91³Ô¹ÏÍø

Always On and Data Fusion

Our interaction with devices in the home, at the workplace and in public is changing. In the past we used device interfaces like a keyboard, mouse, or touch screen. Now the trend is changing, with devices being developed to interact with humans via a human interface including voice, motion and vision. These human-machine interfaces are computationally intensive and use large, power-hungry processors, which greatly reduces battery life. To mitigate the power consumption of these larger processors, smaller, ¡®wake-up¡¯ cores are used to detect multiple sensor interface inputs and identify if a human interaction is about to begin, at which time it wakes up the bigger computation core.

The smaller wake up cores can also perform sensor fusion tasks by receiving input from multiple sensors and processing them. These processors require high computation capability to process the sensor input, while consuming minimal area and power for battery-operated devices.

Always-On Function

Application

Synopsys Processor 91³Ô¹ÏÍø

Speech Detection

Home automation assistant

EM9D/11D

Voice Recognition

Smart home, sensors

EM9D/11D

Motion Detection

User interface

EM9D/11D

Gesture recognition

User interface

EM9D/11D

Face detection

Smart home, door bells

EM9D/11D
Article: Flexible Processing for IoT End Node Devices

Context aware recognition

Mobile devices

EM9D/11D