Selecting the right antenna is a critical step for getting the most out of your Meshtastic nodes. The antenna affects range, reliability and overall performance, so it’s important to make an informed choice.
This guide will help you understand the key factors and tools to optimize your setup.
Editors note: This is a general LoRa antenna selection guide. If you're choosing an antenna for Crow or Rook, check the recommended product section on the product page or browse our antennas - each antenna specifies frequency and compatibility.
The Basics You Can't Skip
Frequency Match
Every LoRa antenna is tuned to a specific frequency. If the antenna's resonant frequency doesn't match the frequency your node is transmitting on, two bad things happen: you lose range because the antenna is radiating inefficiently, and you risk sending energy back into the radio rather than out the antenna - which over time can damage the transmitter.
LoRa operates in regional ISM (Industrial, Scientific, and Medical) bands:
- 868 MHz - Europe (EU868)
- 915 MHz - US, Canada, Australia (US915, AU915)
- 433 MHz - some regional/legacy deployments
In Slovenia and the rest of the EU, you want an antenna specified for 868 MHz. This isn't a minor detail - an antenna designed for 915 MHz has a resonant length tuned to ~8.2 cm per quarter-wave element; the EU868 equivalent is ~8.6 cm. That gap matters more when the antenna is already marginal, which many budget antennas are.
Connector Type
Check that the antenna connector matches your node before ordering anything. The two you'll encounter most in LoRa hardware:
SMA - the most common connector on LoRa modules and nodes. SMA male has a centre pin and external threads; SMA female has a socket and internal threads.
RP-SMA (Reverse Polarity SMA) - identical shell to SMA, but the centre contact is flipped. RP-SMA male has a socket where standard SMA male has a pin. Originally mandated for consumer WiFi equipment, RP-SMA antennas are widely available and cheap - and will not mate with a standard SMA connector even though they look identical. This is one of the most common, easiest-to-miss mistakes when buying antennas.
N-type - a larger, more robust connector used on base station antennas and fixed outdoor installations. Weatherproof when properly torqued, rated to higher power, and common on fiberglass pole antennas.
Always verify the connector on your specific node. Different manufacturers and even different product revisions sometimes change connector types. When in doubt, the product documentation or a quick photo check before ordering avoids a frustrating return.
Antenna Gain - What It Actually Means
Antenna gain is not like amplifier gain. An antenna doesn't add energy to the signal - it redirects it. A higher-gain antenna focuses the radiation pattern toward the horizon, squeezing more signal in the directions you care about (horizontal, toward other nodes) at the cost of coverage above and below.
Gain is measured in dBi (decibels relative to an isotropic radiator - a theoretical point source that radiates equally in all directions). A few reference points:
- ~2 dBi - typical quarter-wave whip or stubby dipole. Broad, roughly spherical pattern. Good for portable nodes and situations where the node might be at varying heights or angles relative to the receiver.
- 3–5 dBi - medium-gain whip or short fiberglass antenna. Pattern starts compressing toward the horizon. Good balance for fixed portable and semi-permanent nodes.
- 5–8 dBi - longer fiberglass or dedicated base station antennas. Pattern is noticeably flattened. Excellent for stationary nodes in flat or open terrain; can actually hurt performance if the receiving node is significantly above or below the transmitting node.
A practical note: cheap high-gain claims are frequently false. Real-world independent testing of 868 MHz antennas - including measurements by the coredump.ch lab - consistently found that budget "9 dBi" or "high gain" antennas from anonymous online sellers performed anywhere from 2 dBi down to deeply negative values. A well-specified, name-brand 2–3 dBi antenna will generally outperform a no-name "5 dBi" antenna from the same supplier. Buy from reputable sources and verify specs where possible.
Antenna Types for Different Deployments
Stub / Whip Antennas
The standard antenna that ships with most LoRa devices. Typically 2–3 dBi, omnidirectional, SMA-connected. The quarter-wave rubber duck antenna in most of these is short (~8–9 cm at 868 MHz), broadband-ish, and tolerant of being used in different orientations.
Best for: handheld and portable nodes, wearable setups, situations where orientation relative to other nodes varies. The forgiving radiation pattern means you don't have to think much about alignment.
Limitation: the antennas that ship with many LoRa devices are often bottom-of-range quality, frequently not tuned precisely for 868 MHz or 915 MHz. Replacing the stock stub with a quality equivalent often provides a noticeable improvement at minimal cost.
Gooseneck / Articulating Antennas
A flexible-neck version of the stub, allowing the radiating element to be angled. Useful when a node is mounted in a fixed position (backpack, vehicle interior, enclosure) but the ideal antenna orientation differs from the mounting angle.
Best for: backpack-mounted nodes, vehicle installs where the connector faces downward or sideways, any deployment where the node position is constrained.
Magnetic Mount Antennas
An external antenna on a magnetic base, attached via coaxial cable to the node. The magnetic base adheres to any ferrous metal surface. Ground plane is provided by the metal surface - which is also why magnetic mount antennas typically perform well on vehicle rooftops, where the roof provides an excellent ground plane.
Best for: vehicle-mounted nodes (attach to roof or bonnet), temporary fixed installations on metal structures, any deployment where you want an elevated antenna separate from the node itself.
Note on cable loss: the coaxial cable between the node and the magnetic mount antenna introduces signal loss - typically 0.5–1.5 dB for short quality cable runs. This is acceptable for most deployments and the gain from antenna height and better ground plane usually more than compensates.
