What Is the Best Frequency for Long Range Communication?

The short answer

There is no single best frequency for long range communication. The right band depends on how far, over what terrain, with what antenna, under what noise floor, and under what licence. As a rough guide, with all of the usual caveats:

The rest of this post explains why each of those is true so you can make the call for your own link.

---

Lower frequency travels further. Up to a point.

Free space path loss increases with frequency. Doubling the frequency adds 6 dB of loss over the same distance, before you account for anything else. So in pure path loss terms, lower is better.

But that is only true when antenna gain is held constant. For a given physical aperture, antenna gain scales as roughly 1/λ², so higher frequencies recover that 6 dB (and often more) through increased directivity. This is why a 60 cm dish on a 6 GHz microwave link can comfortably outperform an isotropic comparison at 600 MHz, despite the higher path loss. Microwave links remain viable over long distances precisely because gain offsets propagation loss when aperture is fixed.

The “best” frequency is therefore the one where the combination of path loss, antenna gain, noise floor, ionospheric or tropospheric behaviour, and licensing gives you the most margin for the distance you need.

---

HF (3 to 30 MHz): the only band that goes truly long range without satellites

HF is the band that, under suitable ionospheric conditions and with efficient antennas, can let a 100 W radio in Perth talk to a 100 W radio in London. Not reliably, not on demand, but often enough to remain operationally useful. It does this in two ways:

• Skywave. Signals are gradually refracted through the plasma gradients of the ionosphere and returned to Earth hundreds to a few thousand kilometres away on each hop. It is not a mirror like bounce; the wave bends through a refractive index that varies with electron density. The usable frequency window is bounded by the Maximum Usable Frequency (MUF) and the Lowest Usable Frequency (LUF), both of which vary with time of day, season, and solar activity.
• NVIS (Near Vertical Incidence Skywave). A specialised use of skywave at low HF, typically 2 to 8 MHz, where signals are launched at high elevation angles and return within a radius of typically tens to a few hundred kilometres, depending on critical frequency, antenna pattern, and D layer absorption. NVIS is the workhorse for regional comms in difficult terrain.
• Ground wave. At the lower end of HF, signals follow the curvature of the Earth and diffract over terrain. Useful range is highly dependent on frequency and ground conductivity: tens of kilometres over average land, with longer reach over seawater or salt pan paths and at the very low end of HF, but rarely the “few hundred kilometres” sometimes claimed.

The catch with HF: antennas are large (a quarter wave at 7 MHz is roughly 10 m), bandwidth is narrow, and propagation conditions change hour by hour. HF rewards engineers who plan around MUF and LUF rather than picking a single fixed channel.

This is why defence, maritime, aviation, and remote area operators still rely on HF for beyond line of sight comms. No infrastructure, no satellite fees, global reach, when the ionosphere co-operates.

---

VHF (30 to 300 MHz): the practical sweet spot for regional line of sight

VHF is where most land mobile, marine, aviation, and emergency services live, and for good reason:

• Path loss is moderate
• Antennas are a manageable size (a quarter wave at 150 MHz is 0.5 m)
• The band penetrates light foliage and diffracts reasonably over modest terrain
• Noise floor is generally lower than HF in built up areas

VHF is essentially line of sight, with a useful slice of diffraction over hills and around obstacles. With a tower of reasonable height, VHF links of 30 to 80 km are routine. Beyond that you are relying on troposcatter, ducting, repeaters, or terrain that happens to co-operate. Troposcatter in particular is a real mechanism used in defence and commercial systems for reliable 100 to 300 km non line of sight links, at the cost of high transmit power and large antennas at both ends.

UHF is equally valid in much of the VHF range and is often preferred in practice where bandwidth and antenna size matter more than maximum range.

---

UHF (300 MHz to 3 GHz): shorter range, more bandwidth, smaller antennas

UHF includes most modern two way radio, LTE, and the lower end of microwave point to point. Compared to VHF:

• More path loss for the same distance
• More bandwidth, which means higher data rates
• Smaller antennas, easier to make high gain
• Bulk material penetration generally favours lower frequencies, but UHF often delivers better indoor coverage in practice through aperture coupling, reflections, and multipath exploitation

For long range, UHF is usually a worse choice than VHF unless you need the bandwidth or you are using high gain directional antennas on towers. With a 30 dBi dish at each end, a 5 GHz microwave link can comfortably cover 50 km. Without that gain, range collapses to a few kilometres under typical conditions.

---

Microwave (3 to 30 GHz and above): long range only with directional antennas

Above 3 GHz, free space loss is severe, but antenna gain scales with the square of frequency for a given aperture size. A 60 cm dish has roughly 8 dB more gain at 6 GHz than at 2.4 GHz. That is why long haul microwave backbones, satellite links, and high capacity point to point radios all live in this band.

