Nordic Demos Bluetooth HDT on Silicon: 3.8 Mbps in 2 MB Transfer

Nordic Semiconductor has demonstrated Bluetooth High Data Throughput (HDT) on prototype silicon, moving a 2 MB file at roughly 3.8 Mbps — nearly four times the ceiling of today's Bluetooth Low Energy links. The company announced the demonstration on October 29, positioning itself among the first silicon vendors to show the upcoming enhancement running in working hardware.

The stakes are architectural. Bluetooth LE currently tops out at 1 Mbps and 2 Mbps physical layers, which constrains devices that must move firmware updates, richer sensor data, or larger files. Manufacturers facing those limits have often added a Wi-Fi radio, which brings extra hardware, power draw, and integration complexity. HDT, expected to enter the Bluetooth specification in late 2026, aims to close much of that gap without a second radio.

What did the demonstration actually measure?

Nordic ran the test between two prototype systems-on-chip in an office environment, using 508-byte payloads. The results:

• HDT (fastest configuration): about 3.8 Mbps average throughput, completing a 2 MB transfer in roughly four seconds.
• Bluetooth LE 2M: about 1.4 Mbps, 12 seconds for the same file.
• Bluetooth LE 1M: about 0.76 Mbps, 22 seconds.

Nordic cautions that these are preliminary numbers, not guaranteed real-world performance. More detailed laboratory benchmarking will follow once the silicon and the specification mature.

The takeaway: HDT works on real hardware, but the measured figures remain early demonstration data rather than shipping-product specifications.

How does the speed gain work?

Unlike earlier Bluetooth LE speed increases, HDT changes both the radio and the packet structure. It introduces additional physical layers built on more advanced modulation schemes, combined with forward error correction. It also supports longer packets, which reduce the share of airtime consumed by protocol overhead.

The specification defines five new PHYs — HDT2, HDT3, HDT4, HDT6, and HDT7.5 — each trading throughput against radio sensitivity and operating conditions. Faster modes need better link quality; slower modes hold up when signal conditions degrade. A link-quality feedback mechanism supports dynamic switching between PHYs as conditions change.

In plain terms, the radio adapts its speed to whatever the link can actually carry at a given moment.

Where are the trade-offs?

Higher throughput does not help every application equally. Nordic's own technical analysis indicates HDT may perform worse than existing LE 2M for very short packets, because the efficiency gains come mainly from longer transmissions using the available bandwidth more fully.

That matters for the bulk of battery-powered sensors. A device sending occasional temperature readings or short status messages may see little or no benefit over current Bluetooth LE. The clear winners are applications that periodically exchange substantial data: connected healthcare devices and wearables synchronizing stored information, smart home products taking firmware updates, and industrial equipment transferring diagnostic files during maintenance.

There is also a hardware constraint. Because HDT requires new modulation capabilities, existing Bluetooth LE SoCs cannot gain full HDT support through firmware updates alone. New silicon is required — at both ends of the connection.

What does this change for device makers?

For OEMs, the most consequential possibility is dropping the second radio. Some products currently pair Bluetooth LE for low-power connectivity with Wi-Fi for bulk transfers. If HDT's bandwidth covers an application's needs, that dual-radio design could collapse into a single-chip architecture, cutting cost and complexity — though actual power and cost benefits depend on future HDT-capable chipsets and their implementation.

Nordic also intends to integrate HDT into its established software and development environment, so customers can work with familiar tools rather than a new connectivity stack.

The announcement is a technology demonstration, not a product launch. Nordic has not disclosed a release date or detailed specifications for HDT-capable SoCs. The Bluetooth Special Interest Group is expected to finalize the HDT specification in late 2026, and Nordic plans to showcase the technology at electronica 2026 in Munich.

The open question is ecosystem adoption: higher throughput requires compatible hardware in smartphones, gateways, and other hosts, so HDT's commercial impact will hinge on how quickly the receiving end of Bluetooth links catches up.