Positioning

Bluetooth 6.0 Channel Sounding vs UWB: Node Cost and IP Exposure

What each radio really delivers on a warehouse floor, what a tag and anchor cost per node, and which patents you inherit the day you ship.

Hayat Amin, President of IP, Position Imaging Hayat AminPresident of IP, Position Imaging 4 min read
The short answer

Bluetooth 6.0 Channel Sounding ranges at 10 to 50 cm in open line of sight but degrades to 1 to 3 m around steel racking, while UWB under 802.15.4z holds 10 to 30 cm in the same aisles. A Channel Sounding tag costs roughly half a UWB tag. Neither Bluetooth SIG membership nor IEEE RAND terms license the system-level positioning patents, the solvers, sync, and NLOS handling, that builders inherit with either radio.

Key takeaways

  • Channel Sounding stitches about 80 MHz of effective bandwidth; UWB uses a 499.2 MHz channel, so it separates direct path from reflections roughly six times more finely.
  • Expect 1 to 3 m from Channel Sounding around metal racking, 10 to 30 cm from UWB in the same aisles.
  • A Channel Sounding tag builds on one SoC near $2 at volume; a UWB tag needs a UWB radio plus a BLE companion and lands at 2x or more.
  • PoE cable drops at $150 to $300 each often cost more than the anchor hardware, so anchor density drives system cost more than radio choice.
  • Bluetooth SIG and IEEE 802.15.4z licensing stop at the spec; multilateration, time sync, NLOS mitigation, and fusion patents apply to both radios.
  • Licensing granted positioning IP, such as US 11,774,249, covers the system layer no spec license touches.

What accuracy do Channel Sounding and UWB actually deliver in a warehouse?

Bluetooth 6.0, ratified by the Bluetooth SIG in September 2024, added Channel Sounding: phase-based ranging hopped across up to 72 channels in the 2.4 GHz band, cross-checked with round-trip timing to resist spoofing. Stitching those hops together yields roughly 80 MHz of effective bandwidth. UWB under IEEE 802.15.4z uses a single 499.2 MHz channel with pulses about 2 ns wide, and since light travels 30 cm per nanosecond, UWB can separate the direct path from a reflection off racking that arrives a few nanoseconds later. Channel Sounding cannot.

That difference sets the field results:

  • Open line of sight: silicon vendors like Nordic (nRF54 series) and Silicon Labs demonstrate Channel Sounding at 10 to 50 cm. UWB does the same.
  • Real aisles, with steel racking, forklifts, and foil-lined product: phase-based ranging drifts to 1 to 3 m because strong reflections corrupt the phase slope. UWB with the 802.15.4z scrambled timestamp sequence holds 10 to 30 cm.

If your use case is pallet to aisle, 1 to 3 m works. If it is pallet to slot, it does not. UWB simply starts with six times the bandwidth.

What does each radio cost per tag and per anchor?

A Channel Sounding tag is one chip. An nRF54L15-class BLE SoC runs around $2 at volume, so with battery, antenna, and enclosure a tag BOM lands near $4 to $6. A UWB tag needs a dedicated UWB radio (Qorvo DW3110 or NXP Trimension class parts, roughly $3 to $6 at volume) plus a BLE SoC for discovery and a calibrated antenna, pushing the BOM to $8 to $12. That is the 2x gap everyone quotes.

Infrastructure is where the quote gets misleading. BLE gateway hardware runs $30 to $80; commercial UWB anchors run $150 to $300. But a PoE cable drop costs $150 to $300 per run in a typical warehouse, often more than the anchor it feeds. And the architectures scale differently:

  • UWB TDoA: a tag blinks once, four or more anchors spaced 15 to 30 m apart timestamp it. Thousands of tags per anchor cell.
  • Channel Sounding: each range is a connected two-way exchange, so update rate per node drops as tag count climbs.

Price the install, not the chip.

The spec license stops at the radio, not your positioning system

Bluetooth SIG membership includes a reciprocal, royalty-free license among members to patent claims necessary to implement adopted specifications. That covers the Channel Sounding procedure itself: the tone exchanges, the security steps, the packet formats. IEEE 802.15.4z is thinner still, resting on RAND letters of assurance with no royalty-free pool, and FiRa certification adds interoperability testing, not patent peace.

