Bluetooth Low Power (BLE) indoor location has gained increasing attention due to its low cost, low power, and ubiquitous availability in mobile devices. In this paper, a new RSSI data labeling method is studied in the offline phase of indoor positioning. In order to achieve better positioning results, two dimensional reduction algorithms such as t-SNE and PCA are used to optimize the data. Then the preprocessed data is used to test the location. Two simple but widely used localization algorithms, weighted centroid and kNN fingerprint recognition, are used to compare the influence of different regions, acquisition equipment and transmitting power on positioning accuracy. In addition, meta-heuristic based optimization techniques were used to determine the optimal TxPower configuration settings for BLE devices. The results reported in this work show an accuracy of nearly 90.7%, a 45% improvement over positioning accuracy using the original RSSI database. This also shows the feasibility of the scheme in developing more accurate indoor positioning system.
No takes yet. Share an insight, caveat, or question.
Nie et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: