Smart farming with IoT is revolutionising Indian agriculture by enabling data-driven decisions that boost yield and reduce waste. From soil moisture tracking in Rajasthan’s arid fields to humidity monitoring in Kerala’s spice plantations, IoT sensor networks give farmers real-time visibility into conditions that once relied on guesswork. In this comprehensive guide, we walk you through building a complete smart farming IoT system tailored for Indian agricultural conditions.
Table of Contents
- Why Smart Farming Matters for India
- System Architecture Overview
- Choosing the Right Sensors
- ESP32 and LoRa Gateway Setup
- Soil Moisture and Nutrient Monitoring
- Field Weather Station Node
- Cloud Dashboard and Alerts
- Frequently Asked Questions
- Conclusion
Why Smart Farming Matters for India
India’s agricultural sector supports over 58% of rural households, yet average productivity per hectare lags behind global benchmarks. Erratic monsoon patterns, groundwater depletion, and rising input costs make traditional farming increasingly risky. IoT-based smart farming addresses these challenges by:
- Reducing water usage by 30-50% through precision irrigation triggered by actual soil moisture readings rather than fixed schedules
- Early pest and disease detection by monitoring microclimate conditions that favour fungal growth or pest breeding
- Optimising fertiliser application with soil nutrient data, saving costs and preventing groundwater contamination
- Enabling remote monitoring for farmers who manage multiple plots across different villages
Government schemes like the Digital Agriculture Mission 2025-2030 are actively funding precision agriculture adoption. For progressive farmers and agri-tech startups, building a DIY IoT sensor network is both affordable and practical with components available from Indian suppliers.
System Architecture Overview
A smart farming IoT system typically consists of three layers:
- Sensor Nodes: Battery or solar-powered units placed across the field, each measuring soil moisture, temperature, humidity, and light levels. These communicate wirelessly via LoRa (long range) to a central gateway.
- Gateway: An ESP32-based hub with LoRa receiver and WiFi/GSM connectivity. It collects data from all sensor nodes and uploads to the cloud every 15-30 minutes.
- Cloud Platform: A dashboard (ThingSpeak, Blynk, or custom) where data is stored, visualised, and triggers SMS/WhatsApp alerts when thresholds are breached.
For a typical 2-acre Indian farm, you need 4-6 sensor nodes placed at different zones, one LoRa gateway with GSM backup, and a solar panel to keep the gateway running in areas without reliable grid power.
Choosing the Right Sensors
Selecting sensors for Indian agricultural conditions requires careful consideration of temperature extremes (summers exceed 45°C in many states), monsoon humidity, and dust exposure.
Soil Moisture Sensors
The capacitive soil moisture sensor is preferred over resistive types because it does not corrode in wet Indian soil. Resistive sensors degrade within weeks during kharif season, while capacitive sensors last multiple seasons. Connect the analogue output to an ESP32 ADC pin and calibrate against known dry (air) and wet (submerged) readings.
Temperature and Humidity
The DHT22 offers a good balance of accuracy and cost for outdoor deployment. For higher precision, the BME280 provides temperature, humidity, and barometric pressure in a single I2C module.
Light Intensity
A BH1750 digital light sensor measures lux levels for tracking photosynthetically active radiation (PAR). This is valuable for greenhouse and polyhouse operations where light management affects crop quality.
Rain Detection
A simple rain sensor module detects rainfall onset, triggering irrigation shutoff to avoid overwatering. Combined with a tipping bucket rain gauge, you can also measure cumulative rainfall.
ESP32 and LoRa Gateway Setup
The ESP32 is the ideal microcontroller for smart farming because it has built-in WiFi, Bluetooth, deep sleep capability (as low as 10µA), and sufficient ADC channels for multiple sensors.
Wiring the Sensor Node
// Sensor Node Wiring (ESP32)
// DHT22: VCC→3.3V, GND→GND, DATA→GPIO4 (10kΩ pull-up)
// Capacitive Soil Moisture: VCC→3.3V, GND→GND, AOUT→GPIO34
// BH1750: VCC→3.3V, GND→GND, SDA→GPIO21, SCL→GPIO22
// LoRa SX1276: VCC→3.3V, GND→GND, MOSI→GPIO23, MISO→GPIO19,
// SCK→GPIO18, NSS→GPIO5, RST→GPIO14, DIO0→GPIO2
Arduino Code for Sensor Node
#include <DHT.h>
#include <LoRa.h>
#include <Wire.h>
#include <BH1750.h>
#define DHT_PIN 4
#define SOIL_PIN 34
#define SLEEP_TIME 900 // 15 minutes in seconds
DHT dht(DHT_PIN, DHT22);
BH1750 lightMeter;
void setup() {
Serial.begin(115200);
dht.begin();
Wire.begin();
lightMeter.begin();
LoRa.begin(433E6); // 433 MHz for India
LoRa.setSpreadingFactor(10);
}
void loop() {
float temp = dht.readTemperature();
float hum = dht.readHumidity();
int soilRaw = analogRead(SOIL_PIN);
float soilPercent = map(soilRaw, 4095, 1500, 0, 100);
float lux = lightMeter.readLightLevel();
String payload = String(temp) + "," + String(hum) + ","
+ String(soilPercent) + "," + String(lux);
LoRa.beginPacket();
LoRa.print(payload);
LoRa.endPacket();
esp_deep_sleep(SLEEP_TIME * 1000000ULL);
}
Soil Moisture and Nutrient Monitoring
Soil moisture is the single most impactful metric for Indian agriculture. During rabi season, wheat fields in Punjab and Haryana need consistent 50-60% soil moisture, while during kharif season, paddy fields in West Bengal require near-saturation levels.
