An example of geolocation in cellular network © Péter
An example of geolocation in cellular network © Péter

Location-based service is used in a great variety of contexts, such as search and rescue operations, navigation and tracking, and location-based mobile advertising. The position information, as a key component of location-based services, needs to be acquired from different positioning systems.

One class of positioning systems is the global satellite navigation system (GNSS), which includes the global positioning system from US (GPS), Russia (GLONASS), Europe (Galileo) and China (Beidou), which are particularly suitable for outdoor positioning.

The other class of positioning systems, named as local positioning systems, are required to extend the reach of GNSS to provide position information in areas where GNSS signal cannot penetrate or in order to enhance the positioning precision of the GNSS. Similar to the satellites used in GNSS, beacons with given positions, such as base stations in cellular networks or WiFi access points in WiFi networks, are used as reference positions in local positioning systems.

The target’s position can be inferred from the position-related measurements, including angle-of-arrival (AOA), received-signal-strength (RSS), time-of-arrival (TOA) and time-difference-of-arrival (TDOA). For instance, TOA is defined as the propagation time between a beacon and a target, which is related to the relative distance between them and subsequently related to their positions.

For more information on Geolocation see the sections below or contact the respective research associates.


An localization application on smartphone ©
An localization application on smartphone ©

The position-related measurements are contaminated due to multipath propagation and non-line-of-sight (NLOS) propagation, i.e., blockage of the direct path between sender and receiver. These measurements can be characterized using a statistical measurement model, where an additive measurement error accounts for these undesirable effects.

Undoubtedly, knowledge about the measurement error statistics is required for an accurate localization result. These error statistics, however, are difficult to obtain in many applications. Offline calibration requiring extensive manual effort can be conducted to deliver information on measurement error statistics. However, environment change, such as people and furniture movement, will make the previously estimated measurement error statistics invalid.

In this project, we assume that no or only partial empirical knowledge about the measurement error statistics is available and investigate how accurate localization in such cases can be obtained.

The Signal Processing Group is collaborating with the Division of Automatic Control in Linköping University, Linköping, Sweden.

For more information on this research project contact Di Jin.