9 CT1, CT3, CT4 and CT6: where most of it happens

9.1 Why stage 3 lands in four groups instead of one

Stage 3 is the bytes on the wire: the message fields, the encodings, the error cases, the exact interface one network function offers another. It is what a product is built from, and it is the CT groups' job.

Four of them appear in this record, and the division between them is not a committee habit. It follows what a change actually touches.

• CT1 does what the phone says to the network: mobility, call control and session management CT1.
• CT3 does what this network says to a network outside it CT3.
• CT4 does what one core network function says to another CT4.
• CT6 does the smart card CT6.

A single feature usually crosses at least two of those, which is why work items in this subject so often come in a family of rows with one row per group. The map of all fourteen groups is The map: fourteen groups, 309 meetings and 374 names; this chapter takes the four CT ones apart.

9.2 CT1: 2169 documents and the busiest meeting in the record

CT1 carries 2169 IMS documents over 66 meetings, from C1-90 to C1-162, with its first stored document on 2015-03-26 and its most recent on 2026-08-04 CT1.

That first date is where this store begins and not where CT1's IMS work began. The subject opened with a Rel-5 row IMS, and no IMS document in this store carries a work item acronym before 2015, so every count in this chapter is a floor.

The six work item acronyms the register names against CT1 say what the group spends its time on: the Rel-15 move onto the 5G core, and five numbered phases of the protocol alignment work CT1.

9.2.1 What CT1 owns

TS 24.229 is the SIP and SDP call control protocol, stage 3, owned by CT1 TS 24.229. It is the largest target in this whole register: 1680 IMS documents name it as the document they change, and 1675 of those are a change request, a change-request pack or a draft change request ^[2].

The register stores a version of it for every release from Rel-5 at 5.26.0 to Rel-20 at 20.0.0, and thirty different work item acronyms have declared a change to it TS 24.229. A document that has been changed under thirty separate work items across sixteen releases is the load-bearing wall of the subject.

9.2.2 The busiest meeting

CT1#112 met at West Palm Beach, Florida, from 2018-08-20 to 2018-08-24, and 206 IMS documents were on its table — more than at any other single meeting in this record C1-112 ^[3].

The meeting before it, CT1#111 at Osaka in May 2018, had 101 C1-111. The Rel-15 work of Releases 15 to 17: IMS moves onto the 5G core was on CT1's table across that stretch: its 476 documents begin at meeting C1-105 ^[4].

At the other end of CT1's span, C1-90 met at Sorrento in February 2015 with 159 IMS documents C1-90 — the oldest meeting the group's row reaches, and again a floor rather than a beginning.

9.2.3 What a busy year looks like from the meeting rows

The oldest year this store covers is already a heavy one for CT1, and its meeting entries show the rhythm:

Meeting	Place	Days	IMS documents
C1-90	Sorrento	2015-02-02 to 2015-02-06	159
C1-91	Bratislava	2015-04-13 to 2015-04-17	116
C1-92	Sanya	2015-05-25 to 2015-05-29	115
C1-95	Anaheim	2015-11-16 to 2015-11-20	125

Four meetings of that year have an entry here, each of them five days long and each with over a hundred IMS documents on the table. The register describes some meetings and not all 309 of them, so this is a sample of a working year rather than the whole diary.

Three years later the same group hit 206 documents at a single meeting C1-112, which is what closing a release on top of the ordinary load looks like.

9.2.4 What a meeting entry does and does not carry

This is worth being exact about, because a meeting entry looks more complete than it is.

• Days and a place for the meetings this register describes: C1-112 at West Palm Beach, C1-90 at Sorrento, C1-162 at Prague C1-162.

• Nothing at all for four older meetings named elsewhere in this register — SP-8, SP-9, SP-15 and NP-20. The meeting table has no row for those codes, so a chapter may name them and may not say when or where they were.

• No registration count for 31 of them. The attendance store reaches back only a little way. C1-162 carries 149 registered C1-162; C1-112 carries none at all C1-112.

9.3 CT4: 746 documents, and both restoration studies

CT4 carries 746 IMS documents over 48 meetings, C4-68 to C4-135 CT4. The whole IMS record starts at this group's table: the earliest IMS document anywhere in this store carrying a work item acronym went up on 2015-01-23, at meeting C4-68 ^[5].

That meeting sat at Sorrento in the same week as CT1#90, with seventeen IMS documents on CT4's table against 159 on CT1's C4-68 C1-90. Neither number says anything about the ten years of IMS work before it; they say where this store's labelling begins.

CT4's subject is what one core network function says to another, and in this record it has two recurring jobs.

Restoration and resiliency. Both studies are CT4's. TR 29.866, the study on IMS disaster prevention and restoration enhancement, carries 106 IMS documents TR 29.866. TR 29.867, the study on IMS resiliency, carries 73 TR 29.867.

The work item that turns the first of those into rules is IMS_RES-CT, "IMS Disaster Prevention and Restoration Enhancement" — four rows across CT1, CT3 and CT4, with 101 documents IMS_RES-CT ^[6].

