SUMMARYThis paper proposes a method for using an accelerometer, microphone, and GPS in a mobile phone to recognize the movement of the user. Past attempts at identifying the movement associated with riding on a bicycle, train, bus or car and common human movements like standing still, walking or running have had problems with poor accuracy due to factors such as sudden changes in vibration or times when the vibrations resembled those for other types of movement. Moreover, previous methods have had problems with has the problem of high power consumption because of the sensor processing load. The proposed method aims to avoid these problems by estimating the reliability of the inference result, and by combining two inference modes to decrease the power consumption. Field trials demonstrate that our method achieves 90% or better average accuracy for the seven types of movement listed above. Shaka's power saving functionality enables us to extend the battery life of a mobile phone to over 100 hours while our estimation algorithm is running in the background. Furthermore, this paper uses experimental results to show the trade-off between accuracy and latency when estimating user activity.

No abstract

Abstract. In a complex indoor environment such as a huge station in an urban area, sometimes the direction and distance relative to another person are more important for pedestrians than their absolute positions, e.g. to search for a lost child. We define this information as the position relation. Our goal is to develop a position relation estimation method on a mobile phone with built-in motion sensors. In literature, methods of cooperative navigation using two pedestrians' positions estimated by pedestrian dead reckoning and a range sensor have been proposed. However, these methods cannot be applied to a mobile phone because pedestrian dead reckoning does not work well when a mobile phone is in a bag, and because there is no range sensor in a phone. In fact, no Bluetooth is reliable as a substitute range sensor. This paper proposes another approach to estimate the position relation of pedestrians. Our method finds the timing when two pedestrians are in close proximity to each other and walk together by using Bluetooth as a proximity sensor and corrects the parameters of position updates dynamically, even if absolute positions are unknown. The algorithm and evaluation results are presented in this paper.

This paper proposes a method for estimating the movement of a user using an acceleration sensor called UME (User Movement Estimator). By considering time-axis changes in the power spectrum, the method is able to avoid performance loss due to factors such as sudden vibration or times when the power spectrum resembles different types of movement. Testing results show that the method has significantly better performance than past methods.

No abstract