Relative Localization of Multinode Asynchronous Decawave Ultrawideband Systems

This article aims to improve the capacity of asynchronous ultrawideband (UWB) systems to accommodate more nodes and to enhance the computational efficiency of relative positioning among these nodes. First, a new type of node is introduced that generates passive measurements by listening to round-trip communications of other active nodes. Due to their passive nature, increasing the number of these nodes does not affect the refresh rate. Then, we devise a two-stage relative positioning algorithm. In the first stage, a pseudolinear solution is derived to position active nodes, which involves reparameterizing nonlinear equations into linear ones through squaring operations and introducing auxiliary variables. This linearization enables efficient attainment of optimal solutions without requiring initial values. In the second stage, the estimated active node positions and a derived pseudolinear-iterative solution are used to locate passive nodes, where the pseudolinear solution derived based on the passive measurements is iteratively refined. Simulations show that the proposed two-stage relative positioning algorithm is superior to traditional methods in terms of computational efficiency and reaches the Cramer-Rao lower bound (CRLB) within the considered noise level range. Furthermore, a robust exchange mechanism is designed to address the data coupling when the traditional polling mechanism collects measurements, thereby increasing the probability of acquiring intact measurements. The real-world experiment validates this mechanism and the proposed positioning method. Finally, other approaches to realize relative positioning based on UWB are briefly discussed.