Skip to Main Content
One of the main objectives of localization algorithms is to compute accurate estimates of sensor positions. This task is usually performed using measurements exchanged with neighboring sensors. However, when erroneous measurements occur, the localization process may yield wrong estimates, which leads to unreliable information for location-based applications. This paper proposes a robust localization technique that works efficiently, even under unreliable measurements assumptions. The proposed method uses belief function theory to estimate sensors locations. Assuming that the reliability of sensors measurements is known, the method combines all the available information to make a final decision about the positions. Each measurement is then used to define a belief function based on the reliability information. Experiments with simulated data demonstrate the effectiveness of this approach compared with state-of-the-art methods using different combination rules.
Date of Publication: May 2011