Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Human centric situational awareness
Context awareness is an approach that has been receiving increasing focus in the past years. A context aware device can understand surrounding conditions and adapt its behavior accordingly to meet user demands. Mobile handheld devices offer a motivating platform for context aware applications as a r...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
AUC Knowledge Fountain
2015
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| Summary: | Context awareness is an approach that has been receiving increasing focus in the past years. A context aware device can understand surrounding conditions and adapt its behavior accordingly to meet user demands. Mobile handheld devices offer a motivating platform for context aware applications as a result of their rapidly growing set of features and sensing abilities. This research aims at building a situational awareness model that utilizes multimodal sensor data provided through the various sensing capabilities available on a wide range of current handheld smart phones. The model will make use of seven different virtual and physical sensors commonly available on mobile devices, to gather a large set of parameters that identify the occurrence of a situation for one of five predefined context scenarios: In meeting, Driving, in party, In Theatre and Sleeping. As means of gathering the wisdom of the crowd and in an effort to reach a habitat sensitive awareness model, a survey was conducted to understand the user perception of each context situation. The data collected was used to build the inference engine of a prototype context awareness system utilizing context weights introduced in [39] and the confidence metric in [26] with some variation as a means for reasoning. The developed prototype's results were benchmarked against two existing context awareness platforms Darwin Phones [17] and Smart Profile [11], where the prototype was able to acquire 5% and 7.6% higher accuracy levels than the two systems respectively while performing tasks of higher complexity. The detailed results and evaluation are highlighted further in section 6.4. |
|---|