Scalable ad-hoc UWB network adjustment

UWB-based ranging offers alternative positioning technique in GPS/GNSS-compromised or-denied environments. The UWB positioning relies on the set of static UWB nodes with known coordinates, called UWB network. The coordinates of UWB nodes for an ad-hoc network cannot be measured with alternative technique, such as total station, due to the short time window of the deployment. This is the case in many applications, such as navigating firefighters in a burning building or supporting indoor military operations. Thus, the UWB network has to calculate the node coordinates by itself using the mutually measured ranges among the nodes. Common solution is to use a central unit that collects all observations from the nodes, and then, it solves the network adjustment. Note that this solution is not scalable. For this reason, in this study, we assume that the network nodes share the measured ranges and other local information with their immediate neighbors, and this way, the need for a central unit is eliminated. Two distributed algorithms, namely the consensus subgradient and the accelerated weighted gradient methods, are investigated to solve the network adjustment problem. The algorithms are tested on three network configuration using simulations. We found that the accelerated weighted gradient methods outperforms the consensus subgradient methods in terms of convergence on the network adjustment problem, if global information of the network structure is available.