Ambient IoT vs UWB After 3GPP Release 19: What Should You Build On?
Release 19 made batteryless tags real, but it standardized presence, not position. A builder's rule for choosing between ambient IoT and UWB RTLS.
Hayat AminPresident of IP, Position Imaging
4 min read
Build on UWB if your product must report where an item is: IEEE 802.15.4z ranging delivers 10 to 30 cm today with shipping silicon. Build on 3GPP Release 19 ambient IoT only when the question is whether an item is present in a zone, because the spec defines inventory and command operations, not ranging, so location resolves to the read zone of whatever base station or phone heard the tag. Most tracking products at scale will end up pairing both.
Key takeaways
- 3GPP Release 19 ambient IoT defines two backscatter device classes, roughly 1 microwatt and a few hundred microwatts, for inventory and command operations, not for ranging.
- Batteryless tag location accuracy is the read zone, often 10 to 50 meters indoors, while UWB holds 10 to 30 cm in production.
- Ambient IoT tag economics target cents per tag with zero battery maintenance, which wins at five and six figure tag counts.
- UWB silicon ships in volume today (Qorvo DW3000 series, NXP Trimension); standards-based ambient IoT silicon and network support land around 2027 at the earliest.
- The durable architecture is hybrid: batteryless tags on inventory, UWB tags on the forklifts, carts, robots, and people that need live position.
What did 3GPP Release 19 actually standardize for ambient IoT?
3GPP approved the Release 19 ambient IoT work item in March 2024, built on the Release 18 study in TR 38.848, and the specification work closed out across late 2025 and 2026. The scope is deliberately narrow. Two backscatter device classes made the cut: one drawing about 1 microwatt with no energy storage at all, and one drawing up to a few hundred microwatts with a small store of harvested energy and an amplified reflection link. Both operate indoors in licensed FR1 spectrum, in two topologies: tag talking to a base station directly, or tag talking to an intermediate device such as a phone acting as reader. The supported operations are inventory (which tags are in range) and command (read or write a small payload), at data rates in the low kilobits per second. What is missing matters more to a product team than what is included: no fine timing measurements, no ranging procedure, no positioning reference signals for these devices. Release 19 standardized presence, not position.
How precisely can a batteryless tag be located?
Location accuracy is bounded by what the tag can do, and a backscatter tag cannot generate precise timestamps. Ranging resolution scales with signal bandwidth. UWB under IEEE 802.15.4z uses 499.2 MHz channels near 8 GHz, which is why two-way ranging and TDoA systems hold 10 to 30 cm in real buildings, through forklift traffic and racking. A Release 19 ambient tag reflects a narrowband carrier instead. The network learns which reader heard it and roughly how strong the reflection was, and that resolves to a read zone: a dock door, an aisle, a room, commonly 10 to 50 meters across indoors. That is the granularity RAIN RFID portals have delivered for years, now with longer range and a path onto carrier and private 5G networks. Academic systems have pushed backscatter ranging much tighter using carrier phase across frequency hops, but none of that is in the spec you can build against today. If your datasheet promises sub-meter, a batteryless Release 19 tag cannot sign it. The zone is the resolution.
Ambient IoT wins on cost per tagged item, not on accuracy
The case for ambient IoT is unit economics at large tag counts. A RAIN RFID label costs 5 to 15 cents, and ambient IoT tags target the same order once volumes arrive, because the bill of materials is an antenna and a tiny chip with no battery, no crystal, no charger. A UWB tag runs $10 to $30 and carries a battery you must track, replace, and dispose of. That turns into standing labor: a fleet of 50,000 UWB tags on a three year battery life means more than 60 battery swaps every working day, forever. So the sweet spot for batteryless tags is high item counts with zone-level questions: returnable transport items, apparel on the sales floor, pharmacy totes, tool cribs with thousands of small assets. The question those operations ask is "is it here, and how many", and a dock-door or aisle answer closes the ticket. Battery-free BLE tags from Wiliot already serve this niche ahead of the standard, with the same zone-level location ceiling. At a million tags, batteries lose.
UWB is still the only standard that ships sub-meter today
Build on UWB when position is the product. Forklift collision warning needs the distance between a vehicle and a worker to about 30 cm, under 100 ms, several times a second. AMR fleets consume continuous pose corrections. Hospital workflow tools need bay-level and bed-level certainty, not a room guess. None of those survive zone resolution or the latency of an inventory round. The supply side points the same way in 2026: Qorvo DW3000 series and NXP Trimension parts ship in volume, FiRa 2.0 certifies interoperable ranging between vendors, and Apple has put UWB in every flagship iPhone since 2019, with a second generation chip since the iPhone 15. Standards-complete ambient IoT is at a different stage of life. Chipsets, reader infrastructure, and carrier or private network support all arrive after the spec freezes, which puts dependable commercial deployments around 2027 or 2028. A product shipping in 2026 cannot wait for that stack, and a safety function should not run on version one of it regardless. For live position, UWB has no peer.
The practical answer is a hybrid, and the hard part is the fusion layer
The decision rule is short. If your product answers "do we have it, and roughly where", build on ambient IoT or RAIN RFID and take the cents-level tags. If it answers "where exactly is it right now", build on UWB. Most deployments at scale will do both: batteryless tags on the inventory, UWB tags on the movers, with one location engine fusing zone reads and precise ranges into a single model of the building. That fusion layer, and the RF ranging math under it, is where engineering teams burn 12 to 24 months, and it is where the granted patents already sit. Position Imaging licenses granted US patents across radio-frequency ranging, real-time positioning, and vision-aided tracking, including US 11,774,249, US 12,079,006, US 12,066,561, and US 12,000,947, from a portfolio cited by Apple and Bosch in their own filings. Licensing buys the hybrid stack a head start measured in quarters, plus freedom to operate while the ambient IoT market finds its footing. License the map, then go build.
Frequently asked questions
What positioning accuracy does 3GPP Release 19 ambient IoT support?
Zone level only. Release 19 defines inventory and command operations for backscatter tags, with no ranging procedure or positioning reference signals. Location resolves to the read zone of the base station or intermediate device that heard the tag, typically 10 to 50 meters across indoors.
Can ambient IoT replace a UWB RTLS?
Only where zone presence is the actual requirement, such as counting inventory or confirming a tote passed a dock door. Anything that needs sub-meter, low-latency position, like forklift safety or robot coordination, still requires UWB's 499.2 MHz ranging. The two solve different problems and often ship together.
When can I actually deploy standards-based ambient IoT?
The Release 19 specification work is done, but chipsets, readers, and carrier or private 5G network support follow the spec, so dependable commercial deployments look like 2027 to 2028. If you need batteryless zone tracking sooner, pre-standard options such as Wiliot's battery-free BLE tags work today.
What does an ambient IoT tag cost compared with a UWB tag?
Ambient IoT tags target cents per unit at volume, in the same range as RAIN RFID labels at 5 to 15 cents, with no battery to replace. UWB tags run $10 to $30 each plus ongoing battery maintenance, which is why batteryless tags win at high item counts and UWB stays on the smaller population of assets that need precise position.
Which should a startup build on in 2026?
If position is your product, build on UWB now; the silicon, FiRa 2.0 interoperability, and phone support already exist. Keep your location engine tag-agnostic so you can ingest ambient IoT zone reads when the Release 19 hardware arrives, rather than rearchitecting in 2028.
Send us your tracking spec and we will map it against the portfolio, patent by patent, so you know what to license before you write the firmware.
Tell us the product. We map the exact scope, what a license covers, and how fast you can ship, all in a 20-minute call.
Book a 20-minute call