Always-on, low-power biometrics partnership formed by AIStorm and Tower Semiconductor
AIStorm and Tower Semiconductor have announced a new partnership focusing on the development of a new always-on edge imager technology.
Always-on imagers, which draw negligible power until an event like the presentation of a fingerprint or face biometric, a wake word, or a specific action is detected, can perform machine learning operations on the device after wake-up to eliminate the need for additional components, according to the company announcement.
As part of the collaboration, AIStorm’s new AI-in-imager products will incorporate AIStorm’s electron multiplication architecture and Tower’s Hi-K VIA capacitor memory, instead of digital calculations.
This will allow the components to perform artificial intelligence (AI) computation at the pixel level, as well as help manufacturers save on the silicon real estate, multiple die packaging costs, and power required by competing digital systems.
The Hi-K VIA capacitors will be built in the metal layers, to allow the AI to operate directly from the pixel matrix and retain optimal performance on both pixel density and size.
“This new imager technology opens up a whole new avenue of always-on functionality,” said Dr. Avi Strum, SVP of the sensors and displays business unit at Tower Semiconductor.
In traditional imager solutions, AI processors are usually placed outside the pixel matrix, causing always-on imaging solutions to continuously have to detect pixel changes and forward digital information to memory and an AI subsystem located outside the imager. This, in turn, results in numerous false alerts and high power consumption.
In AIStorm’s solution, however, electrons are multiplied directly instead of being converted to a digital number, thanks to Tower Semiconductor’s low-leakage VIA capacitor technology.
This capability for local AI pixel coupling introduces a new dimension to edge imaging, enabling an immediate, intelligent AI response to pixel changes for the first time.
“Instead of periodically taking a picture and interfacing with an external AI processor through complex digitization, transport, and memory schemes, AIStorm’s pixel matrix is itself the processor & memory,” Strum explained. “No other technology can do that.”
The edge imager market is worth $7 billion annually, the companies say.