The radio · chapter 6 of 14 · 14 minutes
6 Bands, and why RAN4 carries more of this than anybody
A large share of the short names in this subject are one radio band each, every one led by RAN WG4, and together they are where the biggest document counts come from.
6.1 Why a band gets a piece of work all to itself
A satellite that talks straight to a handset has to do it on some slice of radio spectrum, and the slices set aside for talking to satellites are not the slices a mast uses.
So none of the frequency ranges 3GPP had already written down fitted. Each new range has to be described from scratch: what a device may transmit, what it must be able to hear, how it must behave next to whoever else is using the neighbouring frequencies.
3GPP does that one range at a time, and that is why this subject has so many short names. A band is a row in the work plan, a short name of its own, and a list of changes to the same small shelf of documents.
That work belongs to RAN WG4, the group that writes what a transmitter and a receiver actually have to do. RAN4 carries 7490 documents of this subject over 34 of its meetings, more than any other group anywhere in it.
6.2 What that weight looks like in a room
The document count is abstract until you look at single meetings. RAN4#116 in Bengaluru sat from 2025-08-25 to 2025-08-29 with 633 people registered, and it had 555 documents of this subject on the table — more than any other meeting anywhere in the record [5] R4-116.
The meetings on either side of it are the same shape rather than exceptions: 450 at RAN4#115 in Malta R4-115, 405 at RAN4#117 in Dallas R4-117, 391 at RAN4#118 in the Gothenburg area R4-118 and 319 at RAN4#119 in China R4-119.
Five meetings of one group, each with three to five hundred papers of this one subject on the table. That is what it looks like when a dozen bands are being written at the same time.
6.3 The three parts a radio item is cut into
Look at any radio work item in the work plan and you find the same shape: the
feature row, then a row whose short name ends -Core, usually one ending
-Perf, and sometimes one ending -UEConTest.
The register carries each of those as a separate short name and does not define what the split means. What it does carry is who each row names as its lead and which documents each row promises to change, and those two are enough to read the pattern.
-
The
-Corerows of the band items name RAN WG4, and they are where the documents pile up.NR_NTN_Ku_bands-Corecarries 411 of them [8];NR_NTN_LSband-Corecarries 164 [9]. -
The
-Perfrows name RAN WG4 too and are much smaller. The L and S band performance row carries none at all NR_NTN_LSband-Perf and the Ku band one carries 58 NR_NTN_Ku_bands-Perf. -
The
-UEConTestrows name RAN5 and are the subject of Proving it works — the conformance shelf. Not every band has one, which is the surprise of this chapter and comes at the end of it.
6.4 The NR bands, one at a time
6.4.1 The L and S bands, which came first
NR_NTN_LSband is a Release 18 item: the satellite L and S bands for NR, led
by RAN WG4, planned from 2022-12-13 to 2023-12-12
NR_NTN_LSband. Its own report is TR 38.741, and that report's
whole statement of scope is one line:
That is the register's word for word text and it is all there is. A band report is a working document for the group, not an explanation for outsiders.
The feature row carries 34 documents, its core part 164 and its performance part none NR_NTN_LSband [9] NR_NTN_LSband-Perf.
6.4.2 A second S band, in Release 19
NR_NTN_Sband adds another S band, named in the row's own title as 2000 to
2020 MHz going up and 2180 to 2200 MHz coming down NR_NTN_Sband. It
names RAN4 as its lead, it was planned from 2024-06-06 to 2025-06-06, and it
lists five documents to change.
Its feature row carries 12 documents and its core part 79 NR_NTN_Sband-Core.
6.4.3 The extended L band, the combined L band, and the row that says STOPPED
NR_NTN_combinedLband is the Release 19 item for the extended L band and for
the combined range that joins it to the older one. Its own change table gives
the extended part as 1668 to 1675 MHz going up and 1518 to 1525 MHz coming
down NR_NTN_combinedLband.
