%0 Conference Proceedings
%T CSI based indoor localization using Ensemble Neural Networks
%+ Département Réseaux et Services Multimédia Mobiles (TSP - RS2M)
%+ Institut Polytechnique de Paris (IP Paris)
%+ Réseaux, Systèmes, Services, Sécurité (R3S-SAMOVAR)
%+ Centre National de la Recherche Scientifique (CNRS)
%+ Laboratoire d'Informatique Gaspard-Monge (LIGM)
%+ Wireless Networking for Evolving & Adaptive Applications (EVA)
%A Sobehy, Abdallah
%A Renault, Eric
%A Mühlethaler, Paul
%< avec comité de lecture
%( Machine Learning for Networking: second IFIP TC 6 international conference, MLN 2019, Paris, France, December 3–5, 2019, revised selected papers
%B MLN 2019 : 2nd IFIP International Conference on Machine Learning for Networking
%C Paris, France
%I Springer
%3 Lecture Notes in Computer Science book series (LNCS)
%V 12081
%P 367-378
%8 2019-12-03
%D 2019
%R 10.1007/978-3-030-45778-5_25
%K Deep learning
%K Channel state information
%K MIMO
%K Neural networks
%K Indoor localization
%Z Computer Science [cs]
%Z Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]
%Z Computer Science [cs]/Machine Learning [cs.LG]
%Z Computer Science [cs]/Robotics [cs.RO]
%Z Computer Science [cs]/Programming Languages [cs.PL]
%Z Computer Science [cs]/Software Engineering [cs.SE]Conference papers
%X Indoor localization has attracted much attention due to its indispensable applications e.g. autonomous driving, Internet-Of-Things (IOT), and routing, etc. Received Signal Strength Indicator (RSSI) was used intensively to achieve localization. However, due to its temporal instability, the focus has shifted towards the use of Channel State Information (CSI) aka channel response. In this paper, we propose a deep learning solution for the indoor localization problem using the CSI of a 2 × 8 Multiple Input Multiple Output (MIMO) antenna. The variation of the magnitude component of the CSI is chosen as the input for a multi- layer Perceptron (MLP) neural network. Data augmentation is used to improve the learning process. Finally, various MLP neural networks are constructed using different portions of the training set and different hy- perparameters. An ensemble neural network technique is then used to process the predictions of the MLPs in order to enhance the position estimation. Our method is compared with two other deep learning so- lutions; one that uses Convolutional Neural Network (CNN), and the other uses MLP. The proposed method yields higher accuracy than its counterparts.
%G English
%2 https://hal.science/hal-02334588/document
%2 https://hal.science/hal-02334588/file/v1_%20CSI_based_Indoor_localization_using_Ensemble_Neural_Networks.pdf
%L hal-02334588
%U https://hal.science/hal-02334588
%~ INSTITUT-TELECOM
%~ ENPC
%~ CNRS
%~ INRIA
%~ INRIA-ROCQ
%~ TELECOM-SUDPARIS
%~ PARISTECH
%~ LIGM
%~ TESTALAIN1
%~ LIGM_LRT
%~ IFIP-LNCS
%~ IFIP
%~ INRIA2
%~ IFIP-TC
%~ IFIP-TC6
%~ IP_PARIS
%~ INSTITUTS-TELECOM
%~ IFIP-LNCS-12081
%~ IFIP-MLN
%~ UNIV-EIFFEL
%~ U-EIFFEL
%~ JSE2024