At Sensors Expo 2019 in San Jose, CA, Chet Jewan, Vice President of Sales and Business Development at Eta Compute, describes the demos in his booth. The sensor node consists of the Eta Compute ECM3531 application specific integrated circuit (ASIC), a ROHM Semiconductor RF device and sensors – all running off of a small solar panel. The asynchronous low power ECM3531 is always on and consumes only 250 µA.
Light and temperature data can be easily transmitted with the board powered simply by solar energy with no external batteries to provide additional power in the internal lighting environment.
The company also offers a power efficient key word spotting (KWS) application on an eval board with a microphone for voice detection. Trained with a Google data set, the board is always on for continuous detection. It also consumes only 250 µA.
Based on the EtaCore asynchronous low power ASIC and novel spiking neural nets-based artificial intelligent (AI) software. The low power approach operates in an asynchronous mode for both learning and information processing.
More desirable for edge device applications where the amount of resources (both memory and computing) are limited, asynchronous mode operation also offers a solution for protecting and securing personal information without the need for relying on the cloud for learning.
The ECM3531SF ASIC includes machine algorithms for sensor fusion of gyro, magnetometer and accelerometer sensors.
For environmental sensing applications based on chemical sensors, the EtaCore enables sensors hubs to perform more extensive sensor algorithms providing real time data and updates from mobile and Internet of Things (IoT) networked devices.