8 What the system was asked to do

8.1 Why somebody writes the requirements down first

Before anybody argues about how a satellite connects to a core network, one group has to say what the finished thing must be able to do — and say it in a form other groups can be held to.

That is SA1's job. What the register shows of it is narrow and consistent: every SA1 item in this subject names TS 22.261, the 5G service requirements, as a document it changes, and none of them names an architecture or a protocol document at all.

SA1 carries 711 documents of this subject over 29 of its meetings, running from meeting S1-80 to meeting S1-114. That is a small number beside the radio groups of Bands, and why RAN4 carries more of this than anybody.

8.2 Phase 1: a study, then the requirements

FS_5GSAT is where the system half of this subject begins. It is a Release 16 SA1 study, planned from 2017-09-08 to 2018-06-15, carrying 88 documents over five meetings FS_5GSAT. Its report is TR 22.822, which exists at one version and whose newest file went up on 2018-07-06.

This school cannot quote that report. Its register entry records no parsed text at all TR 22.822, so the report can be named and counted here and nothing more.

5GSAT is the work item that followed — titled "Stage 1 of 5GSAT" in the work plan, Release 17, carrying 109 documents over 22 meetings ^[7].

Its change table names one document, TS 22.261, and describes the change as elaborated requirements for putting satellite components into the 5G system 5GSAT.

Its planned days are worth noticing: 2018-06-06 to 2018-09-14 5GSAT. Yet those 109 documents run from meeting S1-83 to meeting S1-114, the last of them uploaded on 2026-05-26 ^[7].

The plan closed in 2018 and the documents run to 2026, which is one more reason the completion column and the meeting range answer different questions.

8.3 The phase that is missing

Read SA1's items in order and the numbering jumps: phase 1, phase 3, phase 4.

There is no SA1 requirements item for phase 2 in this register. Phase 2 is architecture work about coverage that comes and goes, its study names SA2 as its lead, and it is The architecture, phase by phase — and backhaul, which runs the other way. Why SA1 opened no item of its own that release is not recorded anywhere this register reads.

8.4 Phase 3: the things that can wait

FS_5GSAT_Ph3 is the Release 19 study, planned from 2022-06-02 to 2023-09-09, carrying 205 documents over 16 meetings FS_5GSAT_Ph3. Its justification is two flat sentences that place the work rather than argue for it:

The requirements exist and the radio exists. Its objective then names what the study went looking for on top of that:

The report it produced is TR 22.865, and that report's own scope opens on the same idea in the document's own words:

Store and forward is the idea a satellite can take a message now and deliver it when it next has somewhere to deliver it to. This register carries that phrase and the sentence above; what the study concluded about it is in the report and not here.

5GSAT_Ph3 is the work item that turned the study into requirements: Release 19, 42 documents over nine meetings, sitting on two rows of the work plan, and naming TS 22.261 as the document it changes 5GSAT_Ph3.

8.5 Phase 4: several orbits at once

FS_5GSAT_Ph4 is the Release 20 study, planned from 2024-04-02 to 2024-12-12, carrying 277 documents over seven meetings ^[8]. Its report is TR 22.887, which stands at 20.0.0 and whose newest file went up on 2025-06-27.

The scope of that report opens with the same clause as phase 3's and then names something else first:

Two reports, one opening sentence, two different first items. That is as close as this record comes to telling you what each phase was really about: phase 3 led with the service that can wait, phase 4 with using more than one orbit at once.

5GSAT_Ph4-REQ is the work item after it — Release 20, 33 documents over six meetings, planned from 2024-12-05 to 2025-06-06, and naming TS 22.261 for new requirements and performance targets about satellite access 5GSAT_Ph4-REQ.

8.6 Two words phase 4 had to define for itself

A study that describes something new usually has to name it, and TR 22.887 defines two phrases in its own terms clause.

The first is about getting a message to an ordinary phone when the link is bad. The second is about a fixed dish on the ground. That is a different kind of customer from a handset, and the study defining it in its own terms clause is the register's evidence that phase 4 has both in view.

8.7 What happened to the two reports afterwards

The two quotable study reports have very different afterlives in this record, and the difference is worth reading.

TR 22.865 carries 176 documents of this subject, 14 of them change requests, and stands at 19.2.0 with its newest file uploaded on 2023-12-22.

TR 22.887 carries 269 documents and not a single change request, stands at 20.0.0, and its newest file went up on 2025-06-27.

A report with change requests against it has had changes proposed to it after it was first written. A report with none has not, as far as this record goes. That is all the difference says — not that one was better, not that either is finished, and not that any of the proposed changes was made.

8.8 Regulation, which turns up on both halves of the subject

One phrase in the phase 3 objective is easy to read past: "new use cases and related regulatory requirements". Rules made outside 3GPP are named in the work item's own objective, which is not usual wording for a requirements study.

The same pressure shows up on the radio side. The Ku band item's own justification says the band is allocated for primary satellite use in all three international regions, and that the work exists to bring 3GPP's documents into line with those definitions — see Bands, and why RAN4 carries more of this than anybody.

Requirements and radio, arriving at the same place from different directions. What the register does not hold is any account of which regulator, which proceeding, or what was actually required; only these two sentences, in the work items' own words.

8.9 The four small items, each asking for one thing

Beside the phases sit items with one narrow job each. A reader meets these acronyms and has nothing to look them up against, so here they are.

• SCVS, Release 18 — a 5G system with satellite access supporting control and video surveillance. Eight documents over five meetings, planned from 2021-06-11 to 2021-09-08, naming TS 22.261 SCVS.

• INS_SAT, Release 20 — indirect network sharing on a satellite access network. One document, one meeting, naming TS 22.261 INS_SAT.

• MPS_Sat, Release 20 — priority service where the access is satellite. No document at all, and the one item here whose change table names TS 22.153 rather than TS 22.261 MPS_Sat. That document is not in this register, because nothing else in this subject points at it.

• PWS_NTN, Release 19 — public warning messages over satellite radio. Its leadership line names CT1 rather than SA1, and its change table names TS 23.041, the cell broadcast document, rather than the requirements one. Three work plan rows, 60 documents, CT1 and CT4 PWS_NTN.

8.10 Every SA1 item on one page

Nine rows, five releases, and one document at the end of almost all of them.

Item	Release	Documents	Planned finish	What it produced
FS_5GSAT	Rel-16	88	2018-06-15	TR 22.822
5GSAT	Rel-17	109	2018-09-14	edits to TS 22.261
SCVS	Rel-18	8	2021-09-08	edits to TS 22.261
FS_5GSAT_Ph3	Rel-19	205	2023-09-09	TR 22.865
5GSAT_Ph3	Rel-19	42	2025-09-09	edits to TS 22.261
FS_5GSAT_Ph4	Rel-20	277	2024-12-12	TR 22.887
5GSAT_Ph4-REQ	Rel-20	33	2025-06-06	edits to TS 22.261
INS_SAT	Rel-20	1	2024-09-09	edits to TS 22.261
MPS_Sat	Rel-20	0	2024-12-12	edits to TS 22.153

The pattern is hard to miss once it is laid out. The study is heavier than the work item that follows it in two of the three phases that have both. Phase 3 has 205 documents against the study and 42 against the work item; phase 4 has 277 against the study and 33 against the work item.

Every one of those four numbers was counted the same way — rows whose work item column names that short name — so they can honestly be set beside each other.

Phase 1 is the exception, with 88 against the study and 109 against the work item, and its work item row is the odd one already noted: a plan that closed in 2018 with documents against it into 2026.

8.11 What the record will not tell you about any of this

Three things a reader reasonably wants here, and the register holds none of them.

• Q1. Why any of it was decided. The stores keep a document's outcome word and never the reasoning of the room. Nothing here says why store and forward came first in phase 3 or multi-orbit access in phase 4, only that each report's scope names them in that order.

• Q2. Whether any requirement was contested. No column of any store here records a vote or a tally, so this record can never show anything being carried by a majority.

• Q3. When anything froze or was published. The register holds meeting codes, version numbers, the work plan's planned days and the day a file was uploaded to a server. None of those is a publication date, and the work plan carries no status word at all.

What it does hold is the completion column, and every SA1 item in the table above stands at 1.

8.12 Where all of it lands

Five work items name TS 22.261 in their own table of what they change: 5GSAT, 5GSAT_Ph3, 5GSAT_Ph4-REQ, INS_SAT and SCVS. The document carries 122 documents of this subject and 120 of them are change requests TS 22.261.

That ratio is the signature of a requirements document. Almost nothing is discussed against it; the discussion happened in the study report, and what reaches TS 22.261 is the edit.

8.13 Where this sits in the rest of the course

The requirements say what the system must do. The next question is how it is put together to do it.

• R1. The architecture, phase by phase — and backhaul, which runs the other way — SA2, phase by phase, and satellite backhaul, which runs the other way.

• R2. What the phone says to the network — the protocol half, and what a device actually sends.

• R3. TS 22.261 first, then TR 22.887 for what phase 4 is still arguing about. This school carries the text of neither.

• R4. The overview course treats the requirements side in one section, and the quick-start course names only TS 22.261.