mmWave (millimeter wave) is the portion of the radio spectrum with wavelengths between 1 and 10 millimeters, roughly 30 to 300 GHz, and in 5G networks the term covers the Frequency Range 2 (FR2) bands from about 24.25 GHz to 71 GHz. It gives mobile operators very wide channels of 100 to 400 MHz, which is what makes multi-gigabit peak speeds and very high cell capacity possible.
The price of that bandwidth is short range and fragile propagation. Millimeter-wave signals lose more energy over distance, are blocked by walls, bodies and foliage, and depend on beamforming with large antenna arrays to work at all. In telecommunications practice, mmWave is therefore a capacity layer for dense hotspots, not a coverage layer for whole regions.
How mmWave radio works in 5G NR
mmWave radio in 5G NR relies on wide bandwidth, short wavelengths and highly directional beams. Because the wavelength is only a few millimeters, a gNodeB can pack hundreds of antenna elements into a panel the size of a book, and a phone can carry several small arrays at its edges. These arrays steer narrow beams toward each user, which compensates for the high path loss through antenna gain.
The 5G NR standard adapts its numerology to these frequencies. FR2 uses subcarrier spacing of 120 kHz for data (with 240 kHz for some synchronization signals in earlier releases), which shortens symbols and slots to 0.125 ms. Shorter slots reduce latency and make the system more tolerant of phase noise, a significant impairment of oscillators at high frequencies. Almost all FR2 deployments use time division duplex (TDD), where uplink and downlink share one channel in alternating time slots.
Beam management is the defining procedure. The base station sweeps a set of beams carrying synchronization signal blocks, the device reports the strongest one, and both sides refine the pair through beam refinement and beam tracking. When a person or vehicle blocks the path, a fast beam failure recovery procedure switches to a reflected path or to another panel within milliseconds.
Frequency bands and channel parameters
FR2 bands are defined by 3GPP and assigned by national regulators, so availability differs by country. The table summarizes the bands that see the most commercial use and their main parameters.
| Band | Frequency range | Typical use |
|---|---|---|
| n257 | 26.5-29.5 GHz | Widely used 28 GHz band in Asia and the Americas |
| n258 | 24.25-27.5 GHz | Common 26 GHz band in Europe and other regions |
| n260 | 37-40 GHz | 39 GHz deployments, mainly in North America |
| n261 | 27.5-28.35 GHz | Subset of 28 GHz used by some US operators |
| n262 | 47.2-48.2 GHz | Newer band with limited deployment |
| n263 | 57-71 GHz | Extended FR2 range added in later releases |
Channel width is the main source of mmWave capacity. A single FR2 carrier can be 50, 100, 200 or 400 MHz wide, and carrier aggregation combines several carriers into 800 MHz or more of usable spectrum. By comparison, a mid-band carrier in the 3.5 GHz range is usually 40 to 100 MHz wide.
| Parameter | Typical value or range | What it affects |
|---|---|---|
| Carrier bandwidth | 100-400 MHz per carrier | Peak throughput and cell capacity |
| Subcarrier spacing | 120 kHz for data | Slot length, latency, phase noise tolerance |
| Cell radius | About 100-500 m in urban streets | Number of sites needed |
| Base station antenna elements | Hundreds per panel | Beam width and link gain |
| Peak device downlink | Several Gbit/s in good conditions | User experience in hotspots |
Propagation limits and why coverage is short
The short reach of mmWave comes from physics, not from immature equipment. Free-space path loss rises with frequency for antennas of a fixed physical size, so a 28 GHz link loses far more energy than a 3.5 GHz link over the same distance unless beamforming gain makes up the difference. Each doubling of distance costs at least another 6 dB in free space.
Obstacles matter even more. Brick, concrete and low-emissivity coated glass attenuate millimeter waves strongly, so outdoor-to-indoor coverage is poor and indoor users are usually served by dedicated indoor cells. A human body or a hand on the phone can cause body blockage of 15 to 30 dB, and trees in leaf weaken signals noticeably. Rain adds attenuation that becomes relevant over longer links, while oxygen absorption near 60 GHz limits that band to very short ranges.
