• Socionext and Innatera have jointly developed a human-presence detection system that integrates 60-GHz FMCW radar sensing with neuromorphic edge AI , providing consistent, reliable detection while significantly reducing power consumption. • This technology will be demonstrated live at Embedded World 2026 in Nuremberg (Booth 4A-628). • The system combines Socionext’s compact 60-GHz FMCW radar , which acquires detailed three-dimensional environmental data regardless of lighting or weather conditions, with Innatera’s ultra-low-power spiking neural processor that processes radar signals directly at the sensor edge. • The neuromorphic processor is capable of differentiating human motion from non-human movement, including detecting stationary people, achieving detection accuracy exceeding 99% while consuming less than a milliwatt of power. • This capability allows devices, such as cameras and radios, to remain in low-power sleep modes until actual human presence is confirmed, thereby extending battery life by a factor of three to six. • At Embedded World 2026, visitors will have the opportunity to observe how this combined solution: Accurately detects human presence, including stationary individuals Filters out irrelevant motion caused by animals, plants, or environmental noise Functions entirely on-device without reliance on cloud processing Supports compact, battery-efficient designs suitable for consumer, industrial, and automotive applications Potential use cases include smart doorbells

Socionext and Innatera have jointly developed a human-presence detection system that integrates 60-GHz FMCW radar sensing with neuromorphic edge AI , providing consistent, reliable detection while significantly reducing power consumption.

Article Summaries:

  • Socionext and Innatera have jointly created a human‑presence detection system that fuses a 60‑GHz FMCW radar with Innatera’s ultra‑low‑power neuromorphic edge AI. The radar captures detailed 3‑D environmental data regardless of lighting or weather, while the spiking neural processor analyzes the signals at the sensor edge, distinguishing human motion-including stationary people-from non‑human movement. The solution achieves over 99 % detection accuracy while drawing less than one milliwatt, enabling devices to stay in low‑power sleep until human presence is confirmed and potentially extending battery life by 3‑6×. The technology will be demoed live at Embedded World 2026 and is aimed at consumer, industrial, and automotive applications.
  • Socionext and Innatera have jointly developed a human-presence detection system that integrates 60-GHz FMCW radar sensing with neuromorphic edge AI , providing consistent, reliable detection while significantly reducing power consumption. This technology will be demonstrated live at Embedded World 2026 in Nuremberg (Booth 4A-628). The system combines Socionext’s compact 60-GHz FMCW radar , which acquires detailed three-dimensional environmental data regardless of lighting or weather conditions, with Innatera’s ultra-low-power spiking neural processor that processes radar signals directly at th

Sources: