Cooperative UWB-Based Positioning Systems: CDAP Algorithm and Experimental Results

Abstract

The need for accurate positioning has attracted significant interest in recent years, especially in cluttered environments where signals from satellite navigation systems are not reliable. Positioning systems based on ultrawide bandwidth (UWB) technology have been considered due to the property of UWB signals to resolve multipath and penetrate obstacles. However, range-based localization techniques typically lack accuracy in dense cluttered environments, due to line-of-sight blockage and excess propagation delay through material. In this paper, we consider positioning in a real indoor environment scenario, where anchor nodes are deployed in known locations, and one or more target nodes wish to determine the own position. A range measurement error models based on measured data from real ranging devices is considered and we propose a multilateration cooperative algorithm that, without the need of measurements between targets, exploit the estimated position of cooperative nodes to increase localization performance. We examine the case where multiple targets are present in the same environment and how the position accuracy is affected by the cooperative nodes positions and by the availability of priori information about the environment and network geometry. Using numerical results based on experimental data, we demonstrate the impact of cooperation on the positioning accuracy. It is shown that, depending on the geometric configuration of the nodes, cooperation is not always advantageous.