LoRa at 10km in the Datasheet, 400m in the Field: The Real Antenna Engineering Story

Author: Srihari Maddula  •  Founder & Technical Lead, Eurth Techtronics Pvt Ltd

Category: Connectivity Reliability  •  Estimated Reading Time: 18–20 minutes

Published: July 2026

The Number on the Box Is Not a Deployment Specification

Every LoRa module datasheet has a range figure. 10 km. Sometimes 15 km. The number is real — under the right conditions, LoRa can achieve that range. Those conditions are: line of sight between antennas, antennas at height, correctly matched 50-ohm antenna systems at both ends, a spreading factor of SF12 (at the cost of very low data rate), and no significant interference on the 865–867 MHz (India) band.

Real deployments rarely have line of sight. Antennas are often at 1–2 metres elevation because that is where the mounting point is. The antenna matching is often 'good enough' rather than characterised. Interference from other LoRa networks, agricultural equipment, and unlicensed band users exists. And SF12 at 10 km means a data rate of approximately 250 bps and a time-on-air of 2.8 seconds per 20-byte packet.

The result is that 10 km on the datasheet becomes 400 metres to 1 km in a typical agricultural or industrial deployment. Not because LoRa technology is misrepresented — it is not. But because the datasheet range is achieved under ideal conditions that field deployments do not reproduce. Understanding the link budget is the engineering tool that bridges the gap between the datasheet and the deployment.

Link Budget: The Full Accounting

A link budget is an accounting of all the signal gains and losses between a transmitter and a receiver. If the total gains minus total losses is positive by at least your required fade margin, the link works. If it is not, it does not — regardless of what the datasheet says.

Starting point: transmit power. For a Semtech SX1276-based module operating in India's 865 MHz ISM band, the maximum legal EIRP is typically +30 dBm (1 watt). Many LoRa modules transmit at +14 to +20 dBm to preserve battery life and respect duty cycle limits.

Add: transmit antenna gain. A monopole on a PCB trace has approximately 0 to 2 dBi. A tuned dipole on a pole has 2–3 dBi. A directional Yagi has 6–12 dBi. Antenna gain directly adds to your effective radiated power and to your link budget.

Subtract: cable and connector loss. A 1-metre RG-58 cable at 868 MHz loses approximately 0.5 dB. Connectors add 0.1–0.3 dB each. A 3-metre cable run with two connectors could cost you 2 dB — which is a significant fraction of your budget.

Subtract: free space path loss. At 868 MHz, free space path loss at 1 km is approximately 92 dB. At 5 km, it is approximately 106 dB. This is the dominant loss term and it is unavoidable — it is the physics of propagating electromagnetic waves through space.

Subtract: additional propagation losses. This is where the field diverges from the datasheet. Vegetation loss — a 3-metre sugarcane field absorbs 3–8 dB per 100 metres of path through the crop. Building penetration loss — 10–20 dB through a concrete wall. Ground reflection and multipath — variable, typically adding 0–10 dB of loss in non-line-of-sight conditions. Fresnel zone obstruction — if the first Fresnel zone is more than 40% blocked by terrain, add 6 dB or more of loss.

Add: receive antenna gain. Same calculation as transmit side — gain adds to the budget.

Subtract: receiver noise floor. At SF12, BW 125 kHz, the LoRa receiver sensitivity is approximately -137 dBm. This is the minimum receivable signal power.

The link margin is: TX power + TX antenna gain - cable loss - path loss - vegetation loss - other losses + RX antenna gain - receiver sensitivity. If this number is greater than your required fade margin (typically 10–15 dB for reliable links), the link works. If it is less, you either shorten the range, increase antenna gain, increase TX power (within legal limits), choose a higher spreading factor (lower SF = shorter range), or redesign the deployment topology.

The Fresnel Zone: What Nobody Draws on the Site Map

The Fresnel zone is an ellipsoidal region around the line-of-sight path between two antennas. RF energy does not travel in a laser-like ray — it propagates through a volume of space, and obstacles within that volume affect signal strength even if they do not directly block the line of sight.

The radius of the first Fresnel zone at the midpoint of a 1 km link at 868 MHz is approximately 9.3 metres. This means that any obstacle within 9.3 metres of the midpoint line-of-sight path — a tree, a building, the crest of a hill — will partially block the Fresnel zone and degrade the link, even if the visual line of sight is clear. Clearing 60% of the first Fresnel zone is the minimum requirement for a reliable link.

This is why antenna height matters so much. Mounting a gateway antenna at 8 metres instead of 2 metres on flat terrain moves the first Fresnel zone clearance point above most vegetation and ground-level obstacles. The 6-metre difference in height may add 10–15 dB to the effective link budget in terrain with vegetation — turning a marginal link into a reliable one, or extending a reliable link by 3–4 km.

Antenna Matching: The 3 dB You Are Leaving on the Table

An antenna that is not well-matched to the 50-ohm transmission line reflects some of the transmit power back toward the radio rather than radiating it. The reflected power is characterised by the antenna's return loss or VSWR. A VSWR of 2:1 means approximately 11% of transmit power is reflected — a loss of 0.5 dB. A VSWR of 3:1 means 25% reflected — a loss of 1.25 dB. A poorly matched antenna with VSWR 5:1 wastes 44% of transmit power — a loss of 2.5 dB.

PCB trace antennas and external antennas with SMA connectors can both have poor matching when the antenna is used at a frequency or in a geometry different from its design conditions — for example, a 915 MHz antenna used at 868 MHz, or an antenna mounted on a metallic enclosure that shifts its resonant frequency.

Characterise your antenna's return loss with a vector network analyser (VNA) at the actual operating frequency and in the actual mounting configuration before you deploy. A low-cost NanoVNA is sufficient for this characterisation. The 2–3 dB you recover through proper antenna matching may be the difference between a reliable link and a marginal one.

Specifying Connectivity Before You Ship Hardware

The correct sequence for a LoRa deployment is: site survey first, link budget calculation second, hardware selection and antenna choice third, and deployment fourth. Not the reverse.

During the site survey, measure the GPS coordinates of each proposed node position and the gateway. Calculate the terrain profile and Fresnel zone obstruction using free tools like Radio Mobile or HeyWhatsThat. Estimate vegetation and building losses based on the actual site. Identify the required antenna height for adequate Fresnel zone clearance. Calculate the link budget for the worst-case node position. Select the spreading factor that provides the required throughput with adequate link margin.

This takes one to two days for a 20-node deployment. It is the difference between a deployment where everything works from day one and a deployment where you spend months discovering why nodes at certain positions never connect reliably.

THE RULE: Never commit hardware to a LoRa deployment without a site survey and a link budget calculation.

The datasheet range is a physics upper bound under ideal conditions. Your deployment conditions are not ideal.

© 2026 Eurth Techtronics Pvt Ltd  |  eurthtech.com  |  All rights reserved.