Enabling intelligent systems through motion

May 29th, 2008

1The possibilities for MEMS sensors are impressive, and their uses are getting much more sophisticated as devices improve. Greg shares his perspective on how a typical handheld device using MEMS sensors might change over time.

Sleep-enabled power-saving accelerometers

The tremendous success of Nintendo’s Wii and Sony’s PlayStation 3 games with motion-based controllers has brought motion awareness front and center in new product designs. Although motion-based game controllers are an obvious application, motion awareness can significantly improve product performance, usability, and functionality in many other ways.

One of the next issues motion awareness is tackling is power management in a mobile handset. As mobile handsets begin to more widely adopt motion-based features, power consumption is becoming a critical concern. To date, most applications, such as pedometer features or tilt-based user interfaces, have relied on the baseband processor to interpret motion information supplied by an accelerometer and perform the necessary calculations to implement the feature. Unfortunately, the baseband processor consumes significant power.

Mobile handset designers are looking to offload the motion-based feature implementation onto a dedicated low-power microprocessor or into the accelerometer itself. Most accelerometers are available with a power management option in which the host can direct the accelerometer to go into a very low-power sleep mode when not in use, further reducing the power associated with motion-based features.

While these options are useful, they only represent the beginning of what can be done with motion awareness. Once again, consider the mobile handset. Making the handset aware of its motion can enhance power management while also improving the product’s usability. That’s where accelerometers equipped with a more sophisticated "sleep" mode come in.

When these sleep-enabled accelerometers are put in very low-power sleep mode, they periodically wake up and see if a motion event occurred since their last awakening. If movement happened, the accelerometer goes into its normal operating state and provides an interrupt to its host indicating that a motion event occurred, enabling the host to return to its normal operating mode. If no motion occurs, the accelerometer returns to its sleep mode until the next sampling interval.

Using this type of intelligent accelerometer, handset designers can implement some interesting power management and usability features. When a user places the handset on a table and does not move it for a period of time, the handset can put itself in a low-power mode until either a call is received or the handset is picked up and moved.

With motion awareness, designers can make this feature even more intelligent. For example, a handset placed face down on a table can be interpreted to mean "I don’t want to be disturbed." The handset will switch to low-power mode and without ringing, go straight to voicemail if a call is received. A face-up placement on a table can be interpreted to mean "I am not using you now, but go ahead and ring if a call comes in." If the handset rings and the user flips it over onto its face, the handset can interpret the motion as "I don’t want to answer the call; please go to voicemail and stop ringing."

In addition to using the position of the handset, such as face up or face down, designers are exploring motion-based features enabled by tapping the handset. For example, imagine the handset is in a user’s pocket and rings. The user is not interested in taking the call at the time so he or she taps the handset, which is interpreted by the device to mean "Stop ringing and go to voicemail." Some accelerometers available now come with interrupts for single-tap and double-tap motions, enabling designers to implement multiple tap-based features.

Smarter portable electronic products

Mobile handsets present an interesting case study in how motion awareness can bring useful new features to a portable electronic product. They also illustrate the system partitioning choices that designers must make when determining where to implement different features to minimize power consumption. Numerous other portable electronic products, such as PDAs, MP3s, and personal navigation devices can similarly benefit from motion-based features.

Accelerometers are now available in very small packages, such as 3 mm x 3 mm x 0.9 mm, with a variety of features and at low cost. Such accelerometers are enabling new product designs that exhibit a high level of intelligence based on motion awareness, resulting in improved usability and enhanced performance. As designers become more comfortable with motion-based features and users begin to expect such features, motion awareness will likely spread rapidly to an ever-wider array of portable electronic products.

Greg Galvin is president and CEO of Kionix, based in Ithaca, New York. He founded Kionix in 1993, and was previously director of corporate research relations at Cornell University. He has a BSEE from the California Institute of Technology as well as an MBA and a PhD in Materials Science from Cornell.

Kionix, Inc.
607-257-1080
info@kionix.com
www.kionix.com

The new shape of industrial computing, networking, and sensing
©MMXIIIndustrial Embedded Systems.
An OpenSystems Media publication.