A Deep Learning Approach to Position Estimation from Channel Impulse Responses
Radio-based locating systems allow for a robust and continuous tracking in industrial environments and are a key enabler for the digitalization of processes in many areas such as production, manufacturing, and warehouse management. Time difference of arrival (TDoA) systems estimate the time-of-fligh...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2019-03-01
|
Series: | Sensors |
Subjects: | |
Online Access: | http://www.mdpi.com/1424-8220/19/5/1064 |
_version_ | 1797999502156627968 |
---|---|
author | Arne Niitsoo Thorsten Edelhäußer Ernst Eberlein Niels Hadaschik Christopher Mutschler |
author_facet | Arne Niitsoo Thorsten Edelhäußer Ernst Eberlein Niels Hadaschik Christopher Mutschler |
author_sort | Arne Niitsoo |
collection | DOAJ |
description | Radio-based locating systems allow for a robust and continuous tracking in industrial environments and are a key enabler for the digitalization of processes in many areas such as production, manufacturing, and warehouse management. Time difference of arrival (TDoA) systems estimate the time-of-flight (ToF) of radio burst signals with a set of synchronized antennas from which they trilaterate accurate position estimates of mobile tags. However, in industrial environments where multipath propagation is predominant it is difficult to extract the correct ToF of the signal. This article shows how deep learning (DL) can be used to estimate the position of mobile objects directly from the raw channel impulse responses (CIR) extracted at the receivers. Our experiments show that our DL-based position estimation not only works well under harsh multipath propagation but also outperforms state-of-the-art approaches in line-of-sight situations. |
first_indexed | 2024-04-11T11:04:25Z |
format | Article |
id | doaj.art-caea1614de514362aad88b25ee975d8d |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-04-11T11:04:25Z |
publishDate | 2019-03-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-caea1614de514362aad88b25ee975d8d2022-12-22T04:28:24ZengMDPI AGSensors1424-82202019-03-01195106410.3390/s19051064s19051064A Deep Learning Approach to Position Estimation from Channel Impulse ResponsesArne Niitsoo0Thorsten Edelhäußer1Ernst Eberlein2Niels Hadaschik3Christopher Mutschler4Machine Learning and Information Fusion Group, Precise Positioning and Analytics Department, Fraunhofer Institute for Integrated Circuits IIS, Nordostpark 84, 90411 Nürnberg, GermanyMachine Learning and Information Fusion Group, Precise Positioning and Analytics Department, Fraunhofer Institute for Integrated Circuits IIS, Nordostpark 84, 90411 Nürnberg, GermanyMachine Learning and Information Fusion Group, Precise Positioning and Analytics Department, Fraunhofer Institute for Integrated Circuits IIS, Nordostpark 84, 90411 Nürnberg, GermanyMachine Learning and Information Fusion Group, Precise Positioning and Analytics Department, Fraunhofer Institute for Integrated Circuits IIS, Nordostpark 84, 90411 Nürnberg, GermanyMachine Learning and Information Fusion Group, Precise Positioning and Analytics Department, Fraunhofer Institute for Integrated Circuits IIS, Nordostpark 84, 90411 Nürnberg, GermanyRadio-based locating systems allow for a robust and continuous tracking in industrial environments and are a key enabler for the digitalization of processes in many areas such as production, manufacturing, and warehouse management. Time difference of arrival (TDoA) systems estimate the time-of-flight (ToF) of radio burst signals with a set of synchronized antennas from which they trilaterate accurate position estimates of mobile tags. However, in industrial environments where multipath propagation is predominant it is difficult to extract the correct ToF of the signal. This article shows how deep learning (DL) can be used to estimate the position of mobile objects directly from the raw channel impulse responses (CIR) extracted at the receivers. Our experiments show that our DL-based position estimation not only works well under harsh multipath propagation but also outperforms state-of-the-art approaches in line-of-sight situations.http://www.mdpi.com/1424-8220/19/5/1064radio-based real-time locating systemstime difference of arrivalchannel impulse responsetime of arrivalposition estimationmachine learningdeep learningconvolutional neural networksdistributed CNN |
spellingShingle | Arne Niitsoo Thorsten Edelhäußer Ernst Eberlein Niels Hadaschik Christopher Mutschler A Deep Learning Approach to Position Estimation from Channel Impulse Responses Sensors radio-based real-time locating systems time difference of arrival channel impulse response time of arrival position estimation machine learning deep learning convolutional neural networks distributed CNN |
title | A Deep Learning Approach to Position Estimation from Channel Impulse Responses |
title_full | A Deep Learning Approach to Position Estimation from Channel Impulse Responses |
title_fullStr | A Deep Learning Approach to Position Estimation from Channel Impulse Responses |
title_full_unstemmed | A Deep Learning Approach to Position Estimation from Channel Impulse Responses |
title_short | A Deep Learning Approach to Position Estimation from Channel Impulse Responses |
title_sort | deep learning approach to position estimation from channel impulse responses |
topic | radio-based real-time locating systems time difference of arrival channel impulse response time of arrival position estimation machine learning deep learning convolutional neural networks distributed CNN |
url | http://www.mdpi.com/1424-8220/19/5/1064 |
work_keys_str_mv | AT arneniitsoo adeeplearningapproachtopositionestimationfromchannelimpulseresponses AT thorstenedelhaußer adeeplearningapproachtopositionestimationfromchannelimpulseresponses AT ernsteberlein adeeplearningapproachtopositionestimationfromchannelimpulseresponses AT nielshadaschik adeeplearningapproachtopositionestimationfromchannelimpulseresponses AT christophermutschler adeeplearningapproachtopositionestimationfromchannelimpulseresponses AT arneniitsoo deeplearningapproachtopositionestimationfromchannelimpulseresponses AT thorstenedelhaußer deeplearningapproachtopositionestimationfromchannelimpulseresponses AT ernsteberlein deeplearningapproachtopositionestimationfromchannelimpulseresponses AT nielshadaschik deeplearningapproachtopositionestimationfromchannelimpulseresponses AT christophermutschler deeplearningapproachtopositionestimationfromchannelimpulseresponses |