Find research datasets worth reusing

Dataset of "Comparison of Localization Methods for Internet of Things in 5G Cellular Networks: A Wide-scale Assessment"

<p>As the 3rd generation partnership project (3GPP) organization pushes out new releases,<br>positioning in heterogeneous mobile networks enables the achievement of the accuracy required<br>in the majority of industrial applications without dependence on global navigation<br>satellite systems (GNSS). This study presents the results gathered during an extensive measurement<br>campaign related to the practical applicability of localization in next-generation<br>heterogeneous networks. We present an accuracy comparison of basic timing advance (TA)<br>localization with the k-nearest neighbor (KNN), decision tree-based random forest (RF),<br>extreme gradient boosting (XGBoost), and long short-term memory (LSTM) recurrent neural<br>network. Our results demonstrate that TA cannot be considered an optimal solution<br>from the perspective of localization accuracy because the error roughly corresponds to the<br>average separation distance from the base station (BS) to the end device (ED). In addition,<br>we found that the LSTM approach is not optimal for the outdoor localization of moving<br>ED because of the combination of multiple factors, with sparse deployment being the most<br>important. The median value of the location error of the LSTM was more than 200m higher<br>than that of the TA for the self-validation dataset. However, a simple KNN regression shows<br>solid results for 5G New Radio (NR) operating in the non-standalone (NSA) mode. KNN<br>provided the most accurate results of all methods, with median error values of approximately<br>12 (k=3) and 82 (k=5) m for the self-validated and cross-validated datasets, respectively.</p>