(Weighted) Adaptive Radius Near Neighbor Search: Evaluation for WiFi Fingerprint-based Positioning

2026-04-17

AI Summary

Fixed-radius nearest neighbor search in WiFi fingerprinting-based indoor localization is constrained by a uniform distance threshold, limiting positioning accuracy. This work proposes Adaptive Radius Nearest Neighbor (ARNN) and its weighted variant (WARNN), which for the first time integrates dynamic search radii with sample weighting to overcome the performance bottlenecks of conventional kNN methods. Systematic evaluation across 22 real-world datasets demonstrates that all three variants of WARNN consistently rank among the top four performers, significantly outperforming or matching 12 state-of-the-art kNN variants and substantially improving regression accuracy in indoor localization tasks.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Machine Learning: Learning with Manifolds
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services

Abstract

Fixed Radius Near Neighbor (FRNN) search is an alternative to the widely used k Nearest Neighbors (kNN) search. Unlike kNN, FRNN determines a label or an estimate for a test sample based on all training samples within a predefined distance. While this approach is beneficial in certain scenarios, assuming a fixed maximum distance for all training samples can decrease the accuracy of the FRNN. Therefore, in this paper we propose the Adaptive Radius Near Neighbor (ARNN) and the Weighted ARNN (WARNN), which employ adaptive distances and in latter case weights. All three methods are compared to kNN and twelve of its variants for a regression problem, namely WiFi fingerprinting indoor positioning, using 22 different datasets to provide a comprehensive analysis. While the performances of the tested FRNN and ARNN versions were amongst the worse, three of the four best methods in the test were WARNN versions, indicating that using weights together with adaptive distances achieves performance comparable or even better than kNN variants.

Problem

• Fixed Radius Near Neighbor
• WiFi fingerprinting
• indoor positioning
• adaptive radius
• nearest neighbor search

Innovation

• Adaptive Radius Near Neighbor
• Weighted ARNN
• WiFi fingerprinting
• indoor positioning
• nearest neighbor search