Fiberglass Pole Antennas
Long, weatherproof antennas mounted on masts or poles. Available in gains ranging from ~3 dBi up to 8+ dBi, with physical length increasing with gain. Usually use N-type connectors and are designed for permanent or semi-permanent outdoor mounting.
Best for: fixed relay nodes at high points (rooftop, hilltop, tower), base stations for a mesh network operating over a wide area, any application where the node doesn't move and line-of-sight range matters.
Important caveat: as gain increases beyond ~5–6 dBi, the radiation pattern becomes increasingly narrow in the vertical plane. At 8 dBi, the pattern is flat enough that a node only slightly above or below the antenna can fall outside the main beam. This is the counterintuitive part of high-gain omnidirectional antennas - they can extend range horizontally while creating blind spots vertically. For dense mesh networks in varied terrain, moderate gain (3–5 dBi) often works better than maximum gain.
Planning Your Coverage Before You Deploy
Getting antenna placement and type right is much easier when you can simulate it against real terrain data before going into the field. The Meshtastic Site Planner (available at site.meshtastic.org) is a free, open-source tool that does exactly this.
It's built on the ITM/Longley-Rice model - the same propagation modelling approach used in professional cell tower planning - running against NASA SRTM terrain elevation data with 90-metre resolution. You enter a location, antenna height, transmit power, frequency, and antenna gain, and it generates a colour-coded RSSI map showing predicted coverage across the surrounding terrain.
Practically, this means you can:
- Compare node placement candidates: Is the hill to the north a better relay site than the ridge to the east? Simulate both and see the coverage difference.
- Estimate the value of raising antenna height: How much does going from 2m to 5m above ground change the coverage area? The planner shows you quantitatively.
- Identify gaps in a planned mesh: If three nodes cover an area, are there terrain shadows where coverage drops out? Where should a fourth relay go to fill them?
- Set realistic expectations before ordering hardware: A site planner result of strong RSSI (above -110 dBm) across your intended operating area is a solid basis for deployment. Predicted RSSI below -125 dBm in key areas suggests you need higher nodes, higher gain, or more of them.
Key limitations to understand: the model accounts for terrain elevation but not for trees, buildings, or other above-ground obstructions. In heavily forested or urban environments, predicted coverage will be optimistic compared to real-world results. Always validate with field testing after planning - the site planner narrows down your options and sets realistic targets; it doesn't replace real-world verification.
To use it: go to site.meshtastic.org, click your node location on the map, enter your antenna height above ground, set the frequency to your region (868 MHz for EU), input your transmit power and antenna gain, and run the simulation. The receiver sensitivity default (-130 dBm for LongFast) is appropriate for most Meshtastic setups.
Deployment Specifics: Getting the Most From Your Setup
Keep antennas vertical. LoRa antennas are vertically polarised. A receiving antenna at the same polarisation as the transmitting antenna maximises signal strength; mismatched polarisation (one vertical, one horizontal) introduces up to 20 dB of polarisation loss - effectively reducing your link to a fraction of what it should be. For wearable nodes, this means the antenna should ideally point upward from a chest rig or backpack shoulder strap, not lie flat.
Height matters more than gain, up to a point. Elevation above obstructions is the single most effective way to extend coverage. A 2 dBi antenna at 10m elevation will typically outperform a 6 dBi antenna at 1m elevation in any terrain with ground-level obstructions. When placing relay nodes, prioritise getting them high before worrying about antenna selection.
Watch cable length and quality. Every metre of coaxial cable between a node and an elevated antenna introduces signal loss. At 868 MHz, common RG58 cable loses around 4-5 dB per 10 metres - enough to completely negate the benefit of antenna height if the cable run is long. For anything over 1–2m of cable, use lower-loss coax (LMR-200 or better). Keep connector count low; each connector adds ~0.2 dB.
Test with real hardware before committing. Simulate first, then deploy two nodes in your intended area and walk the coverage boundary. Record RSSI values at the edges. Compare to simulation predictions. If real coverage is significantly worse than predicted, suspect foliage density, building attenuation, or multipath effects that the model didn't account for — and adjust node placement accordingly.
Common Mistakes Worth Avoiding
Using a 915 MHz antenna on an 868 MHz node - or vice versa. Many budget antennas are advertised as "868/915 MHz compatible". This usually means they work at both, with degraded performance at both. If you're deploying in Europe, use an antenna specified and measured for 868 MHz.
Trusting gain specs on no-name antennas. As independent antenna measurements have consistently shown, claimed gain and measured gain can differ dramatically on cheap antennas. An antenna from an unknown seller claiming 9 dBi that actually delivers -7 dBi is actively worse than a quality 2 dBi antenna. Stick to antennas from reputable suppliers with actual test data.
Ignoring connector type. SMA and RP-SMA will not mate. Verify before ordering.
High gain for a mobile node in varied terrain. A 6 dBi antenna on a node being carried through a forest with significant elevation changes will miss nodes that a 2–3 dBi antenna would have reached, because the compressed vertical radiation pattern misses targets above and below. High gain belongs on stationary, elevated relay nodes - not carried hardware.
Skipping field validation. Simulation and spec sheets give you a starting point, not a guarantee. Real terrain, vegetation, and interference can differ significantly from model predictions. Deploy, test, and adjust.
Summary
Your antenna is the interface between your node and the physics of radio propagation. Every other decision you make about a Meshtastic network assumes the antenna is doing its job. Getting it right means: matching frequency (868 MHz in EU, 915 MHz in USA/CA, etc.), verifying connector type, choosing gain appropriate to the deployment context, and validating with the Meshtastic Site Planner before committing to placement.