Products

Neural Sensor Platform is a Complete Solution for Micropower Artificial Intelligence at the Sensor Endpoint

ECM3531

ASIC based on the ARM Cortex-M3 implemented with Eta Compute’s patented continuous voltage and frequency scaling technology (CVFS) for dramatically lower power consumption. The processor includes a 16b Dual MAC DSP processor and Eta Compute’s high efficiency Analog to Digital Converter (ADC) sensor interface and highly efficient PMIC circuits. The chip also includes I2C, I2S, GPIOs, RTC, PWM, POR, BOD, 128kB SRAM, 64kB SRAM dedicated to the DSP and 512kB on chip FLASH.

Documentation

ECM3532

Standalone AI platform includes a multicore processor, that includes flash memory, SRAM, I/O, peripherals and a machine learning software development platform. The patented CVFS substantially increases performance and efficiency for edge devices. The self-timed CVFS architecture allows to automatically and continuously adjusts internal clock rate and supply voltage to maximize energy efficiency for the given workload. The ECM3532 multicore NSP combines an MCU and a DSP, both with CVFS, to optimize execution for the best efficiency making it an ideal solution for IoT sensor nodes.

Documentation

Neural Network API

Neural net library that supports CNN, RNN, LSTM, GRU, and others. Also included are many ready-to-deploy reference designs to accelerate customer product developments of algorithms for fusion of gyro, magnetometer, and accelerometer sensors as well as image sensors and others. Customers can write code entirely in their preferred programming environment and target execution of the machine learning functions to either the DSP or the M3. This allows a further power optimization.

Software Development Kit

Software Development Kit includes direct programming access for custom and proprietary application development with support for popular machine learning training frameworks such as TensorFlow. The SDK allows for simple or automatic control of the power management functions enabling significant power reduction when the application workload decreases for maximum product efficiency which extends battery life and lowers cost.

energy harvesting smart sensor board
fingerprint smart sensor board