The trade offs:

• Rain fade becomes significant above ~10 GHz
• Atmospheric absorption peaks at specific frequencies dominate link budgets in certain bands. Water vapour absorption peaks around 22 GHz and oxygen absorption peaks near 60 GHz. The 60 GHz peak is so steep that it is sometimes used deliberately for short range secure links
• Line of sight is mandatory. Almost no diffraction or penetration
• Fresnel zone clearance becomes critical (see our Path Loss and Fresnel Zones post)
• Regulatory EIRP limits often become the dominant constraint on achievable range in licensed microwave bands. The physics may permit more, but the licence does not
• Multipath fading is often the dominant impairment on long terrestrial paths, particularly below ~10 GHz, and is the reason ITU-R P.530 style fade margins exist

Microwave is not really a “long range frequency” in the sense people usually mean. It is a long range technique that uses high gain antennas to compensate for high path loss.

---

LF and VLF (below 300 kHz): the unusual case

Very low frequencies follow the Earth’s curvature, penetrate ground and seawater, and propagate globally with extraordinary stability. This is why they are used for submarine communications, time standard broadcasts, and some navigation systems.

The downsides are immense: data rates are extremely low (bits per second in some cases), antennas are kilometres in size, and transmit powers are measured in megawatts. Not relevant for most users, but worth knowing the band exists when the question is “what reaches everywhere?”

---

Antenna size: the constraint that quietly decides everything

Antenna gain is proportional to physical aperture divided by λ², so for a fixed gain the antenna gets larger as frequency drops, and for a fixed aperture the gain rises with frequency. A 10 dBi antenna at 7 MHz is enormous. The same gain at 2.4 GHz fits in a coffee cup. This is often the deciding factor for portable or vehicle mounted operation.

Engineers picking a frequency for long range often start by deciding what antenna they can physically deploy, then work backwards to the band. A handheld asking for HF skywave performance is not going to deliver. A truck with a 10 m whip might.

---

Noise and interference: the other quiet constraint

Lower frequencies have higher atmospheric and man made noise floors. ITU-R P.372 captures this with separate models for galactic, atmospheric, and man made noise across frequency. In practice, urban HF noise floors regularly exceed −90 dBm in narrow bandwidths (a few kHz, typical of voice or data channels) from switch mode power supplies, LED drivers, plasma displays, and Ethernet cabling. That alone can make HF unusable in the middle of a city at midday despite its “long range” reputation.

Higher frequencies are quieter in this sense, which sometimes means a 70 cm UHF link outperforms a 40 m HF link for short range comms in built up areas.

Always characterise the noise floor at your chosen site before committing to a band.

---

Licensing: often the real constraint

The frequency that would technically work best is sometimes simply not available to you. Apparatus licences, class licences, frequency coordination, and international band plans all narrow the choices. In Australia specifically, the ACMA RRL determines what is allocated and where, and a coordination check often eliminates the textbook best frequency before any engineering happens. Most other regulators apply broadly similar mechanisms under different names.

For a refresher on what is licensed and how, see our posts on apparatus and class licensing and the 2025 LIPD class licence.

---

A quick decision framework

1. How far? Under 50 km LOS, start at VHF or low UHF. Hundreds of km without infrastructure, look at HF (NVIS or skywave). Global, accept HF or satellite.
2. What terrain is in the way? Lower frequencies diffract better. Open paths give you more freedom.
3. What antenna can you actually deploy? This rules out more options than people expect.
4. What is the noise floor at the site? Characterise it across candidate bands. This rules out more again.
5. What is licensed and available? This usually rules out the rest.
6. Define required availability. A link engineered for 99.9% is not the same link engineered for 99.99%. The fade margin grows substantially as availability targets tighten, particularly above 10 GHz where rain statistics dominate.
7. Run the link budget and verify margin under worst case conditions. FSPL, antenna gain, environmental loss, fade margin sized to your availability target, with rain, multipath fading, noise, and interference all considered at their worst likely values rather than typical ones.

The unifying point underneath all of this: long range communication is not really a frequency problem. It is an SNR at the receiver problem. At its simplest (in dB form), received power is

Pr = Pt + Gt + Gr − Lpath − Lother

Receiver SNR is then determined by Pr minus the system noise floor (kTB + NF) and any in band interference, and link viability follows from that SNR relative to the demodulator threshold. Every band, every antenna, every terrain effect, every licence constraint just changes the values on the right hand side. The “best” frequency is the one that maximises received SNR after all propagation, antenna, and noise effects are accounted for, subject to whatever licensing, antenna, and availability constraints you are working within.

---

Tools that help

Typical workflow uses tools such as a Link Planner (VHF, UHF, and microwave path loss, Fresnel clearance, link budget), an HF Frequency Planner (MUF, LUF, ionospheric reliability, skywave coverage), a frequency database search (what is actually licensed at a site, e.g. the ACMA RRL in Australia), and an antenna sizing calculator (gain and aperture across bands). The noIM₃ platform provides each of these.

---

Key takeaway

There is no single best frequency for long range communication. HF reaches further than anything else without infrastructure but demands large antennas and patience with the ionosphere. VHF is the practical sweet spot for regional line of sight. UHF and microwave reach long distances only with high gain directional antennas. LF and VLF reach everywhere but carry almost no data.

The right answer for your link is the band where path loss, antenna gain, noise floor, availability target, and licensing all line up with the distance you need. Working through that comparison, anchored to received SNR rather than chasing a single magic number, is what makes a link close.