Its core part carries 81 documents NR_NTN_combinedLband-Core, and it is one of the few that also name the release-independence document TS 38.307.
Its performance part is the one row in this whole subject whose title tells you what happened to it. It is called "STOPPED - Perf. part: …" in the work plan NR_NTN_combinedLband-Perf.
That word is part of a title somebody typed, not a field the work plan keeps. The same row still carries a completion value of 1 and a planned finish of 2025-09-09. The completion column says the plan for that row is complete and nothing else. Two documents were ever filed against it.
6.4.4 The Ku band, and the reason given for it
NR_NTN_Ku_bands is a Release 19 item, led by RAN WG4, planned from 2024-06-06
to 2025-03-03, with 80 documents on the feature row and 411 on the core part
NR_NTN_Ku_bands [8]. That makes it
the heaviest band item here by a distance.
Its work item description says why it was proposed, and the reason is regulatory rather than technical:
The objective picks up the same regulation, and is about living beside the neighbours rather than about the band itself:
The quotation stops where the register's stored text stops, mid-list. The frequency ranges themselves are in the document and not in this register.
6.4.5 One channel width, and one guard band
Not every RAN4 item is a band. NR_NTN_CBW_30MHz is a Release 18 item for a
30 MHz channel width in frequency range 1, as the row's own title spells it.
It was planned from 2023-06-07 to 2023-09-08 and carries four documents on the feature row and 15 on its core part NR_NTN_CBW_30MHz NR_NTN_CBW_30MHz-Core. It names three documents to change and has no performance row at all.
Release 20 has two more of the same shape: an extended guard band for both radios NR_NTN_IoT_NTN_ext_guard, and higher power classes with a 3 MHz channel width for the bands that already exist NTN_bands_Features_R20.
The first carries no document yet and stands at a completion value of 0; the second carries 47 and stands at 0.34.
6.5 The same exercise on the older radio
The internet-of-things side repeats the band work on the LTE radio, and its band items point at a different shelf of documents.
-
IoT_NTN_extLband, Release 18 — the extended L band, 1668 to 1675 MHz up and 1518 to 1525 MHz down in the row's own title. 30 documents on the feature row, 49 on the core part, four on the performance part IoT_NTN_extLband IoT_NTN_extLband-Core. -
IoT_NTN_FDD_LS_band, Release 18 — an LTE band across L and S, filed as FDD, which is a paired band: one range up, a separate one down. 41 documents on the feature row and 96 on the core part, the heaviest of the three IoT_NTN_FDD_LS_band IoT_NTN_FDD_LS_band-Core. -
IoT_NTN_FDD_S_band, Release 19 — the same 2000 to 2020 and 2180 to 2200 MHz pair the NR item took, done again for this radio. 13 documents and 44 on the core part IoT_NTN_FDD_S_band IoT_NTN_FDD_S_band-Core.
All three name RAN4 as their lead, spelled RAN WG4 on some rows and RAN4 on others. All three collect into the same report, TR 36.764, and that report's scope is as short as the NR one:
6.6 A broadcast band over a satellite that stands still
Two items sit slightly apart from the rest. Their own titles say broadcast rather than access, and both name a geosynchronous satellite.
LTE_band_5G_bcast_GSO is a Release 19 item for a new LTE band for 5G
broadcast in international region 3, over a geosynchronous satellite
LTE_band_5G_bcast_GSO. "Geosynchronous" is the row's own word and
this register does not define it.
What it does define is the geostationary orbit: the circular one 35,786 km above the equator, where a satellite has the same turning time as the earth and so appears motionless to somebody standing on the ground [10].
Its leadership line names RAN WG4 as primary and no secondary group. Six documents sit on the feature row, 49 on the core part and 28 on the performance part LTE_band_5G_bcast_GSO-Core LTE_band_5G_bcast_GSO-Perf.
LTE_band_246_5G_bcast_GSO_BWs is the Release 20 follow-up, adding new
channel widths to that band LTE_band_246_5G_bcast_GSO_BWs.
