Bluetooth 6.0 Silicon Is Shipping: Which Ranging Stack to Bet On
Bluetooth 6.0 channel sounding chips are in volume production. Here is how to pick a ranging stack your product can live with for five years.
Hayat AminPresident of IP, Position Imaging
4 min read
Bet on UWB if your product needs accuracy under 50 cm: its 500 MHz channels deliver 10 to 30 cm where Bluetooth channel sounding delivers 0.5 to 1 m indoors. Bet on BLE 6.0 channel sounding when 1 m is enough and tag cost rules, since shipping silicon like Nordic's nRF54L15 ranges on the same $1 to $2 chip that already runs telemetry. Mixed estates should run both and invest in the fusion layer above the radio.
Key takeaways
- Bluetooth 6.0 channel sounding delivers 0.5 to 1 m indoors on $1 to $2 silicon; UWB holds 10 to 30 cm but adds roughly $3 to $5 per tag.
- Bandwidth sets the floor: UWB's 499.2 MHz channels resolve paths about 30 cm apart, while channel sounding's roughly 72 MHz cannot separate close reflections.
- If you range against phones, UWB owns the installed base, shipping in every iPhone since 2019; channel sounding handsets only started arriving in 2025.
- Tag-to-anchor systems in buildings you control can ignore phone support entirely and choose on tag cost and accuracy alone.
- Hybrid wins mixed estates: channel sounding for zone-level tracking, UWB anchors only in the cells where sub-30 cm pays for itself.
- The durable investment is the fusion and tracking layer above the radio, which can be licensed instead of built over 18 to 24 months.
What does Bluetooth 6.0 channel sounding actually ship in 2026?
The Bluetooth SIG published Channel Sounding in Core 6.0 in September 2024. Two years later the silicon is in volume: Nordic's nRF54L15 and NXP's KW47 and MCX W72 run the feature in production firmware, with more vendors sampling parts. The method combines phase-based ranging, which measures phase across tones hopped over roughly 72 MHz of the 2.4 GHz band, with round-trip timing that acts as a check against relay attacks and gross multipath errors.
Vendor demos in open line of sight show 30 to 50 cm. On a real floor with racking, people, and reflective ceilings, plan on 0.5 to 1 m, with occasional multi-meter outliers when the direct path is blocked. That is still a large step up from RSSI beacons, which drift 3 to 5 m, and it arrives on the same $1 to $2 SoC that already handles your sensor telemetry and firmware updates. No second radio, no second antenna, no FiRa certification cycle. The chips are real. So is the meter.
Why does UWB still win below half a meter?
Range resolution scales with bandwidth. The rule of thumb is c divided by twice the bandwidth: UWB's 499.2 MHz channels under IEEE 802.15.4z resolve paths about 30 cm apart, while channel sounding's roughly 72 MHz of measured spectrum resolves about 2 m and relies on phase super-resolution math to do better. Super-resolution works when the direct path is strong. It degrades exactly where you need accuracy most: dense racking, metal, crowds.
Field behavior follows the math:
- UWB two-way ranging: 10 to 30 cm typical, usually holding under 50 cm without line of sight, because the wideband pulse separates the first arrival from its reflections.
- Channel sounding: 0.5 to 1 m in line of sight, degrading to several meters when the direct path is attenuated, because overlapping reflections inside 72 MHz cannot be pulled apart.
If your product promise is "which shelf", "which stall", or "keep the forklift 1 m from people", that gap is your product. Bandwidth is the spec that cannot be negotiated.
Phones, not datasheets, decide which stack matures first
UWB has a seven-year head start on handsets. Apple has shipped a UWB chip in every iPhone since the iPhone 11 in 2019 and exposes ranging through Nearby Interaction. Premium Android devices from Samsung and Google carry UWB as well. Channel sounding on phones only started in 2025: Android 16's RangingManager API puts channel sounding, UWB, and Wi-Fi RTT behind one interface, and the first handsets with Bluetooth 6.0 radios are reaching the market now. Apple has said nothing about channel sounding support.
