Available methods that make explicitly use of multipath propagation, i.e., by generating scattering models with statistics, by simultaneous target and multipath positioning, by using training signals to model a random variable together with a floor plan to enhance the tracking filter or by using large-scale MIMO do not scale well or do not improve with an increasing number of available training signals, which may often be acquired easily in practice. But they only work well if there is still a strong line-of-sight (LoS) signal. This, however, requires a measurement campaign to capture the propagation effects on which we then need to solve a highly non-linear optimization problem.Īpproaches that explicitly address multipath propagation include (unscented) Kalman filters, channel classification, subsample interpolation, sub-space approaches, or cone constraints. In practice, we often use much more infrastructure nodes than we would actually need for unambiguous position estimation (which results in an over-determined system) and try to install them at positions where those unwanted effects are minimal. This leads to a erroneous position estimation. These make the ToA-estimation non-linearly distorted, and if the effects are strong it increasingly becomes difficult to estimate the correct ToA from the radio burst signal. The omnipresence of metallic surfaces, i.e., machinery, pipes, vehicles, racks etc., in industrial environments leads to effects such as signal reflections (multipath propagation), scattering, obstruction, shadowing, and attenuation. However, achieving the application requirements through TDoA-based positioning may also become difficult in practice. Nevertheless, their better positioning accuracy makes them still attractive for many use-cases, including the tracking of goods in warehouses or virtual and augmented reality in various applications. While RSS- and AoA-based localization usually come at a low cost but also lower accuracy, T(D)oA-based systems require synchronization schemes and have a more complex system setup, which usually makes them more expensive. Under the hood, RF-based positioning systems may be implemented using a multitude of different technologies, which include angle of arrival (AoA), received signal strength (RSS), time of arrival (ToA), and time difference of arrival (TDoA). As localization is also discussed for standardization in 5G, we can expect to see RF-based tracking becoming more and more ubiquitous especially in GPS-denied areas such as indoor environments. In contrast to highly accurate vision-based tracking systems, which often also raise privacy concerns and are prone to dirt and weather conditions, they guarantee a robust position tracking traded for a lower accuracy. Radio-based real-time locating systems (RLTSs) are key to drive automation and digitalization in many applications in warehouse management, production, and manufacturing.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |