Nordic Semiconductor Integrates AI and Multi-Protocol Radios on Single Chip for Ultra-Low-Power Edge Devices

Nordic Semiconductor has unveiled its nRF71 Series of Systems-On-Chip (SoCs), which integrate ultra-low-power Wi-Fi, Bluetooth Low Energy (LE), Thread, and edge AI capabilities onto a single chip. The new series boasts up to 20 times lower standby power consumption compared to Nordic's previous nRF70 Series Wi-Fi products. The nRF7120 and nRF7120E models both feature a 256 MHz Arm Cortex-M33 application processor and a 256 MHz RISC-V coprocessor. Crucially, the nRF7120E variant includes an integrated Axon Neural Processing Unit (NPU) for hardware-accelerated edge AI, enabling machine learning models to run 10 to 17 times faster than on the Cortex-M33 processor alone. This development is critical for practitioners in the embedded systems and IoT space. By consolidating multiple wireless protocols, application processing, and AI acceleration into a single, power-efficient chip, Nordic is addressing key challenges in battery-powered device design. Developers can now create more sophisticated edge AI applications that operate for extended periods without frequent recharging, reducing operational costs and expanding deployment possibilities. The simplified bill of materials and reduced board space also translate to faster development cycles and more compact product designs, which is particularly beneficial for wearables and small form-factor IoT devices. This move aligns with the broader trend of pushing AI capabilities closer to the data source – the edge. As AI models become more efficient and specialized, the demand for on-device inference is growing across various industries. This reduces latency, enhances privacy by processing data locally, and decreases reliance on continuous cloud connectivity. The integration of multiple wireless standards on a single chip also reflects the increasing complexity and interconnectedness of IoT ecosystems, where devices often need to communicate using different protocols. In practice, this means developers should explore the nRF71 Series for their next generation of battery-powered smart devices. The ability to perform AI inference directly on the chip opens up possibilities for real-time anomaly detection, keyword spotting, and image classification in applications like smart home sensors, industrial monitoring equipment, and advanced wearables. Practitioners should evaluate the power savings and performance gains offered by the integrated NPU against their specific application requirements. While the nRF71 Series is currently sampling with selected customers, its eventual general availability will likely accelerate the adoption of more intelligent and autonomous edge devices.