Cost-effective filtering of unreliable proximity detection results based on BLE RSSI and IMU readings using smartphones

Katarzyna Filus,Joanna Domańska,Sławomir Nowak,Jakub Duda

doi:10.1038/s41598-022-06201-y

Katarzyna Filus, Joanna Domańska + Show 2 more

Open Access

https://doi.org/10.1038/s41598-022-06201-y

Copy DOI

Abstract

Indoor environments are a major challenge in the domain of location-based services due to the inability to use GPS. Currently, Bluetooth Low Energy has been the most commonly used technology for such services due to its low cost, low power consumption, ubiquitous availability in smartphones and the dependence of the signal strength on the distance between devices. The article proposes a system that detects the proximity between static (anchors) and moving objects, evaluates the quality of this prediction and filters out the unreliable results based on custom metrics. We define three metrics: two matrics based on RSSI and Intertial Measurement Unit (IMU) readings and one joint metric. This way the filtering is based on both, the external information (RSSI) and the internal information (IMU). To process the IMU data, we use machine learning activity recognition models (we apply feature selection and compare three models and choose the best one—Gradient Boosted Decision Trees). The proposed system is flexible and can be easily customized. The great majority of operations can be conducted directly on smartphones. The solution is easy to implement, cost-efficient and can be deployed in real-life applications (MICE industry, museums, industry).

Highlights

Indoor environments are a major challenge in the domain of location-based services due to the inability to use global positioning system (GPS)
As an answer to the problem formulated in “Problem and aim definition”, we propose a system based on Bluetooth technology and Intertial Measurement Unit (IMU) readings from smartphones for detecting proximity to a static point and evaluating the quality of this prediction
The results show that by taking into consideration only features with the coefficient value greater than 0.3, we were able to minimize the number of features used for training from 54 to 17

Summary

Introduction

Approaches based on the received signal strength allow a reasonably accurate proximity estimation, especially when beacons are individually calibrated and appropriately distributed in space This approach is considered the most practical[10] and the vast majority of indoor distance localization and estimation works developed in recent years are based on this method, with different approximation models and methods used to filter incoming frames. The domain of indoor proximity and location estimation is dominated by approaches using Bluetooth Low Energy RSSI, the Bluetooth signal in indoor environment has low stability and sensitivity to interference (many obstacles, signal reflections etc.)[11] To minimize these problems sophisticated mathematical methods (e.g. smoothing filters and wavelet filters) are used in the systems proposed in the literature, which make them impractical in situation, in which data for numerous nodes has to be analyzed. The work described was conducted as a part of the AIMeet project under the Bridge Alfa program, an initiative

Objectives

Methods

Results

Discussion

Conclusion