A New Set of Wi-Fi Dynamic Line-Based Localization Algorithms for Indoor Environments

Localization is of great importance in several fields such as healthcare and security. To achieve localization, GPS technologies are common for outdoor localization but are insufficient for indoor localization. This is because the accuracy and precision of the users’ indoor locations are influenced by many factors (e.g., multi-path signal propagations and signal fading). As a result, the methodologies and technologies for indoor localization services need continuous refinement. Another challenge is the time complexity of the methodologies which impacts the performance of mobile phones’ limited resources. To address these challenges, a new set of fingerprinting algorithms called Fingerprinting Line-Based Nearest Neighbour is designed and implemented. The new set shifts some points such grid points possibly towards target nodes based on a deterministic range of percentages. Furthermore, the new algorithms run in 0(j * d) (i.e., the term y refers to the number of grid points and the term d refers to the number of Received Signal Strength Indicators collected at the online stage). This paper presents the following: 1) the accuracy of the new set of algorithms has a theoretical upper bound in terms of the distance errors, 2) the new set of algorithms (e.g., 90% shifting percentage) improves the accuracy compared to those for several existing Nearest Neighbour (NN)-based algorithms such as the Dual-Scanned Fingerprint Localization and the Soft-Range-Limited KNN (SRL-KNN) by 106% and 68%, respectively, 3) the algorithm improves the precision compared to those for the NN-based algorithms such as the Path-Loss-Based Fingerprint Localization and the SRL-KNN 76% and 73%, respectively, and 4) the new set has lower probabilities related to the position errors compared to those for the existing NN-based algorithms. Therefore, the new set of algorithms is considered reliable and efficient for indoor location services.