Multi-Head Attention Neural Network for Smartphone Invariant Indoor Localization

  title={Multi-Head Attention Neural Network for Smartphone Invariant Indoor Localization},
  author={Saideep Tiku and Danish Gufran and Sudeep Pasricha},
—Smartphones together with RSSI fingerprinting serve as an efficient approach for delivering a low-cost and high-accuracy indoor localization solution. However, a few critical challenges have prevented the wide-spread proliferation of this technology in the public domain. One such critical challenge is device heterogeneity, i.e., the variation in the RSSI signal characteristics captured across different smartphone devices. In the real-world, the smartphones or IoT devices used to capture RSSI… 

Figures and Tables from this paper


LearnLoc: A framework for smart indoor localization with embedded mobile devices
The best technique improves upon prior work, achieving indoor localization accuracy between 1-3 meters, and is implemented on real smartphones and evaluated across several diverse real-world indoor environments.
Adapting Convolutional Neural Networks for Indoor Localization with Smart Mobile Devices
This paper presents a novel approach to transform Wi-Fi signatures into images, to create a scalable fingerprinting framework based on Convolutional Neural Networks (CNNs), which is validated across several indoor environments and shows improvements over the best known prior works.
SHERPA: A Lightweight Smartphone Heterogeneity Resilient Portable Indoor Localization Framework
This work proposes a framework called SHERPA that enables efficient porting of indoor localization techniques across mobile devices, to maximize accuracy and shows that it can deliver up to 8× more accurate results as compared to state-of-the-art localization techniques for a variety of environments.
WiDeep: WiFi-based Accurate and Robust Indoor Localization System using Deep Learning
WiDeep combines a stacked denoising autoencoders deep learning model and a probabilistic framework to handle the noise in the received WiFi signal and capture the complex relationship between the WiFi APs signals heard by the mobile phone and its location.
Improved CNN-based Magnetic Indoor Positioning System using Attention Mechanism
This study proposes a CNN-based indoor magnetic-only fingerprinting system using Recurrence Plots as sequence fingerprints and approaching the localization problem from a regression perspective, showing that the architecture involving attention mechanism clearly outperforms the initial RP-based CNN, yet results in much higher prediction latency.
Ensemble based real-time indoor localization using stray WiFi signal
An ensemble based learning technique to localize a user in an apartment (located in a residential complex) at room-level granularity by training an ensemble of classifiers on a weighted averages of RSSI of WiFi signals within a time window is presented.
Handling Device Heterogeneity in Wi-Fi based Indoor Positioning Systems
  • Yongyong WeiRong Zheng
  • Computer Science
    IEEE INFOCOM 2020 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS)
  • 2020
This work proposes two novel solutions to mitigate device heterogeneity for representative localization approaches using Gaussian Process regression and neural network, respectively, which are effective and achieve higher accuracy than that of two baseline approaches in most cases.
Applying Deep Neural Network (DNN) for large-scale indoor localization using feed-forward neural network (FFNN) algorithm
A Feed-Forward Deep Neural Network (FF-DNN) Algorithm is proposed for indoor localization and received Signal Strength (RSS) is used due to minimum cost, less operational complexity, and easy usages.
SIABR: A Structured Intra-Attention Bidirectional Recurrent Deep Learning Method for Ultra-Accurate Terahertz Indoor Localization
A Structured Intra-Attention Bidirectional Recurrent (SIABR) deep learning method is proposed to solve the CSI-based three-dimensional (3D) THz indoor localization problem with significantly improved accuracy.
A Robust Indoor Positioning System Based on the Procrustes Analysis and Weighted Extreme Learning Machine
A robust and precise IPS is developed by integrating the merits of both the STI and weighted extreme learning machine (WELM) and a performance comparison with existing solutions verifies the superiority of the proposed IPS in terms of robustness to device heterogeneity.