It names sixteen documents it would change, which is how far a restoration change reaches through a core network.

Two of those sixteen are CT4's own: TS 23.380, the IMS restoration procedures, with 22 IMS documents against it TS 23.380, and TS 23.334, with 25 TS 23.334.

The second of those two opens its scope clause by pointing at an annex of the stage 2 document and describing what that annex sets up — the first words are cut here and the rest is exact:

Two named boxes and a list of jobs between them, and that is the whole of what this school can say about it. The register holds no text of this document beyond that sentence, so what NAPT-PT, gate control and traffic policing actually do has to be read in the document itself.

The subscriber database's service interface. TS 29.562, the HSS services of the 5G system, is CT4's and carries 230 IMS documents — the second-largest target in this register TS 29.562. Two acronyms are named against it: the Rel-16 service-based architecture work and the Rel-19 restoration work.

9.4 CT3: 719 documents, and the two interworking books

CT3 carries 719 IMS documents over 47 meetings, C3-81 to C3-144 CT3. Its subject is what an IMS network says to a network that is not this one, and two documents carry almost the whole of it.

TS 29.163 is interworking between IMS and circuit-switched networks — the rules for a call that starts in an internet session and ends on an old telephone exchange. 228 IMS documents change it, and it is owned by CT3 TS 29.163.

TS 29.165 is the interface between two IMS networks, the inter-IMS network-to-network interface. 190 IMS documents change it TS 29.165.

Look at what has declared changes to those two and a pattern appears: phase after phase of the protocol alignment work, each one carrying the same instruction — changes to this specification consequent on changes to TS 24.229 made under this work item TS 29.163.

That is the shape of the CT1-to-CT3 relationship in one line. CT1 changes the SIP profile; CT3 changes the two interworking documents to match.

TS 29.163 carries the earliest release label of any document in this chapter. Its Rel-4 and Rel-5 rows both read 1.9.0, and above them sits a version for every release up to Rel-19 TS 29.163.

Two release labels sharing one version number is what the catalogue holds. It explains nothing about that, and neither does this course.

CT3's newest stored IMS document is dated 2025-11-28 and CT4's 2026-05-08 CT3 CT4. Neither date says a group has stopped: the register carries no status field for a work item and no closing date for anything.

9.5 The alignment work is where CT1 and CT3 meet

IMSProtoc7 is four work plan rows across CT1 and CT3, and 364 documents carry it IMSProtoc7 ^[7]. It is the largest of the alignment phases inside Releases 10 to 14, which Releases 10 to 14: the browser, telepresence and the middle years lays out in full.

The register lists five alignment phases against CT1 and one against CT3 CT1 CT3, and both groups' rows on the map are largely made of them. Protocol maintenance is not a side activity here; it is the main road.

9.6 Why a CT work item arrives as a family of rows

Look at the CT work items in this subject and the same shape keeps appearing: one acronym, one description document, and a row apiece for the groups that have to do something about it.

Acronym	Rows	Groups the rows name
NWK-PL2IMS-CT	4	CT1, CT3, CT4
IMS_WebRTC-CT	3	CT1, CT4, and two IETF rows
5GS_Ph1-IMSo5G	4	CT1, CT3, CT4
eIMS5G_SBA	6	SA2, CT1, CT3, CT4
FS_eIMS5G2	4	CT1, CT3, CT4
IMS_RES-CT	4	CT1, CT3, CT4

The rows exist so that each group's own plan carries its own share, with its own target date and its own completion figure. They are bookkeeping, and they are why a row count is never a count of work items.

9.7 CT6: six documents

CT6, the smart card group, carries six IMS documents CT6. Two work item acronyms are named against it: IMS impacts on the card's application aspects IMS_UApAs, and communication control for IMS by the USIM CC_IMS_USIM.

Six is a real number and not an error. Two acronyms, one Rel-10 and one Rel-12, is the whole of what this subject has asked of the card in this record.

9.8 How to find any of this yourself

The four rows of this chapter are all reached the same way, and the method is worth having because it works for any group and any subject.

• Start at the group, not the document. A group entry carries its document count, its meeting count, its first and last meeting, and the acronyms that appear most against it CT4.

• Take an acronym from that list and read its work plan entry. That gives the rows, the groups, the paper that opened it, the rapporteur and the documents it said it would change IMS_RES-CT.

• Then read the documents it names. Each one carries its owner, its stored versions and how many IMS documents change it TS 23.380.

Three steps, each of them a fact somebody counted, and none of them a summary anybody wrote. Where that runs out — and it runs out at every question about why — the honest answer is that the record does not hold it.

9.9 Where this meets the rest of the course

The SA groups that decide what these four are implementing, the two RAN groups that test it and the plenaries that approve it are SA1 to SA5, the two RAN groups, and the plenaries above them.

For what the CT groups were doing when their counts were highest, read Releases 15 to 17: IMS moves onto the 5G core for the Rel-15 to Rel-17 stretch, and Releases 10 to 14: the browser, telepresence and the middle years for the five releases before it.