That split shapes the bet. If your product ranges against the user's phone, a finder tag, keyless entry, tap-to-locate, then UWB covers the paying installed base today, and channel sounding covers the midrange Android base in two to three years. If your product ranges tag-to-anchor in a building you control, phone support is irrelevant and you can choose purely on tag cost, accuracy, and update rate. Most teams conflate these two cases and buy the wrong radio for it. Phones decide the consumer market, not datasheets.
How do you choose a stack you can live with for five years?
Write the accuracy requirement as a user-visible promise before opening a datasheet, then apply three rules.
- 1 m or coarser, battery first: channel sounding. Zone-level asset tracking, room-level wayfinding, presence. One SoC, one antenna, tag BOM near $2, and a CR2032 lasts years at low update rates.
- Under 50 cm, or safety-rated: UWB. Forklift proximity stops, picking to the slot, finding one tote in a dense rack. Budget $3 to $5 more per tag plus FiRa interop testing.
- Mixed estate: hybrid. Run channel sounding for the 90 percent of assets that need zone accuracy, and reserve UWB anchors for the aisles or cells where sub-30 cm pays for itself.
Then price the update rate, not just the tag. A UWB two-way ranging exchange takes under a millisecond of air time; a full channel sounding procedure across its hopped channels takes tens of milliseconds, which caps how many tags one anchor can serve per second at a given accuracy. Pick the cheapest radio that meets the spec.
What happens if you bet on the wrong radio?
Radios will keep churning. Channel sounding silicon will improve, UWB is spreading into cheaper phones, and 802.11bf Wi-Fi sensing is coming up behind both. The part of your stack that should not churn is the system layer above the radio: fusing ranges into positions, rejecting multipath outliers, scheduling thousands of tags per anchor, and keeping one track continuous when an asset crosses from a UWB cell into a channel sounding zone. That layer takes most teams 18 to 24 months to build, and it is where most positioning patents sit, regardless of which radio produced the range.
That is the hedge Position Imaging sells. Its portfolio of hundreds of granted patents in real-time positioning, RF ranging, computer vision, and machine learning, including US 11,774,249, US 12,079,006, US 12,066,561, and US 12,000,947, is cited by Apple, Bosch, and other major firms. Licensing the system layer means your ranging estimates can come from a UWB timestamp today and a channel sounding phase sweep next year without rebuilding the product or re-clearing freedom to operate. Hedge the radio bet at the system layer.
Frequently asked questions
How accurate is Bluetooth channel sounding indoors?
Plan on 0.5 to 1 m in line of sight on a real floor, even though vendor demos in open air show 30 to 50 cm. Accuracy drops to several meters when the direct path is blocked, because the roughly 72 MHz of measured spectrum cannot separate overlapping reflections. For zone-level or room-level answers that is enough; for shelf-level it is not.
Does BLE 6.0 ranging replace UWB?
No. UWB's 499.2 MHz channels under IEEE 802.15.4z deliver 10 to 30 cm and hold up better without line of sight, which a 2.4 GHz phase measurement cannot match. Channel sounding replaces RSSI beaconing and competes with UWB only where 1 m accuracy is acceptable. Many deployments will end up running both.
Which phones support Bluetooth channel sounding in 2026?
The first Android handsets with Bluetooth 6.0 radios began arriving in 2025, and Android 16 exposes the feature through the RangingManager API alongside UWB and Wi-Fi RTT. Apple has not announced channel sounding support; every iPhone since the iPhone 11 in 2019 carries UWB instead. If your product ranges against iPhones, UWB is the only option today.
How much does switching a tag from UWB to channel sounding save?
Roughly $3 to $5 of BOM, because the UWB transceiver, its antenna, and front end come off the board and ranging moves onto the $1 to $2 BLE SoC you already carry for telemetry. You also skip the FiRa certification cycle. In exchange you give up the difference between about 30 cm and about 1 m, so price that against what your spec actually requires.
Can one system run both channel sounding and UWB?
Yes, and for mixed estates it is usually the right answer. Use channel sounding tags for the majority of assets that need zone accuracy, and place UWB anchors only where sub-30 cm precision pays, such as picking faces or forklift safety zones. The hard part is the fusion layer that keeps one track continuous across both radios, which is system software worth licensing rather than building.
Tell us the accuracy your product promises and we will map it to the portfolio, radio by radio, in one working session.
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