Nordic Semiconductor launches nRF71 series for power-efficient edge AI Wi-Fi devices

Nordic Semiconductor has introduced the nRF71 series of System-on-Chips (SoCs), targeting a reduction in standby power consumption by up to 20 times compared to previous iterations. The new hardware architecture integrates Wi-Fi 6/6E with Bluetooth Low Energy and Thread connectivity, aiming to address the high energy demands typically associated with Wi-Fi-enabled battery devices.

A key feature of the series is the onboard Axon Neural Processing Unit (NPU), which supports machine learning inference locally on the device. By eliminating the need for external AI processors, the chipset aims to improve responsiveness in applications such as industrial sensors, smart home hubs, and wearables. The hardware is designed to handle tasks like image classification and anomaly detection.

"Edge AI will be part of every connected device, but for those that run on a battery, Wi-Fi has demanded a power budget they could not afford," said Kjetil Holstad, EVP of Wi-Fi and Power Management at Nordic Semiconductor. The integration allows developers to manage wireless communication and AI processing on a single platform.

The nRF7120 and nRF7120E models include dual and tri-band Wi-Fi support, 4 MB of non-volatile memory, and 1 MB of RAM. Nordic Semiconductor has confirmed that the series is currently sampling to select customers, with further demonstrations scheduled for the upcoming Electronica trade show.