This blog contains the papers I have read for my summer research: Applying machine learning to indoor fingerprinting positioning, especially using deep learning.

开篇入门(chinese version):

Please skip this part if you don’t speak chinese.
机器学习与室内定位技术

Overview of indoor positioning

  1. A Survey of Indoor Localization Systems and Technologies
  2. Overview of indoor positioning system technologies
  3. Indoor Fingerprint Positioning Based on Wi-Fi: An Overview

Dataset used

Crowdsourced WiFi database and benchmark software for indoor positioning

Long-Trem WiFi Fingerprinting Dataset for Reserach on Robust Indoor Poitioning

Papers:

deep learning

  1. Low-effort place recognition with WiFi fingerprints using deep learning
    https://scholar.google.com/scholar?oi=bibs&hl=en&cites=8184808048122111760

    This article use deep neural networks and Autoencoders to do the floor Classification, but no positioning prediction. I have done the positioning part, get a good result.
    code of this paper: github

  2. A Deep Learning Approach to FingerprintingIndoor Localization Solutions
    https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8215428

    It use two methods to solve the small dataset problem. one is using data augmentation. The sequence of the APs will change(very doubt at this method) ; Anther method is to use transfer learning, only similar dataset can help.
    github:https://github.com/MaiziXiao/IndoorLocalization

  3. Large-Scale Location-Aware Services in Access: Hierarchical Building/Floor Classification and Location Estimation Using Wi-Fi Fingerprinting Based on Deep Neural Networks
    https://www.tandfonline.com/doi/abs/10.1080/01468030.2018.1467515

    get more advance based on paper 1. Not only floor detection, but also positioning estimation.
    http://kyeongsoo.github.io/research/projects/indoor_localization/index.html

  4. Indoor Fingerprint Positioning Based on Wi-Fi: An Overview
    http://www.mdpi.com/2220-9964/6/5/135/htm

    This is an overview. Two keywords: fingerprint, WiFi

  5. Learning the Localization Function: Machine Learning Approach to Fingerprinting Localization
    https://arxiv.org/abs/1803.08153

    basically the same as paper 2.

  1. CNN based Indoor Localization using RSS Time-Series
    https://www.researchgate.net/publication/325678644_CNN_based_Indoor_Localization_using_RSS_Time-Series

    using CNN to deal with long-term(time—series) dataset

  2. todo

KNN

  1. Clustering benefits in mobile-centric WiFi positioning in multi-floor buildings
    https://ieeexplore.ieee.org/abstract/document/7533846/

    cluster method for our dataset1

  2. 基于K均值聚类算法的位置指纹定位技术(chinese version)
    https://wenku.baidu.com/view/941a46e0192e45361166f505.html

    该文章很好的整理了为什么会用到kmeans和knn,以及提到kriging方法可用于创建指纹库

  3. Comprehensive analysis of distance and similarity measures for Wi-Fi fingerprinting indoor positioning systems
    https://www.sciencedirect.com/science/article/pii/S0957417415005527*

    The UJI kNN algorithm for dataset.
    This article mainly concentrates on the Wi-Fi Indoor positioning systems based on fingerprinting and k-NN. It mentions Non-heard data processing, data preprocessing, and all kinds of distance calculating.

  1. Adaptive K-nearest neighbour algorithm for WiFi fingerprint positioning
    Website: https://www.sciencedirect.com/science/article/pii/S240595951830050X

    This article focus on how to improve K nearest neighbour algorithm

Kriging algorithm:

  1. Method for yielding a database of locationfingerprints in WLAN
    https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1522067

    using kriging to generate fingerprint data

  2. Fingerprint Space Building Algorithm with Kriging for Large Positioning Regional Environment

  3. Applying Kriging Interpolation for WiFiFingerprinting based Indoor Positioning Systems

  4. kriging tutoril(Chinese version)
    克里金(Kriging)插值的原理与公式推导

other algorithms need arrange

  1. Dealing with Insufficient Location Fingerprints in Wi-Fi Based Indoor Location Fingerprinting
    https://www.hindawi.com/journals/wcmc/2017/1268515/

other maybe useful articles or resources: