5 The system half: 5GSAT, phase by phase

5.1 Why there had to be a second half at all

Getting bits between a phone and something in orbit is a radio problem, and the RAN groups took it. It leaves a pile of questions the radio never answers.

Which network should a phone pick when the coverage overhead belongs to somebody else? What should the system do while the satellite is below the horizon? Who is billed, who may listen in, and what does the phone say?

Those are system questions. In 3GPP they sit with SA — the groups that decide what the system must do and how it is put together — and with CT, the groups that write the actual messages. The short name their work items carry is 5GSAT, not NTN.

Nothing in the register labels two halves. The split shows in the short names and in the leadership line each work item carries, and that line is where every claim here about who led what was read.

5.2 Phase 1 went requirements, then architecture, then protocol

SA1 writes what the system has to do. It studied using satellite access in 5G into TR 22.822 FS_5GSAT TR 22.822, then wrote the result into TS 22.261, the 5G service requirements 5GSAT TS 22.261.

SA2 owns the shape of the system. Its study produced TR 23.737 FS_5GSAT_ARCH TR 23.737, and the work item after it put the answer into TS 23.501, TS 23.502 and TS 23.503 5GSAT_ARCH TS 23.501 TS 23.502 TS 23.503.

CT1 led the protocol half, and that piece alone carries 912 documents 5GSAT_ARCH-CT ^[6]. Its own study phase had one question, and the work item description states it plainly.

A PLMN is one operator's network — the thing your phone shows as a network name. Choosing between them when the coverage comes from orbit was enough of a question to need a study of its own, and it produced TR 24.821 TR 24.821.

5.3 Phase 2 is about coverage that comes and goes

A satellite in a low orbit is overhead for a few minutes and then it is not. Phase 2 is what the system does in between.

SA2 studied it into TR 23.700-28 FS_5GSAT_Ph2 TR 23.700-28 and then wrote the rules, in a piece of work carrying 810 documents 5GSAT_Ph2 ^[7].

The Evolved Packet System is the older, 4G core network. It is named there beside the 5G one because a great deal of satellite work lands in the old core rather than the new one.

5.4 Phase 3 named three things and went after them

The phase 3 protocol work item opens by saying what SA2 had already settled, and it lists the three subjects by name.

A regenerative payload is a satellite that demodulates, decodes and re-encodes the signal, so a real piece of the base station is in orbit rather than on the ground ^[8].

Store-and-forward means the satellite takes the traffic, carries it and hands it over later — for a service that can wait, which is what TR 22.865's own scope names first ^[9].

Phone-to-satellite-to-phone is the one where the traffic never touches the ground at all ^[10].

Underneath those three sit the ordinary two steps. SA1's study produced TR 22.865, and its work item put the requirements into TS 22.261 FS_5GSAT_Ph3 TR 22.865 5GSAT_Ph3.

SA2's study produced TR 23.700-29, whose scope names the regenerative payload architecture as its first aim FS_5GSAT_Ph3_ARCH TR 23.700-29 ^[11].

CT1 again led the protocol side, in a piece carrying 862 documents, and its description lists 19 documents it changes — TS 24.301, what a device says to the older core, among them 5GSAT_Ph3_ARCH ^[12] TS 24.301.

5.5 Phase 4 is the one running now

SA1's study produced TR 22.887 FS_5GSAT_Ph4 TR 22.887, SA2's produced TR 23.700-19 FS_5GSAT_Ph4_ARC TR 23.700-19, and the work item turning that into rules is 5GSAT_Ph4-ARC 5GSAT_Ph4-ARC.

Its study's first work task is narrow enough to picture.

Read backwards, that is a phone call, carried by IMS — the telephony core — over a geostationary satellite, on the narrowband radio built for sensors, landing in the old 4G core.

A geostationary orbit is the circular one 35,786 km above the equator, where a satellite appears to stand still ^[13].

The justification says where the pressure came from: satellite work had run through four releases already, and SA1 was writing fresh use cases while phase 4 was being proposed ^[14].

Phase 4's other pieces have barely started. The CT work item carries one document 5GSAT_Ph4-CT, the security and application ones carry none at all 5GSAT_Ph4-SEC 5GSAT_Ph4-APP, and the charging study stands at a completion value of 0.4 FS_5GSAT_Ph4_CH.

What those numbers are, and what they are not, is Where it stands, how to follow it, and what the record cannot say.

5.6 Satellite backhaul runs the other way round

Here nothing reaches the phone from orbit. The mast is an ordinary mast; the satellite carries the link behind it, back towards the operator's network.

SA2 studied it into TR 23.700-27 FS_5GSATB TR 23.700-27 and specified it in Release 18 5GSATB, with a charging item and a management item beside it 5GSATB_CH 5GSATB_OAM.

Release 20 adds one more, for quality of service on a mobile wireless access backhaul carried over satellite, split between SA2 and CT TEI20_QoS_MWAB_SAT TEI20_QoS_MWAB_SAT-CT.

5.7 The trades around the edge

Four more groups have a piece of this, and each has its own run of studies.

SA3 does security and carries 551 documents of this subject SA3. Its studies produced TR 33.700-28, TR 33.700-29 and TR 33.700-30 TR 33.700-28 TR 33.700-29 TR 33.700-30.

SA5 does charging and management and carries 751 SA5. The charging reports are TR 28.844, TR 28.846 and TR 28.894 TR 28.844 TR 28.846 TR 28.894; the management ones are TR 28.808, TR 28.841 and TR 28.874 TR 28.808 TR 28.841 TR 28.874.

SA6 does the application layer and carries 402 SA6, with TR 23.700-01 in Release 19 and TR 23.700-02 in Release 20 TR 23.700-01 TR 23.700-02.

SA4 does speech coding, appears only in Release 20 and carries 272 SA4. Its study is on an ultra low bit rate speech codec, 325 documents of it, aimed at voice over a geostationary satellite FS_ULBC ^[15] TR 26.940.

For scale against those, SA1 carries 711 documents of this subject SA1, SA2 carries 3588 SA2 and CT1 carries 2164 CT1.

5.8 Where this meets the rest

The radio half built its own shelf of documents, and the system half mostly edited documents that already existed. Which is which, and how heavy each one got, is What is on the shelf, and why the bands weigh so much.

How far any of these phases has actually got is a separate question, and the record answers it in a narrower way than most people expect: Where it stands, how to follow it, and what the record cannot say.