It has three documents, is planned from 2026-03-03 to 2026-12-12, and a completion value just under 0.15 on the feature row against just under 0.30 on the core part and 0 on the performance part.
Its own change table names three documents LTE_band_246_5G_bcast_GSO_BWs, and they are the internet-of-things shelf again — the device document, the node document and the node's test document.
6.7 Why five documents carry most of the weight
Here is the point the rest of this course needs. The band items do not each write a new document. They each edit the same short list, and they say so in their own change tables.
The NR band items name the same five: the device requirements TS 38.101-5, the node requirements TS 38.108, the radio resource management requirements TS 38.133, the node's conformance tests TS 38.181 and the coexistence report TR 38.863.
The internet-of-things band items name their own five: the device document TS 36.102, the node document TS 36.108, the radio resource management one TS 36.133, the node's tests TS 36.181 and the band report TR 36.764.
That repetition is where the counts come from. TS 38.133 is the most worked-on document in the whole register with 1426 documents naming it, 1306 of them change requests TS 38.133; TS 38.101-5 is next with 838 documents and 771 change requests TS 38.101-5.
6.8 All of them on one page
Ten items, three releases, one group. The document column is the feature row's, not the core part's.
| Item | Release | Documents | Planned finish | Report it collects into |
|---|---|---|---|---|
| NR_NTN_LSband | Rel-18 | 34 | 2023-12-12 | TR 38.741 |
| NR_NTN_CBW_30MHz | Rel-18 | 4 | 2023-09-08 | none |
| NR_NTN_Sband | Rel-19 | 12 | 2025-06-06 | TR 38.863 |
| NR_NTN_combinedLband | Rel-19 | 12 | 2025-06-06 | TR 38.863 |
| NR_NTN_Ku_bands | Rel-19 | 80 | 2025-03-03 | TR 38.863 |
| IoT_NTN_extLband | Rel-18 | 30 | 2025-06-06 | TR 36.764 |
| IoT_NTN_FDD_LS_band | Rel-18 | 41 | 2025-03-03 | TR 36.764 |
| IoT_NTN_FDD_S_band | Rel-19 | 13 | 2025-06-06 | TR 36.764 |
| LTE_band_5G_bcast_GSO | Rel-19 | 6 | 2026-03-03 | none |
| LTE_band_246_5G_bcast_GSO_BWs | Rel-20 | 3 | 2026-12-12 | none |
Two things fall out of that table.
The feature row is almost never where the documents are. The Ku band's feature row carries 80 and its core part 411 [8].
The core part is the heavier of the two in nine of the ten rows above; only the Release 20 one runs the other way, with three documents on the feature row against two on the core part LTE_band_246_5G_bcast_GSO_BWs-Core.
The reports are shared, not one per band. Only the first NR band item produced a report of its own; the three Release 19 ones all collect into the coexistence report TR 38.863, and all three internet-of-things ones into TR 36.764.
That is one more reason those two documents carry the counts they do — 238 documents against the first and 31 against the second TR 38.863 TR 36.764.
6.9 What the band items do not carry
Two absences are worth knowing before you go looking for them.
No NR band item here has a conformance row. The internet-of-things ones do:
IoT_NTN_extLband-UEConTest, IoT_NTN_FDD_LS_band-UEConTest and
IoT_NTN_FDD_S_band-UEConTest, each naming RAN5 and each with a completion
value of 1 IoT_NTN_FDD_LS_band-UEConTest.
What the NR band items do name is the node's own test document TS 38.181, so part of the testing is written inside the band item. The rest is Proving it works — the conformance shelf.
No band item's objective is quoted except the Ku band's. The register quotes the objective of the 41 short names carrying at least 200 documents, and the band items are almost all smaller than that. For the rest, this course can say what the work plan calls them and not what they promised.
6.10 Where this sits in the rest of the course
RAN4 carries more documents of this subject than any other group, and the band items are one part of that load. The work items its own documents carry most are the core and performance parts of the big phase items, not the bands RAN4. Three places to go on:
-
R1. Proving it works — the conformance shelf — RAN5, the conformance rows and the test documents behind them.
