Putting Your Calls Into Context

Posted by: Gremelin

Putting Your Calls Into Context - 09/27/03 07:46 PM

Incessant calling and voicemails might become a thing of the past.

In Pittsburgh, a research team at Carnegie Mellon University's Institute of Technology, or CIT, has developed a new context-aware mobile-phone technology called the SenSay. The SenSay cellular phone, still in prototype stage, keeps tabs on e-mails sent, phone calls made and the user's location. The phone also adapts to the user's environment.

"SenSay is a huge productivity boost," said Dr. Asim Smailagic, a senior researcher at Carnegie Mellon's Institute for Complex Engineered Systems. "Because people can see when you are available, the time it takes to hand off or receive information is greatly reduced."

In addition to automatically manipulating ringer volume, vibration and phone alerts, SenSay (PDF) can give callers the capability to communicate the urgency of their calls, Smailagic said.

For instance, if the phone's location system indicates the user is in a conference room and the to-do list shows a conference at 11 a.m., the ringer would be automatically disabled. When a call comes in, SenSay will combine the information from the location-sensing system, schedule and to-do list, then make a decision to route the call directly into voicemail.

An emergency call would still be sent to voicemail, but once the system recognizes the word "urgent," SenSay can send a text message asking the caller to call back, then vibrate the SenSay and send a message to alert the user to expect an important call in three minutes.

To provide data about the user, SenSay uses motion sensors (accelerometers), a microphone, a heat-flux sensor (to measure the heat coming from the user's body) and galvanic skin-response sensors. These sensors are housed in a light, stretchy wireless armband. A GPS device helps to determine the user's position, both outdoors and inside a building.

In a demonstration at the CIT lab, the results of the sensor's learning process are monitored on a laptop screen. A white dot on the screen represents Maria, a student, and shows what the system thinks she is doing. The dot moves from a region labeled "normal" to one labeled "walking," then "jogging," "running" and "driving" as she acts out each activity. If Maria entered a nightclub, the microphone would detect the loud music, the light sensor would indicate low light conditions, and the heat-flux sensor could recognize whether she were dancing. "This is pretty revolutionary stuff," Smailagic said.

The system won't be available commercially for at least two years and will cost around $200. Already Intel, which helped fund the research, has expressed interest in becoming the group's manufacturing partner. The military is also interested in the technology.

Hurdles remain. The system needs to be integrated into one piece, and its storage and computation space must be increased, since the more people SenSay users allow to access their personal data, the more tasks the system must juggle. Another drawback is that callers can receive text prompts only if they own a phone with a text screen.

Read the full story here .