Always-On Sensing Changes EverythingOctober 11, 2013 By: Tom Kevan
Sensors have long played key roles on mobile platforms, but a fundamental improvement in their operating parameters suggests big changes are afoot. Last month, a number of new mobile phones entered the market, distinguished by their always-on sensor functionality. In the past, these sensors were turned off periodically to avoid draining precious battery power and to free up the main processor as much as possible. Now, with the introduction of additional technology, that's all changing, and this shift promises to usher in a raft of new features and applications.
The catalyst for this change is the introduction of power-efficient co-processors that offload work from the device's main processor. The chips, aided by specialized firmware, aggregate and analyze data from a full spectrum of sensors, ranging from accelerometers, gyroscopes, and magnetometers to voice and pressure sensors. This allows the sensors to stay active all the time, even when the phone is asleep.
One of the leading examples of this development is the iPhone 5S. Here Apple has added the M7 co-processor, paired with the CoreMotion API. Together, these technology components offload sensor-hub functions from the application processor, reduce overall power consumption, and enable data from motion sensors to be used by any application that requests it.
Another example is Motorola's Moto X cell phone. In this case, Motorola has added a low-power processor devoted entirely to speech recognition, which also allows the sensors to function continuously.
As these two products featuring always-on sensors entered the market, two other technology providers—STMicroelectronics and Movea—joined forces to provide another microcontroller-firmware combination to enable always-on sensor functionality. This joint effort integrates Movea's SmartMotion technology with STMicroelectronics'STM32F401 low-power sensor-hub microcontroller.
Movea's firmware provides tools and motion-processing models that together gather data from sensors and enable the STM32F401 platform to perform advanced motion functions. In addition, the software augments the sensors with auto calibration and 3D orientation, supporting gesture and activity monitoring.
A cursory look at the developments may prompt the question: What's the big deal? When in fact, the combination of low-power intelligence and continuous sensory input processing promises to usher in a new class of user interfaces and advanced health and fitness tracking, contextual awareness, and pedestrian navigation services.
As the MIT Technology Review article "What Apple's M7 Motion-Sensing Chip Could Do" points out, with always-on sensory input, a mobile user interface could use simple gestures or voice commands to launch applications or changing settings without having to unlock the phone. And an always-on microphone could determine when a noisy environment merits making the ringtone louder.
Add physiological sensors to the mix, and continuous real-time monitoring could provide insight into an individual's health, fitness, and well-being. According to the HIT Consultant article "How Sensors Are Transforming the Life Sciences Industry Landscape," this kind of data could enable early detection or even prediction of changes in health conditions such as heart disease, epilepsy, Parkinson's disease, stroke, dementia, and cancer.
Finally, devices equipped with motion sensors and GPS devices could provide additional services based on the user's activity, location, and surroundings. For example, they might enable advanced indoor pedestrian navigation and a variety of location-based services.
The introduction of power-efficient co-processors that support continuous sensory input portend the emergence of a new level of mobile "ambient intelligence." Because of the ubiquity of mobile devices, this development provides a glimpse of the benefits promised by the Internet of Things.
ABOUT THE AUTHOR
Tom Kevan is a New Hampshire-based freelance writer specializing in technology.
Most Read Articles