Place capacitive soil moisture sensors at two depths: 10 cm (root zone for shallow crops like wheat) and 30 cm (for deep-rooted crops like sugarcane and cotton). The ESP32 reads both sensors and transmits the gradient, which indicates whether water is percolating properly or pooling at the surface.
For soil nutrient monitoring (NPK), the RS485 soil NPK sensor provides nitrogen, phosphorus, and potassium readings. Connect it to the ESP32 via a MAX485 transceiver module. While not as accurate as laboratory testing, trend monitoring over weeks helps optimise fertiliser application timing.
Field Weather Station Node
A dedicated weather node at the farm entrance provides hyperlocal weather data far more relevant than IMD forecasts for distant city stations. Include:
- BME280: Temperature, humidity, barometric pressure (pressure drops indicate approaching rain)
- Anemometer: Wind speed for spray application decisions (avoid spraying above 15 km/h)
- Rain gauge: Tipping bucket with reed switch, each tip equals 0.2mm rainfall
- UV sensor: VEML6075 for UV index monitoring, relevant for high-altitude farms in Himachal Pradesh and Uttarakhand
Cloud Dashboard and Alerts
ThingSpeak (free for up to 3 million messages/year) is an excellent starting point. Configure channels for each sensor node with fields for temperature, humidity, soil moisture, and light. Set up MATLAB Analysis in ThingSpeak to calculate evapotranspiration (ET) using the Penman-Monteith formula, which tells you exactly how much water your crop is losing daily.
For alerts, use ThingReact or a simple webhook to send SMS via Twilio or WhatsApp messages via the WhatsApp Business API. Critical alerts include:
- Soil moisture below 30% (irrigation needed)
- Temperature above 42°C (heat stress, activate foggers)
- Humidity above 85% with temperature 20-25°C (fungal disease risk)
- No data received for 2 hours (node battery or connectivity failure)
The total cost for a 2-acre smart farming setup with 4 sensor nodes, 1 gateway, and solar panels comes to approximately ₹8,000-12,000 — a fraction of the potential savings from optimised irrigation and reduced crop losses.
Frequently Asked Questions
What is the range of LoRa for farm sensor networks?
In open agricultural fields (line of sight), LoRa at 433 MHz with spreading factor 10 achieves 3-5 km range easily. In hilly terrain like the Western Ghats, expect 1-2 km. For larger farms, use LoRa repeater nodes.
How long do batteries last on sensor nodes?
With ESP32 deep sleep (15-minute wake intervals), two 18650 lithium cells (3400mAh each) last 4-6 months. Adding a small 1W solar panel makes the node self-sustaining indefinitely.
Can this system work without internet in remote areas?
Yes. The LoRa sensor-to-gateway communication is independent of internet. The gateway can store data locally on an SD card and upload in batches when GSM connectivity is available. You can also set up local alerts using a buzzer or LED at the gateway.
Is this system suitable for all crops?
The sensor types remain the same, but threshold values differ. Cotton requires different soil moisture levels than paddy. Customise your alert thresholds based on crop-specific guidelines from ICAR (Indian Council of Agricultural Research).
What is the total cost of a smart farming IoT setup?
A basic 4-node system for 2 acres costs ₹8,000-12,000 including sensors, microcontrollers, LoRa modules, enclosures, and a solar panel. This is a one-time investment that pays for itself within one season through water and fertiliser savings.
Conclusion
Building a smart farming IoT system is one of the most impactful projects for Indian agriculture. With affordable sensors, ESP32 microcontrollers, and LoRa connectivity, even small and marginal farmers can access the kind of data-driven insights that were once limited to large agribusiness operations. Start with a single soil moisture node, validate the benefits in one season, and expand from there. Browse the complete range of agricultural sensors and IoT modules at Zbotic to get started on your smart farming journey.
Add comment