Daily Dose of 6G

The band that has no number

TR 38.760-4 is the RAN4 part of the 6G radio study — radio frequency requirements, spectrum, coexistence and the tests that decide whether any of it works. It is the last of the four parts this newsletter had not checked, and it is the thinnest document in the set.

Transcribed from 3GPP’s published Release 21 timeline (© 3GPP 2026). Spectrum and band requirements are core RAN4 work, so this edition lands on the same Q4 2028 row as Editions 10 and 11. Note what the table lacks: a Rel-20 RAN4 performance milestone at TSG#116, and no Rel-21 equivalent. RAN4 is the one group whose second deadline is missing from the published picture.

The record, and the set completed

TR 38.760-4, “Study on 6G Radio RAN4 aspects”, 38 series, responsible group RAN4, initially planned Release 20, rapporteur Joseph Schumacher, AT&T. Latest version 0.0.1, uploaded 29 June 2026. Linked work item FS_6G_Radio (1080072) — the same five-deliverable work item that carries the other three parts.

That closes the last of the four version gaps, and the completed table is more interesting than any single row in it.

Version 0.0.1 is a skeleton — headings agreed, little underneath. Edition 11 read RAN3’s high version and old date as a group that banked its decisions early and went quiet. RAN4’s low version and old date says something else: a document that has not started, because its inputs do not exist. You cannot write a coexistence requirement for a waveform whose bandwidth is under study, or a band requirement for a band nobody has been given.

What RAN WG4 owns

3GPP’s group page opens plainly — “The group works on the Radio Frequency (RF) aspects of UTRAN/E-UTRAN/NR” — then lists them: minimum requirements for transmission and reception parameters, radio resource management, channel demodulation and CSI reporting, test procedures for base stations, repeaters, relays and IAB, and over-the-air testing. As with RAN WG3, the formal terms of reference sit in a document the page points at rather than reproduces — RP-210786, approved at RAN#91-e, not opened here.

Five meetings, again

RAN4 keeps its own numbering and sits six behind RAN1, but the weeks and the cities are shared: RAN4#120 ran in Maastricht 24–28 August 2026, the same room-block as RAN1#126, RAN2#135 and RAN3#133. With FS_6G_Radio closing 4 June 2027, five RAN4 meetings remain — #120-bis (South Korea, October), #121 (Calgary, November), #122 (Osaka, January), #122-bis (Lyon, April) and #123 (China, 24–28 May 2027), which ends a week before the deadline. That is the fourth group in a row with exactly five left. The difference is where each starts: RAN1 at v0.4.0, RAN3 at v0.6.0, RAN4 at a skeleton.

What RAN4 actually opened in Maastricht

The same route that worked for RAN3 works here. RAN4’s drafting inbox is on the open web, and its listing carries 19 draft folders for RAN4#120 whose names begin with 6G, spread across 14 numbered agenda items — several split into a part I and a part II, which is itself a signal about volume.

Two things stand out. Spectrum gets an item of its own and a separate one for spectrum sharing — which matches every vendor account of where the difficulty is. And sensing appears twice in the 360s, filed not as architecture, the way RAN3 filed it, but as coexistence and feasibility: sensing that shares spectrum with communication must prove it does not break it.

The numbers 3GPP has not printed

Ericsson’s standardization post of 12 June 2026 — Daniel Chen Larsson, Ricardo Blasco and Per Emanuelsson — carries the only figures with a name attached: system bandwidth from a minimum of 3 MHz in specific spectrum up to 400 MHz, matching the up-to-400 MHz figure Juan Montojo of Qualcomm gave RCR Wireless for the region around 7 GHz. On coexistence the same authors are explicit: “the baseline 6G design assumes a standalone 6G deployment, without any form of aggregation – for example, dual connectivity – between 5G and 6G. To allow a smooth migration to 6G, it is assumed that on the network side that a carrier can be dynamically shared between 5G and 6G.” Separately, they note 3GPP “is assessing how to aggregate millimeter-wave (mmW) and Frequency Range 1 (FR1) spectrum.”

Bands are a different matter, because 3GPP does not have any yet. The ranges in circulation come from regulators and from industry. Nokia’s mid-band explainer — March 2024, two years before this study opened — names upper 6 GHz at 6.425–7.125 GHz and points at 7.125–8.4 GHz and 14.8–15.35 GHz as candidates, with “up to 700 MHz of spectrum in the upper 6GHz band and potentially more than 2 GHz… in the 7–15 GHz range”, subject to a WRC-27 study. The academic framing is wider: Bazzi, Bomfin, Mezzavilla, Rangan, Rappaport and Chafii, in IEEE Communications Magazine, put the upper mid-band at 7.125–24.25 GHz. None of that is a 3GPP band.

One date is worth watching this month. Guy Daniels, in 6G Futures on the March 2026 RAN plenary in Fukuoka, reports that operators asked for preliminary evaluation results on multi-RAT spectrum sharing by September 2026 — to learn whether running 5G and 6G on one carrier costs more overhead than it saves. That plenary round is now upon us: a RAN4-shaped question arriving before RAN4’s document has a first real version.