What Bluetooth 6.0 Channel Sounding Removes From Your RTLS BOM

What did Bluetooth 6.0 actually add?

The Bluetooth SIG adopted Bluetooth 6.0 in September 2024, and its headline feature is Channel Sounding: a standardized two-device ranging method that combines phase-based ranging with round-trip timing. The radio steps across up to 72 narrow channels in the 2.4 GHz band, measures phase at each hop, and reconstructs distance from the composite response. Silicon followed quickly. Nordic's nRF54L15 and NXP's KW47 both run channel sounding on the same die that handles advertising, connections, and over-the-air updates.

What changed is where ranging lives. Before 6.0, sub-meter distance meant bolting a second radio onto the tag, almost always an 802.15.4z UWB chip. Now the radio you were already shipping can measure distance to a few tens of centimeters in clean line of sight, and closer to a meter on a working floor with racking and people.

What did not change is bandwidth. The entire 2.4 GHz ISM band spans about 83 MHz. UWB uses a single 499.2 MHz channel. Every accuracy difference between the two flows from that ratio. The radio changed. The physics did not.

Which BOM lines does channel sounding delete?

Walk a dual-radio UWB tag design and mark what a Bluetooth 6.0 single-chip design removes:

• The UWB transceiver. A Qorvo DW3110 or NXP Trimension part runs roughly $3 to $6 in volume. A channel-sounding-capable BLE SoC lands under $2 and replaces both radios.
• The second antenna and matching network. UWB radiates at 6.5 or 8 GHz, so it needs its own antenna, its own RF layout, and tighter board tolerances than a 2.4 GHz trace antenna.
• Dual-radio coexistence firmware. One radio means no arbitration layer between BLE provisioning traffic and UWB ranging sessions, which shortens bring-up and shrinks flash.
• A certification line. UWB interoperability typically means FiRa certification on top of Bluetooth qualification. Single-radio tags qualify once.

For a tag program shipping tens of thousands of units, the deleted silicon and antenna lines alone often cut tag cost by more than half, before counting the smaller board and enclosure. One chip now does the whole tag.

Which BOM lines survive no matter which radio you pick?

The tag is the cheap half of an RTLS bill of materials. The lines that dominate a deployment quote do not mention the PHY at all: anchors or locators, the PoE switch ports behind them, the cable drops, the lift rental and electrician hours to mount hardware at height, the site survey, and the location engine license. On most warehouse quotes, installation labor and cabling per anchor exceed the anchor hardware itself, and that ratio holds whether the anchor listens for UWB blinks or runs channel sounding sessions.

One line can actually grow. Channel sounding is connection-based and pairwise: a tag ranges with one responder at a time, consuming airtime per measurement. UWB TDoA inverts this, letting a tag transmit one blink that four or more anchors timestamp simultaneously. As tag counts climb into the thousands, channel sounding either accepts lower update rates per tag or adds responders to spread the load. Model that before quoting anchor counts. Cabling does not care which radio you picked.

Where does UWB still earn its premium?

Range resolution scales with bandwidth: roughly c divided by twice the bandwidth. UWB's 499.2 MHz channel resolves reflected paths about 30 cm apart from the direct path. Channel sounding's 80 MHz cannot separate reflections closer than roughly 1.9 m, so a signal bounced off steel racking blends into the direct path and the math has to estimate its way out. That is why UWB holds 10 to 30 cm in cluttered, metal-heavy aisles while channel sounding drifts toward a meter or worse exactly where warehouses need precision most.

UWB also keeps three operational advantages:

• Update rate at scale. TDoA blinks support thousands of tags at multiple hertz each.
• Latency. Sub-100 ms position updates make forklift proximity warnings viable; a half-second-old position does not stop a 3-ton vehicle.
• NLOS behavior. A 2 ns pulse gives the receiver a fighting chance to find the leading edge through partial blockage.

For pallet-level zone tracking, none of this matters. For vehicle safety, docking, or pick verification, all of it does. Below half a meter, UWB still wins.

How should you split the BOM, and where does licensed IP fit?

The 2026 decision rule is simple. If a meter of error never changes the operational answer, such as which bay a tote sits in or which room holds the infusion pump, put channel sounding tags on those assets and pocket the deleted BOM lines. If the workflow breaks below half a meter, or involves anything moving fast, keep UWB on those assets. Most real sites end up hybrid: thousands of cheap single-chip tags, dozens of UWB-equipped vehicles, one location engine fusing both.

Here is the line item neither radio deletes: the intellectual property under the ranging itself. Multipath mitigation, phase-slope estimation, anchor calibration, and RF-plus-vision fusion are patented techniques, and swapping a UWB chip for a Bluetooth 6.0 chip moves you between crowded patent territories, not out of them. Position Imaging licenses granted US patents in radio-frequency ranging and position tracking, including US 11,774,249 and US 12,066,561, from a portfolio cited in filings by Apple and Bosch. Teams that license ship the algorithm layer in weeks instead of reinventing it under freedom-to-operate risk. License the ranging IP, spend the savings shipping.