TAIPEI--With engineers in short supply, and with the most experienced Internet of Things (IoT) designers commanding top salaries when they can be found, Freescale has attempted to lower the expertise bar for developing context-aware applications by offering an open-source toolkit called its intelligent sensing framework (ISF) at Computex. The Xtrinsic ISF makes it easier for designers to differentiate IoT devices by simplifying the backend work of collecting and fusing sensor data for context-aware use cases.
"Sensor fusion remains important, but is no longer enough alone is not enough anymore for context-aware connected devices," said Babak Taheri, vice president and general manager of Freescale’s sensor and actuator solutions division. "Developers need an open platform that enables data fusion and facilitates the creation of differentiated, context-aware devices that drive the Internet of Things."
IHS iSuppli tracks the micro-electro-mechanical system (MEMS) sensors used to provide context-awareness for IoT devices, predicting that the market will experience double-digit growth through 2016 when the total sensor market will top $2.2 billion. The mainstream sensors uses for IoT devices include accelerometers (for orientation), gyroscopes (for fine movements), barometric pressure (for altimeters), magnetometers (for heading) and global positioning system (GPS, for location). However, designers cannot just read-out raw data, but must fuse the outputs of multiple sensors simultaneously, in order to use them to cross-check each other, thereby providing a reliable context for IoT command- and menu-options.
The Xtrinsic intelligent sensing framework (ISF) not only eases these integration tasks--speeding development with customizable libraries of common sensor fusion tasks--but also aims to boost product differentiation for IoT applications by allowing designers to "see the forest" as a whole instead of just thousands of "tree" data points streaming in from sensors. Specific use cases addressed by the ISF include industrial systems, medical equipment, consumer devices and any IoT device leveraging realtime sensor fusion.
Freescale's developers toolkit called the Xtrinsic intelligent sensing framework (ISF) is an open-source' resource for leveraging sensor fusion in context-aware Internet of Things (IoT) applications.
All sensor data is integrated into a single dashboard-like view which provides a unified methodology for obtaining, calibrating, scaling and displaying sensor data. Combined sensor views also simplify applying sensor fusion algorithms to serve specific context-aware domains and use cases.
Pre-built adapters handle all the communications details for collecting sensor data and built-in power-management, host proxy and command interpreter functions as well as simplify the coding effort especially for novel sensor arrays mixing sensors from various manufacturers.
An easy way to get a leg up, for newbies and veterans alike, is to design around smart sensor hubs like Freescale's FXLC95000CL, which combines a three-axis accelerometer with a ColdFire 32-bit microcontroller in the same package running its own realtime operating system (MQX). Such sensor hubs can add-on any number of auxiliary sensors to serve specific applications, but confine all the sensor fusion and context-aware algorithms to the ColdFire processor, thus offloading the IoT device's application processor from these tedious realtime tasks.