Design engineers continue to push the low-power envelope, particularly in mobile devices. Low power was a common theme at Sensors Expo 2014, held for the final time at the Donald E. Stephens Convention Center in Rosemont, IL from June 24th to the 26th.
The show is headed to the west coast in 2015, and while I’m a fan of Long Beach, I’ll miss the easy morning drive down from my office in Madison, WI. I’ll take a long drive from my own bed well before I opt for an artificially cooled sterile box overlooking a parking lot.
Back to Rosemont, where I had a chance to speak with Ashis Bhattacharya, Honeywell’s vice president of global strategic marketing and business development about lower power in smaller spaces. “In mobile devices, everything starts at the sensor,” said Bhattacharya. “Engineers need 500 nA or less. Battery life is huge and we need to keep lowering power consumption.” This, of course, was a convenient lead in to Honeywell’s new magnetoresistive sensor ICs, which only draw 310 nA of current, which means that if you attached a AA battery, it would last for 300 years, give or take. The new IC series stood out at the show, and even made it into ECN’s Top Ten from Sensors Expo 2014 (www.ecnmag.com/sensors-10).
“When Google is going to pay 10¢ per part, how do you differentiate yourself?” asked Susan Adams, vice president of marketing and sales for PNI Sensor Corporation. PNI’s low-power answer for power-hungry sensors in mobile devices is the company’s SENtral motion coprocessor, particularly the SENtral-K sensor hub designed to support all Android 4.4 KitKat sensors. Part of KitKat includes Google’s requirement for standardized sensor outputs as part of a hardware sensor batching initiative (see also: sensor fusion) which could dramatically reduce the amount of power consumed by ongoing sensor activities.
According to the Android developer site, “With sensor batching, Android works with the device hardware to collect and deliver sensor events efficiently in batches, rather than as they are detected. This lets the device’s application processor remain in a low-power idle state until batches are delivered.” In addition to its processing, PNI’s SENtral-K does the math on-chip with additional RAM space for further customization — all while consuming about one percent of the power.
Sensor fusion is the logical next step as we move into an era of absolute connectivity. We already have the sensor hardware in place, but the fusion takes place within the software, which combines the data from each sensor to improve system and/or application performance. According to Kionix, combining this data will correct for the deficiencies of the individual sensors to calculate accurate position and orientation information. For a sample, check out Google Now. If you have an Android, chances are that it is already on your phone — and the app already knows many of your tendencies. When I wake up in the morning, Google Now knows my location and provides route information to the office; it knows that I check the weather and the news so that information is at the top of the page; and it provides local parking and nearby event information — all with no pre-specified user input. I’ve been fused. What I found particularly unnerving were the route suggestions to frequently visited businesses and locations.
The technology is leading the mobile device industry towards more accurate outdoor/indoor location with increasing context- and location-aware systems — all of which is fodder for a future column as we transition culturally from fear of Big Brother intrusions, to the absolute elimination of privacy, all in exchange for ease-of-use functionality.
For more coverage from Sensors Expo 2014, visit ecnmag.com.