4 RAN1 and the radio problem

4.1 What makes this radio hard

A normal radio does two expensive things: it generates a carrier and it runs a receiver. Both cost power the whole time they are on. A device budgeted at about a millionth of a watt can afford neither in the usual way ^[2].

So the design question is not "how do we make the radio smaller". It is "what is left when you take away almost all the power", and whether what is left can be made to work well enough to be worth standardising.

That is why the radio side ran two studies before writing a line of binding text, and why one of them is the largest single item in the whole area.

4.2 The two studies, in order

The feasibility study asked whether the thing was buildable at all. It is FS_Ambient_IoT_RAN, unique id 970078, run under the RAN plenary rather than a working group, approved as RP-232404 at plenary RP-101 and sourced from Huawei and HiSilicon FS_Ambient_IoT_RAN RP-232404 RAN.

It ran to 2023-09-08 and stands complete, and its report is TR 38.848 FS_Ambient_IoT_RAN TR 38.848.

The solutions study then went looking for designs. It is FS_Ambient_IoT_solutions, unique id 1020085, led by RAN1 and approved as RP-240826 at plenary RP-103, sourced from CMCC FS_Ambient_IoT_solutions RP-240826.

Its own objective repeats the feasibility wording and narrows it to the working groups:

That study carries 2357 documents — more than the whole of SA3's contribution to the area, and more than any other item anywhere in it FS_Ambient_IoT_solutions SA3. Its report is TR 38.769 TR 38.769.

4.3 What the study was aiming at

The scope of that report is the closest thing this area has to a design target, and it names two devices, not one:

The full scope continues past that sentence. Read together, it names:

• P1. A device at about one microwatt peak, with energy storage, no amplification in either direction, whose answer is backscattered on a carrier wave provided from outside it.

• P2. A device at up to a few hundred microwatts peak, with energy storage and with amplification in one or both directions.

Both carry an initial sampling frequency offset allowance, and the word "harmonized" in the first line is the whole ambition: one air interface design covering both, with the differences kept as small as they can be ^[2].

4.4 The differentiation requirement

Both radio studies carry a demand that is easy to skim past and shaped everything. The feasibility study says the work "shall provide clear differentiation" ^[5], and the solutions study repeats the sentence word for word ^[1].

Differentiation means addressing cases that existing 3GPP low-power technology cannot.

That is a hurdle, not a slogan. A design that merely did what 3GPP already had, a bit cheaper, would fail it. The pressure it creates points one way — towards the microwatt end — and it is why the target in the scope is as extreme as it is ^[2].

The same sentence reappears, word for word, in the normative work ^[3], in the phase 2 work ^[6] and in the outdoor study ^[4]. Four documents, one requirement, carried forward unchanged.

4.5 From study to specification

When the studies concluded, RAN1 led the normative core work, Ambient_IoT_Solutions-Core, unique id 1061084, approved as RP-252105 Ambient_IoT_Solutions-Core RP-252105. Its objective is the feasibility sentence with the verb changed:

"Study" became "standardize". That one word is the difference between a report nobody has to obey and text that products are built against.

The core work carries 1234 documents and stands complete as of 2025-09-09 Ambient_IoT_Solutions-Core. The specification it produced is TS 38.291, the Ambient IoT physical layer, owned by RAN1 TS 38.291.

TS 38.291 is a striking entry in the register: 12 documents and 8 change requests TS 38.291. The arguing happened in the study, over thousands of papers; by the time the text was written there was not much left to argue about.

4.6 Phase two, and the outdoor study

Two RAN1 items are still running.

Phase 2 core, Ambient_IoT_Solutions_Ph2-Core, unique id 1081086, opened as RP-252894 and carries 534 documents Ambient_IoT_Solutions_Ph2-Core RP-252894. Its parent row is Ambient_IoT_Solutions_Ph2 Ambient_IoT_Solutions_Ph2. What it is for is quoted in Study first, then build, then do it again; what it excludes is quoted below.

The outdoor active-device study, FS_Ambient_IoT_Outdoor_Active, unique id 1080073, opened as RP-253394 at plenary RP-110, sourced from LG Electronics, and carries 353 documents FS_Ambient_IoT_Outdoor_Active RP-253394:

Two warnings about that item, both of them the kind of thing that bites somebody reading the work plan quickly:

• R1. Its recorded start date is a leftover from a template — it reads years before Ambient IoT existed as a subject. Never quote it. Its finish date, 2026-03-03, is real, and the study stands complete FS_Ambient_IoT_Outdoor_Active.

• R2. "Outdoor" and "active devices" are the study's own words for its scope. What 3GPP means by an active device in this context is settled inside the study report, and this school does not carry that text.

4.7 The 6G question, answered honestly

The phase 2 RAN work item's scope excludes "the features of 6G relevant to IoT", and that clause is the whole of what this record says about 6G ^[6].

It is enough to know that the exclusion exists. It is not enough to write a sentence about how Ambient IoT relates to 6G, and this course does not.

4.8 Where to read on

RAN1's own two documents are the hop-off points, and this school carries the text of neither:

• TR 38.769 TR 38.769 — the solutions study report, 198 documents against it, and the source of the microwatt targets. Read this first: it is where the reasoning is.

• TS 38.291 TS 38.291 — the Ambient IoT physical layer, the binding text that came out of it.

The layers above the physical one belong to RAN2, and the interfaces, the radio performance and the tests to RAN3, RAN4 and RAN5 — all of which is RAN2, RAN3, RAN4 and RAN5 — protocols, interfaces, performance, tests.