-
R2. The coexistence report TR 38.863, which is where the band-by-band radio frequency work is actually written down.
-
R3. The overview course takes the same ground in one chapter, and the quick-start course does not go into bands at all.
Where the numbers in this chapter come from
- the scope of 38.741 clause 1 of the parsed text at /var/www/whatthespec.net/data/friendlyspec/json/38741/19.1.0/, read 2026-08-05
- why NR_NTN_Ku_bands-Core was proposed section 3 of the work item description /var/www/whatthespec.net/data/data/wis/1041135/RP-252911_WID revision for the NTN Ku band cl.md, read 2026-08-05
- the objective of NR_NTN_Ku_bands-Core section 4 of the work item description /var/www/whatthespec.net/data/data/wis/1041135/RP-252911_WID revision for the NTN Ku band cl.md, read 2026-08-05
- the scope of 36.764 clause 1 of the parsed text at /var/www/whatthespec.net/data/friendlyspec/json/36764/19.1.0/, read 2026-08-05
- R4-116 handled 555 Satellites documents, more than any other meeting Satellites documents grouped by meeting, largest first, asked of /var/www/whatthespec.net/data/database/api/api.sqlite on 2026-08-05
- 550 meetings have had Satellites documents on the table distinct values of the meeting column over these documents, asked of /var/www/whatthespec.net/data/database/api/api.sqlite on 2026-08-05
- 6568 documents name satellites or NTN in their title and carry no work item documents whose title contains "satellite", "non-terrestrial" or "NTN" and whose work item column is empty — every plenary approval document among them, and every proposal filed before a work item existed, asked of /var/www/whatthespec.net/data/database/api/api.sqlite on 2026-08-05
- 411 documents carrying the work item NR_NTN_Ku_bands-Core rows of the tdoc table whose work item column names NR_NTN_Ku_bands-Core as a whole word; the first and last meeting are those of its earliest and latest upload time, asked of /var/www/whatthespec.net/data/database/api/api.sqlite on 2026-08-05
- 164 documents carrying the work item NR_NTN_LSband-Core rows of the tdoc table whose work item column names NR_NTN_LSband-Core as a whole word; the first and last meeting are those of its earliest and latest upload time, asked of /var/www/whatthespec.net/data/database/api/api.sqlite on 2026-08-05
- "Geostationary Earth Orbit" as 38.108 defines it clause 3.1 of the parsed text at /var/www/whatthespec.net/data/friendlyspec/json/38108/19.4.0/, read 2026-08-05
Check yourself
Answers appear when you pick one, with where they come from.
Q6.1 Which group carries more documents of this subject than any other?
RAN4 carries 7490 documents over 34 of its meetings. RAN2 is next with 6714 and SA2 leads the groups outside RAN with 3588. RAN4
Q6.2 What does the name of the combined L-band performance row begin with?
The row's own title in the work plan begins with the word STOPPED. It is part of a title somebody typed, not a status field, and the same row still carries a completion value of 1. NR_NTN_combinedLband-Perf
Q6.3 Which core part carries the most documents of any band item here?
411 documents against `NR_NTN_Ku_bands-Core`, well ahead of the L and S band core part at 164. 411 documents carrying the work item NR_NTN_Ku_bands-Core
Q6.4 Why do TS 38.101-5, TS 38.108, TS 38.133, TS 38.181 and TR 38.863 carry such large counts?
Each band item names the same short list of documents in its own change table, so the counts are that repetition rather than one huge argument. TS 38.133
Q6.5 Which band items in this register carry a conformance row of their own?
The three internet-of-things band items each carry a row whose short name ends UEConTest, led by RAN5. The NR band items in this register carry none. IoT_NTN_FDD_LS_band-UEConTest
This chapter was built from a source register generated 2026-08-05. A fresher build of the register may hold different numbers.