- Line of sight gives the best results; reflections from buildings can sustain a link at lower quality.
- Diffraction around corners is weak, so a user turning into a side street can lose the signal.
- Devices must fall back quickly to Sub-6 GHz bands to avoid interruptions.
mmWave compared with Sub-6 GHz and LTE
mmWave and Sub-6 GHz are complementary layers of the same 5G network. Mid-band spectrum offers the best balance of capacity and coverage, low-band spectrum provides reach, and mmWave provides extreme local capacity. The comparison below shows the trade-offs an operator weighs.
| Property | mmWave (FR2) | Mid-band Sub-6 GHz | Low band |
|---|---|---|---|
| Frequency | 24-71 GHz | About 2.5-7 GHz | Below 1 GHz |
| Typical channel width | 100-400 MHz | 40-100 MHz | 5-20 MHz |
| Cell range | Hundreds of meters | One to a few kilometers | Many kilometers |
| Indoor penetration | Very poor | Moderate | Good |
| Main role | Hotspot capacity | Capacity and coverage | Wide-area coverage |
Most mmWave service runs as 5G NSA or with a Sub-6 GHz anchor, because the mmWave layer alone cannot guarantee a continuous connection. Massive MIMO in mid-band and mmWave beamforming are related ideas, but mmWave uses mostly analog or hybrid beamforming, since a fully digital chain per element would cost too much power at such bandwidths.
Where operators deploy mmWave
Operators deploy mmWave where demand per square meter is extreme and users stay within a small area. The decision table summarizes the typical cases.
| Scenario | Why mmWave fits | Main constraint |
|---|---|---|
| Stadiums and arenas | Tens of thousands of users in a fixed space | Dense site planning and body blockage |
| Transport hubs | High peak demand in halls and platforms | Indoor installations needed |
| Fixed wireless access | Gigabit links to homes without fiber | Outdoor antenna and clear path |
| Private industrial networks | Precise, high-throughput local links | Licensing and equipment cost |
| Wireless backhaul | Fiber-like capacity between sites | Alignment and weather margins |
For fixed wireless access, a customer premises device mounted on a window or roof can use a stable beam toward the base station, which avoids most of the mobility and blockage problems of handsets. This scenario is often the most economically convincing use of mmWave spectrum.
Limitations and common misconceptions
The most common misconception is that mmWave equals 5G. Most 5G traffic worldwide is carried by mid-band and low-band spectrum, and mmWave is a specialized layer. A second misconception is that mmWave signals are uniquely dangerous; they are non-ionizing radiation that penetrates the skin only superficially and are subject to the same exposure limits as other radio services.
- Site density drives cost: covering a city center may need many small cells on lamp posts and facades, each with power and fiber backhaul.
- Device power and heat increase when the modem drives several antenna modules at high throughput.
- Measured speeds vary widely with line of sight, load and orientation, so peak figures rarely describe typical experience.
- Uplink coverage is weaker than downlink because a phone transmits with far less power than a base station.
The simple explanation
Mobile networks send data over radio waves, and different waves behave differently. mmWave uses very high frequencies where plenty of spectrum is still free, so it can carry enormous amounts of data very quickly. The catch is that these waves are weak travelers: they fade fast, and walls, trees and even a hand can block them.
Think of it as a wide, fast highway that is only a few hundred meters long. It is perfect for a crowded stadium or a busy station, but it cannot serve a whole town. Phones therefore switch between mmWave and ordinary 5G bands automatically, using mmWave only when it is close and visible.
How it is used in practice
A football stadium with about 60,000 seats regularly suffers from slow mobile data during matches, because every fan tries to upload photos and video at halftime. The operator installs mmWave radio units under the roof and along the stands, each covering a section of seats with narrow beams in the 26 GHz band and 400 MHz of spectrum. A mid-band 5G layer remains in place as the anchor and fallback.
During the next season, devices that support FR2 move their heavy traffic to the mmWave layer when they have a clear view of a radio unit. The mid-band network is relieved, and upload and download speeds remain usable even at peak moments. The operator measures the result in sustained throughput per section and fewer complaints, and it confirms that mmWave pays off where crowds are dense and confined.