An Indoor Scene Recognition Algorithm Based on Pressure Change Pattern

The indoor scene accurate recognition can provide valuable environmental information for location based services and context-aware applications. In this paper, we proposed a lightweight indoor scene recognition algorithm. It adopts the quick barometric pressure change to detect indoor floor transition, which is then used to judge the user is in buildings. Because pressure observation is independent of the smartphone placement, the proposed indoor scene recognition algorithm is ubiquitous and robust. Real experimental results demonstrate that the proposed algorithm can accurately determine the user is in indoor scene whenever a floor transition happens with more than 98.8% accuracy and less than 10mW power consumption.