Cheat sheet
Ambient IoT — a tag with no battery
The working reference on 3GPP's Ambient IoT work, across every group that touched it
- Depth
- In depth
- Chapters
- 14
- Reading
- 95 min
- Questions
- 72
- Sources cited
- 176
The shape of it
1 What the work is for
- 1 A tag with no battery, and the gap it was meant to fill
- 2 Study first, then build, then do it again
- 3 Seventeen groups, and where the weight sat
2 The radio groups
- 4 RAN1 and the radio problem
- 5 RAN2, RAN3, RAN4 and RAN5 — protocols, interfaces, performance, tests
3 The system groups
- 6 SA1 and the requirements nobody designed against yet
- 7 SA2 and the specification that did not exist
- 8 SA3, and why a device with no battery is a hard security problem
4 Protocols, management and the layer above
- 9 CT1, CT3, CT4 and CT6 — one work item description, four groups
- 10 SA5 and SA6 — running it, billing it, and building on top of it
- 11 The fifty-three specifications, and what this work put in each
5 Where it stands and where to read on
Keep these in your head
-
1.2 A tag with no battery, and the gap it was meant to fill
Energy harvesting is the whole point. Everything else in this area — the radio, the protocols, the security model, the management — is a consequence of a device that cannot be assumed to be awake, or even to exist, when you speak to it.
-
2.1 Study first, then build, then do it again
Study first, decide, then specify. A study report is a record of the argument; a specification is the settlement. Reading a study as though it were binding is the single most common mistake in this area.
-
3.4 Seventeen groups, and where the weight sat
The order of arrival is the process made visible. Requirements, then architecture and radio, then protocols, then tests. A group that turns up out of order is usually a sign of something being reworked.
-
4.1 RAN1 and the radio problem
Every hard question in this area comes back to one number: a device with roughly a microwatt to spend cannot behave like a phone. The radio, the protocol, the security and the testing all bend around that.
-
5.2 RAN2, RAN3, RAN4 and RAN5 — protocols, interfaces, performance, tests
The words "Topology 1", "Topology 2" and "Device 1" that run through the whole area are defined in the radio overall description, not in the architecture. That is why a change to TS 38.300 ripples into SA2's work and back.
-
6.1 SA1 and the requirements nobody designed against yet
SA1 is the only group in this area that has finished and left. Its first document is 2022-04-29 and its last is 2025-05-09, and there has been nothing since SA1.
- 7.4 SA2 and the specification that did not exist
-
8.1 SA3, and why a device with no battery is a hard security problem
The second sentence is the whole difficulty: mechanisms have to be found that fit inside the energy and complexity budget. Not the best available security — the best security the device can afford.
-
9.4 CT1, CT3, CT4 and CT6 — one work item description, four groups
Stage 2 produced one architecture. Stage 3 then produced 144 change requests to a document that did not exist before it TS 24.369. The pattern is normal: the arguing happens upstream, the writing happens downstream, and it happens quickly.
-
10.3 SA5 and SA6 — running it, billing it, and building on top of it
TS 23.370 having 70 documents and zero change requests is the signature of a specification still being written rather than corrected. The same pattern shows in RAN5's two conformance documents TS 38.591-1.
-
11.2 The fifty-three specifications, and what this work put in each
Fifteen new specifications is the honest measure of how big this area is. A feature that fits into existing documents is an enhancement; one that needs fifteen new documents across five groups is a new subsystem.
-
12.1 Where the work stands, and what a percentage is worth
Nineteen of the 37 work items stand at 100 per cent [1]. That sentence is true of the work plan on 2026-08-04. It is not a claim about the world.
-
13.5 How to follow this area yourself
Reading in that order takes days. Reading in the reverse order — meeting documents first — takes weeks and usually ends in the same place. The reports exist so nobody has to reconstruct an argument from a thousand discussion papers.
-
14.1 What the record cannot tell you
Every item below was checked, not assumed. If a statement is not on this page or somewhere else in this course, it is not that nobody got round to it: it is that the record does not support it.
Easy to get wrong
One from each chapter that has one. The course carries 16 in all.
-
1.5 A tag with no battery, and the gap it was meant to fill
"Battery-less" does not mean "no energy storage". Both the founding definition and the radio scope allow a capacitor. The device stores a tiny amount of energy and spends it; what it does not do is carry a chemical battery somebody has to replace.
-
2.3 Study first, then build, then do it again
A pCR and a CR look the same and mean different things. A pCR edits a draft report that will be published once and then closed. A CR edits live binding text that products are already built against, so it carries a category, a release and a version number. Never quote a pCR as though it changed a specification.
