Lattice sensAI stack optimized for always-on, on-device AI

Lattice sensAI stack optimized for always-on, on-device AI

New Products |
Building on the ultra-low power (1mW to 1W) focus of its sensAI stack, Lattice Semiconductor is releasing new IP cores, reference designs, demos and hardware development kits that provide scalable performance and power for always-on, on-device artificial intelligence (AI) applications.
By Rich Pell


Examples of the architectural choices that sensAI solutions enable include: a stand-alone iCE40 UltraPlus / ECP5 FPGA based always-on, integrated solutions, with latency, security, and form factor benefits. Solutions using the iCE40 UltraPlus as an always-on processor can detect key-phrases or objects to wake-up a high performance AP SoC / ASIC for further analytics only when required, reducing overall system power consumption.

The scalable performance/power benefits of the ECP5 for neural network acceleration could be leveraged along with its IO flexibility to seamlessly interface to on-board legacy devices including sensors and low-end MCUs for system control.

Updates to the sensAI stack include a new CNN Compact Accelerator IP core for improved accuracy on iCE40 UltraPlus FPGA, and an enhanced CNN Accelerator IP core for improved performance on ECP5 FPGAs. The neural network compiler tool has been updated too with improved ease-of-use and both Caffe and TensorFlow support for iCE40 UltraPlus FPGAs.

The company has put together several reference designs including new human presence detection and hand gesture recognition reference designs and demos. The new iCE40 UltraPlus development platforms include the Himax HM01B0 UPduino Shield, and the DPControl iCEVision Board.

Lattice Semiconductor –

Linked Articles