Cell phones might soon have the ability to predict when they’ll be plugged in and could even intelligently estimate how many calls a user is likely to make over a period of time, based on findings in a paper[pdf] by researchers at Intel and Rutgers University. They would then use that information to provide better battery life.
The devised system, dubbed CABMAN (for context-aware battery management architecture
for mobile devices*), would be based on three main principles:
- The availability of crucial applications should not be compromised by non-important ones
- Opportunities for charging should be predicted to allow devices to determine how much energy they can expect to have, instead of simply going by the battery level
- Context, such as location information, can be used to predict charging opportunities
CABMAN would predict where it can be charged by learning which towers are nearby when it’s plugged in. Then, by tracking location and processing call logs, the system knows when to alert a user to plug the cell phone in or to stop using battery intensive applications (or not do anything at all if it thinks the phone will be plugged in soon). This, for example, could know when to turn off a phone’s music player on an airplane if the phone won’t be able to make any calls soon.
A prototype was tested using data from another project. The software was on average 12 minutes away from predicting charging opportunities – a great result. The future call time prediction didn’t work as well, so more time is needed to perfect that aspect.
This research is showing us a very interesting future for cell phones, indeed. With this kind of “smarts” our future phones will be all but bulletproof, predicting our behavior perfectly and adapting themselves to suit our needs. Hopefully the wait will not be too long.
* Yes, we know that “mobile devices” doesn’t start with an “N” and have no idea who was responsible for naming this project and why this slipped through.