-
3.3 Seventeen groups, and where the weight sat
The "work item names led" column counts names, not items. Four separate CT work items share the one name
AmbientIoT-CT, so a group that leads one of the four still shows the name once. Adding this column down the table gives a number that means nothing. -
4.3 RAN1 and the radio problem
Neither device in that scope is "a battery-less phone". The first one does not transmit in the ordinary sense at all — it reflects a wave somebody else provides. If you read the radio specifications expecting a small version of NR, nothing will make sense.
-
5.4 RAN2, RAN3, RAN4 and RAN5 — protocols, interfaces, performance, tests
TS 38.195 tests the network end — the base station and the carrier-wave node. The device end is tested by RAN5's TS 38.591-1 and TS 38.591-2, a different pair of documents owned by a different group. Reaching for TS 38.195 to test a tag is reaching for the wrong document.
-
6.2 SA1 and the requirements nobody designed against yet
A requirement is not a design and cannot be read as one. When TS 22.369 says something must be possible, that is a constraint on the architecture, not a description of it. The architecture is TS 23.369, and it is a different document by a different group TS 23.369.
-
7.7 SA2 and the specification that did not exist
Phase 1 covered Topology 1 and phase 2 studies Topology 2. It is tempting to guess what those two arrangements are. Do not: the definitions live in TS 38.300 and this school does not hold them, so any explanation here would be invention.
-
8.3 SA3, and why a device with no battery is a hard security problem
"Security for isolated private networks" is a scope, not a caveat somebody forgot to remove. Reading TS 33.369 as though it secured battery-less devices on a public network is reading it for a job it does not claim to do.
-
9.2 CT1, CT3, CT4 and CT6 — one work item description, four groups
Every one of those four rows shows 959 documents, and it is the same 959 counted four times. Document counts in this register are per short name, not per work item. Add the four together and you get a number four times too large, describing nothing at all. The same trap sits under
FS_AmbientIoTand underAmbientIoTAmbientIoT-CT AmbientIoT-CT FS_AmbientIoT AmbientIoT. -
10.2.2 SA5 and SA6 — running it, billing it, and building on top of it
There is no objective sentence to quote for this work item. Its description jumps straight from section 3 to section 5, so the section every other work item in this course is quoted from does not exist in it AmbientIoT_Ph2-OAM. That is a fact about the document, not a judgement on the work — and the fix, if you need its scope, is to open SP-260311 itself SP-260311.
-
11.3 The fifty-three specifications, and what this work put in each
A TR is not a weaker TS. It is a different kind of document with a different life: written, published, closed. Quoting a solution out of TR 23.700-13 as though it were the design is wrong even when the solution was the one chosen — the design is in TS 23.369, and the two can differ TS 23.369.
-
12.4 Where the work stands, and what a percentage is worth
Every date in the right-hand column is a target held in the work plan, not a recorded outcome. Nothing in this data says whether phase 2 is finished, and it is not finished. No chapter here says it is complete, and none says it has failed either AmbientIoT_Ph2-ARC Ambient_IoT_Solutions_Ph2-Perf.
-
13.2 How to follow this area yourself
Rows that share a name share a number. The four
AmbientIoT-CTrows all say 959, and it is one set of 959 documents seen four times. The same holds for the twoFS_AmbientIoTrows, the twoAmbientIoTrows and the fiveAmbientIoT_Ph2-CTrows. Adding the documents column down all 37 rows [2] counts the same paper several times over, so never do it. -
14.3 What the record cannot tell you
The temptation with an unexplained term is to guess from context. "Topology 2" plainly means some second arrangement — but which, and with what in it, decides whether a sentence about phase 2 is right or wrong. The definitions are in TS 38.300 and TS 23.369, and this school holds neither TS 38.300 TS 23.369.
The numbers, and where each came from
Every one was counted out of a file on this machine. Hover a row to read the question that produced it.
- 2022-04-29 the first Ambient IoT document was filed 2022-04-29
- 10 10 of them are studies
- 3973 3973 Ambient IoT documents of type discussion
- 3620 3620 Ambient IoT documents of type pCR
- 1538 1538 Ambient IoT documents of type CR
- 351 351 Ambient IoT documents of type LS out
- 37 37 Ambient IoT work items and study items
- 179 179 Ambient IoT documents in Rel-18
- 8200 8200 Ambient IoT documents in Rel-19
- 2176 2176 Ambient IoT documents in Rel-20
- 10771 10771 Ambient IoT documents
- 16 16 groups filed Ambient IoT documents
- 179 179 meetings carried Ambient IoT documents
- 900 900 different sources filed Ambient IoT documents
- 53 53 specifications are touched by Ambient IoT
- 2026-07-30 the newest Ambient IoT document was filed 2026-07-30
- 220 220 Ambient IoT documents filed in 2022
- 414 414 Ambient IoT documents filed in 2023
- 3384 3384 Ambient IoT documents filed in 2024
- 19 19 of the 37 work items stand at 100%
- 4457 4457 Ambient IoT documents filed in 2025
- 2090 2090 Ambient IoT documents filed in 2026
Where to look it up
The 154 named sources it cites, by what they are.
