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

TL;DR

This paper introduces adaptive and weighted variants of Fixed Radius Near Neighbor search, demonstrating that weighted adaptive methods can outperform traditional kNN in WiFi fingerprint-based indoor positioning.

Contribution

The paper proposes ARNN and WARNN methods that adapt distances and incorporate weights, improving upon fixed-radius approaches for positioning accuracy.

Findings

WARNN methods outperform fixed-radius variants in accuracy

Weighted adaptive methods are comparable or better than kNN variants

Comprehensive evaluation on 22 datasets supports effectiveness

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.