u-blox announced the expansion of its collaboration with Nordic Semiconductor through the launch of the ALMA-B2 Bluetooth LE module series. The ALMA-B2 modules are based on Nordic Semiconductor’s nRF54LM20 platform. They combine next-generation Bluetooth LE connectivity with a powerful wireless MCU in a compact form factor and include low-latency Edge Machine Learning (Edge ML) capabilities. This marks u-blox’s first module with a dedicated hardware Neural Processing Unit (NPU).
The modules support Bluetooth LE 6.0 and 802.15.4 technologies such as Thread and Zigbee. They feature Channel Sounding for accurate distance measurement, global certification, and flexible antenna options (embedded and external). ALMA-B21 variants include a built-in Axon NPU for hardware-based machine learning.
Key specifications include a 128 MHz Arm Cortex-M33 processor, 2 MB of non-volatile memory, 512 kB of RAM, a RISC-V coprocessor, and HS-USB interface. Advanced security features comprise TrustZone, cryptographic accelerators, tamper detection, debug port protection, and an immutable boot partition.
The modules operate over a temperature range of -40 °C to 85 °C. They deliver up to 15x faster performance with superior energy efficiency for edge processing compared to MCU-only implementations.
The ALMA-B2 series targets intelligent, low-power IoT applications in industrial automation, healthcare, smart home, and asset tracking.
“With ALMA-B2 built on our nRF54L series, u-blox is demonstrating how low-power wireless, advanced security and dedicated ultra-low-power edge AI acceleration can come together in production-ready designs,” said Wojciech Bober, Technical Product Manager at Nordic Semiconductor. “Our collaboration enables developers to move intelligence closer to the edge without compromising latency, power consumption, or scalability.”
Early samples of the ALMA-B2 modules will be available in the next few weeks.