Specification 53
TR 22.840TR 38.848TS 33.369TS 23.369TS 22.369TS 38.300TR 38.769TS 23.003TS 38.291TS 38.391TS 38.413TS 38.410TS 38.401TS 38.412TS 38.423TS 38.191TS 38.194TS 38.195TS 38.591-1TS 38.591-2TS 22.261TR 23.700-13TS 23.501TS 23.502TR 23.700-30TR 33.713TR 33.71433.340TS 29.569TS 24.369TS 24.007TS 24.501TS 29.522TS 29.369TS 29.518TS 29.510TS 29.504TS 29.506TS 29.571TS 29.505TS 29.50329.568TS 32.240TS 32.254TS 32.291TS 32.298TS 28.540TS 28.541TS 28.552TR 23.700-26TS 23.370TS 23.434TS 23.436
Work item 37
FS_AmbientIoTFS_Ambient_IoT_RANAmbientIoTFS_Ambient_IoT_solutionsFS_AmbientIoTFS_Ambient_IoT_SecAmbient_IoT_Solutions-CoreAmbientIoT-ARCAmbientIoT-SECFS_AmbientIoT_Ph2_ARCAmbientIoT_Ph2-ARCAmbientIoTAmbient_IoT_SolutionsAmbient_IoT_Solutions-PerfAmbient_IoT_Solutions_Ph2Ambient_IoT_Solutions_Ph2-CoreAmbient_IoT_Solutions_Ph2-PerfAmbientIoT_Ph2-CTFS_Ambient_IoT_Outdoor_ActiveAmbient_IoT_Solutions_plus_CT1_SA3-ConTestFS_AIoT_SEC_Ph2AmbientIoT_Ph2-SECAmbientIoT-CTAmbientIoT-CTAmbientIoT-CTAmbientIoT-CTAmbientIoT_Ph2AmbientIoT_Ph2-CTAmbientIoT_Ph2-CTAmbientIoT_Ph2-CTAmbientIoT_Ph2-CTAmbientIoT_Ph2_APP-CTAmbientIoT-CHAmbientIoT_Ph2-OAMFS_AmbientIoT_Ph2_CHFS_AmbientIoT_Ph2_APPAmbientIoT_Ph2-APP
Meeting document 28
SP-220085RP-232404SP-231403RP-240826SP-250377SP-250369RP-252105CP-252196SP-251196SP-240969SP-240506SP-250834SP-260296RP-252894RP-253394RP-253381S2-2605244S2-2605226SP-251246SP-260665S3-262519S3-262517CP-261030CP-261295SP-260311SP-260656SP-250870SP-260315
Quoted wording 19
the gap the founding study namedwhat an Ambient IoT device is, in the founding studywhat an Ambient IoT device iswhat the first study set out to dowhat the RAN feasibility study set out to dothe scope of the RAN solutions study reportthe scope of the architecture specificationwhat an Ambient IoT area iswhat the phase 2 RAN work set out to dowhat the RAN solutions study set out to dowhat the RAN normative work set out to dowhat the outdoor active-device study set out to dowhat the architecture study set out to dowhat the application enablement study set out to dowhat the architecture normative work set out to dowhat the phase 2 architecture study set out to dowhat the security study set out to dowhat the phase 2 security study set out to dowhat the protocol work set out to do
Wg 17
Where to go from here
- The same subject, quick start — Under an hour, four chapters, the shape of the thing.
- The same subject, overview — Enough to follow the conversation and know what to look up.
- Every question of this course — 72 of them, on one page.
- The final test — 20 questions, 70 per cent to pass.
- Every source behind the course — 176 of them, each with its method.
The raised numbers on this page
- 19 of the 37 work items stand at 100% — Work Plan rows of the Ambient IoT work items whose completion field is 1.0, read 2026-08-04
- 37 Ambient IoT work items and study items — folders under /var/www/whatthespec.net/data/data/wis whose metadata.json carries an acronym matching AmbientIoT, Ambient_IoT or AIoT; all 9406 folders were read and 5 other 'ambient' items were rejected, read 2026-08-04