field photos from the DorperNet deployment, hardware assembly, and on-site livestock monitoring workflow.

Deployment Process

How It Was Built & Deployed

01

Hardware Assembly

ESP32, RFID reader, HX711 + load cell, and LoRa radio wired and encased for outdoor farm use. PCB or breadboard prototyping stage.

02

Firmware Development

Embedded code in C++ / Arduino framework handling RFID reads, load cell ADC sampling, averaging, and LoRa packet transmission.

03

Gateway + Cloud

Python script on gateway device receives LoRa packets, parses data, and uploads livestock records to Google Sheets via the Sheets API in real time.

04

Onsite Validation

Calibrated load cells on-site with known weights. Validated RFID read rates, LoRa range across the farm, and end-to-end data integrity in production conditions.

Outcomes

What Was Achieved

4mo Internship — Jul to Oct 2023

Deployed, calibrated, and validated the full system in a real working farm environment during the internship period at D'Impian Agro Farm.

Key Deliverables

Automated livestock weighing replacing manual recording
RFID-based individual animal identification
Long-range LoRa wireless link across the farm site
Cloud-synchronized real-time Google Sheets records
Onsite calibration and production deployment
Delivered a working deployment spanning RFID capture, weight logging, LoRa transfer, and cloud reporting

Additional Information

Notes & Context

Additional context from the real farm deployment, including field constraints, technical decisions, and lessons from operating the system onsite.

Challenges

The deployment had to account for farm interference, outdoor protection for electronics, calibration stability, and reliable wireless links across the site.

Lessons Learned

The project reinforced the value of robust field calibration, simple telemetry pipelines, and designing embedded systems around real operating conditions instead of lab assumptions.

Jejak Inovasi UTeM 2023 — Best Project & Gold Award
← Previous Wound Image Classification All Projects Next → Wound RF Classification