QuickLogic’s sensor hub is Android “KitKat” 4.4 compliant
QuickLogic is addressing “Always-on Context Awareness” – for example, your mobile phone always monitoring surroundings even when idle – with a new CSSP (Customer Specific Standard Product), writes David Manners.
It is an Android “KitKat” 4.4 compliant sensor hub, supporting all commonly used sensors in mobile devices – such as Accelerometer, Gyroscope, Magnetometer, Pressure, Ambient Light and Proximity. It includes Sensor Platform Inc’s FreeMotion library – a sensor fusion and context awareness system optimised for smartphones, tablets, and wearable devices.
Promised to be light on power consumption (less than 1% of a typical smartphone battery, claims the company), it works as part of the wider QuickLogic ArcticLink 3 S1 platform.
The company writes:
Featuring a power and processing-appropriate combination of hard logic and programmable fabric, the ArcticLink 3 S1 can feature a number of QuickLogic PSB (Proven System Block) technologies, including Sensor Manager, Flexible Fusion Engine, and Communications Manager. Due to it’s programmable fabric nature, the Ultra-Low Power Sensor Hub (ULPSH) can be run-time re-configured to run different sensor management algorithms, enabling multiple use cases for mobile devices through a single sensor hub.
Power Efficient Architecture
- Reduces power of context aware sensor hub and sensor subsystem to less than 2% of device power
- Proprietary low power I2C and smart FIFO technologies enable the lowest operating power
- Autonomous initialization and sampling of sensors eliminates
- AP load and lessens overhead
- Accurate time stamping (interrupts) improves the accuracy of sensor fusion
- Programmable sensor data filtering reduces host system processor load
Flex Fusion Engine (FFE) Reconfigurable Computational Core
- Optimized core for sensor fusion algorithms saves space and power
- Different cores available to support different OEM models
- Enables new applications and use cases after deployment
- Different use cases for different contexts (walking, running, day, night, etc.)
- Executes complex algorithms, fusing physical sensors to form virtual sensors
- Sensor data management and buffering offloaded from host processor, saving power and processing resources
- Programmable interface; SPI or I2C can be selected for best system fit and performance
- Dynamic calibration enables instant usability of magnetometer implementation applications