Abstract
With the ongoing diffusion of mobile computing and context-aware applications, knowledge of the current location of an individual can be leveraged in a number of different domains, from personal diaries and fitness-related applications to human behavior analysis and targeted advertising. This article presents a review of past research works describing techniques for utilizing smartphone sensors to identify the environment where a smartphone user is located. The review focuses on studies where user location can be computed autonomously and continuously by a smartphone, without the need for an active involvement of the user, and where issues such as power consumption and dependence of sensor readings from the on-body position of the phone are addressed.
TopBackground
In the last few decades there has been an increasing interest in positioning technologies. The deployment of a number of satellites in the Earth’s orbit enabled satellite-based positioning, whose main use case was vehicle navigation, but due to poor performance of this technology in indoor areas, indoor location methods have to rely on other means. The first indoor location techniques required carrying specialized devices and/or deploying ad-hoc hardware in the environment; then, the continuous enhancement of mobile phone sensing and computation capabilities, and the widespread deployment of infrastructure for wireless communication opened new frontiers for indoor localization; now, an increasing number of location-based services are made possible by different technologies for locating people in indoor environments.
Many pervasive computing applications are enabled or can be enhanced by knowledge of the current user context; while the exact definition of user context can vary between applications, physical location is an important piece of information in defining the context for many applications. Thus, methods for automatic localization of users can be considered as part of the more general issue of user context recognition (Hoseini-Tabatabaei, Gluhak, & Tafazolli, 2013).
Key Terms in this Chapter
Step Detection: A method for automatically recognizing the walking steps of a pedestrian, usually by means of an IMU mounted on a device carried by the pedestrian.
Landmark: A location characterized by specific values or patterns of sensor readings that can be used to uniquely identify the location during successive visits. Measured values can be specific to the local environment (e.g. electromagnetic signals), or related to how the location is accessed by the visitor (e.g. acceleration patterns observed when using a staircase).
Fingerprinting: A localization method where the current location is determined by measuring the local characteristics of one or more signals (usually of electromagnetic nature) and comparing them with a set of values stored in a database. The database contains a map of signal values and their corresponding physical locations.
Dead Reckoning: A localization method that consists of calculating the current location as the result of a displacement function applied to the previous location.
Inertial Measurement Unit (IMU): A sensing device that reacts to changes in its velocity (accelerometer) or orientation (gyroscope). When mounted on a rigid body, it allows measuring the acceleration or orientation of the body.
Map Matching: The technique of matching location estimations coming from sensor readings with constraints or hints coming from a map of the environment.
Simultaneous Localization And Mapping (SLAM): A technique for localizing a person or object in an area and simultaneously building a map of the area by processing sensor reading data. Usually, the map is built and adjusted incrementally during multiple visits to the area, and localization accuracy improves during the map building process.