Neither touches what actually makes an asset tracking product:

  • multilateration and solver logic that turns ranges into coordinates
  • anchor time synchronization for TDoA
  • NLOS detection and multipath mitigation
  • tag scheduling and power management across thousands of nodes
  • fusing RF ranges with inertial, odometry, or camera data
  • zone, geofence, and inventory logic on top

Those are implementation patents held by operating companies and, increasingly, by assertion entities that bought RTLS portfolios. A builder who implements the Bluetooth spec perfectly can still infringe a dozen system-level claims. Spec licenses stop at the radio.

Who holds the ranging patents warehouse builders inherit?

On the UWB side the filings are concentrated. Qorvo acquired Decawave in 2020 and with it foundational two-way ranging and timestamping patents from the DW1000 era. NXP, Samsung, and Apple have filed heavily since Apple shipped the U1 chip in 2019, and Apple's second-generation UWB chip in 2023 signaled that the filing pace is not slowing. Zebra and Siemens hold RTLS system patents from a decade of industrial deployments.

The BLE side is not clean just because the spec is new. Phase-slope ranging in 2.4 GHz shipped in 802.15.4 transceivers years before Bluetooth 6.0, so active patents from that era sit under the new procedure, and fresh filings around Channel Sounding applications are accumulating now.

Position Imaging's granted RF positioning patents are cited as prior art in filings by Apple and Bosch, which is a useful signal of where the foundational claims sit.

A freedom to operate search should cover both the radio you chose and the system layer you built, because the second is where most assertions land. You inherit the filings on day one.

Pick the radio for the job, then license the layer both radios share

The decision rule is short. If aisle-level or zone-level resolution is enough, pallets to aisles, carts to rooms, Channel Sounding wins on tag cost and on reuse of BLE infrastructure you may already own. If the product needs 30 cm, slotting verification, autonomous forklift coordination, tool-level tracking, pay for UWB and its anchors.

Either way you face the same system-layer work: solvers, synchronization, NLOS handling, fusion with other sensors. That is 12 to 24 months of engineering to rebuild, and it is the layer where the patent exposure concentrates. It is also the layer you can license instead of rebuild.

Position Imaging licenses hundreds of granted US patents in RF ranging, real-time positioning, computer vision, and machine learning, including US 11,774,249 and US 12,000,947 in RF-based positioning. A license gives you two things at once: proven methods your team can implement in months rather than years, and freedom to operate on the claims a spec license never covered. Ship the tracker, not the lawsuit.

Patents referenced
US 11,774,249US 12,000,947US 12,066,561US 12,079,006

Frequently asked questions

Is Bluetooth 6.0 channel sounding accurate enough for warehouse asset tracking?

For zone-level tracking, yes. Expect 10 to 50 cm in open line of sight and 1 to 3 m around steel racking and forklifts, which is enough to put a pallet in the right aisle or a cart in the right zone. For slot-level or sub-30 cm requirements, channel sounding's roughly 80 MHz of effective bandwidth cannot separate direct path from reflections well enough, and UWB is the better fit.

Why is UWB more accurate than BLE channel sounding indoors?

Bandwidth. UWB under 802.15.4z uses a 499.2 MHz channel with pulses around 2 ns wide, and light travels 30 cm per nanosecond, so the receiver can pick the first arriving path out of the reflections. Channel sounding reconstructs about 80 MHz by hopping across 2.4 GHz channels, so strong reflections off metal corrupt its phase measurement and the error grows to meters.

What does a UWB anchor cost installed compared to a BLE node?

Commercial UWB anchor hardware runs $150 to $300 against $30 to $80 for a BLE gateway, but the PoE cable drop at $150 to $300 per run often costs as much as either. Total installed cost is driven by anchor density and cabling more than by the radio, so model the building, not the datasheet.

Does Bluetooth SIG membership protect me from positioning patent claims?

Only partially. SIG membership grants a royalty-free license among members to claims necessary to implement the adopted spec, meaning the channel sounding procedure itself. It does not cover system-level patents on multilateration, time synchronization, NLOS mitigation, tag scheduling, or sensor fusion. Those claims apply whether you build on BLE or UWB, and they are where most RTLS assertions land.

How does licensing positioning IP change the build timeline?

Rebuilding the system layer, solvers, sync, NLOS handling, fusion, typically takes an RF team 12 to 24 months before the product is trustworthy. Licensing granted patents and the methods behind them lets a team implement against proven techniques in months, and it converts an unquantified infringement risk into a known line item.

Talk to the IP team

Send us your tag spec and accuracy target, and we will map your product against the portfolio and show which granted patents cover your system layer.

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