Poster: Towards NLOS Ranging Error Detection and Mitigation using Machine Learning on  Embedded Ultra-Wideband Devices

Markus Gallacher; Carlo Alberto Boano; Michael Stocker; Kay Uwe Römer

Poster: Towards NLOS Ranging Error Detection and Mitigation using Machine Learning on Embedded Ultra-Wideband Devices

Markus Gallacher, Carlo Alberto Boano, Michael Stocker, Kay Uwe Römer

Institute of Technical Informatics (4480)

Research output: Contribution to conference › Poster › peer-review

Abstract

We study the ranging error classification and mitigation capabilities of machine learning models used in ultrawideband systems. This is relevant, as distance estimates in non-line-of-sight (NLOS) conditions can be off by several
meters, which may severely compromise the performance of applications that require location awareness. Our ultimate goal is to optimize the size of a convolutional neural network (CNN) used for classifying and mitigating ranging errors such that it can run on constrained embedded devices without affecting its performance. To this end, we present an optimized CNN implementation that, in contrast to resourcehungry machine learning models requiring hundreds of kB
of memory, can classify and mitigate NLOS conditions with 12 kB of RAM and 75 kB of ROM.

Original language	English
Publication status	Published - Oct 2022
Event	19th International Conference on Embedded Wireless Systems and Networks: EWSN 2022 - Linz, Linz, Austria Duration: 3 Oct 2022 → 5 Oct 2022 https://ewsn2022.jku.at/

Conference

Conference	19th International Conference on Embedded Wireless Systems and Networks
Abbreviated title	EWSN 2022
Country/Territory	Austria
City	Linz
Period	3/10/22 → 5/10/22
Internet address	https://ewsn2022.jku.at/

Fields of Expertise

Information, Communication & Computing

Access to Document

https://ewsn2022.pro2future.at/paper/posters+demos/ewsn22posters-final24.pdf

Intelligent & Networked Embedded Systems
Boano, C. A., Römer, K. U., Schuß, M., Cao, N., Saukh, O., Hofmann, R., Stocker, M., Schuh, M. P., Papst, F., Salomon, E., Brunner, H., Gallacher, M., Mohamed Hydher, M. H., Wang, D., Corti, F., Krisper, M., Basic, F. & Petrovic, K.
1/09/13 → …
Project: Research area

Poster: Towards NLOS Ranging Error Detection and Mitigation using Machine Learning on Embedded Ultra-Wideband Devices
Markus Gallacher (Speaker)
3 Oct 2022 → 5 Oct 2022
Activity: Talk or presentation › Poster presentation › Science to science

Cite this

Gallacher, M., Boano, C. A., Stocker, M., & Römer, K. U. (2022). Poster: Towards NLOS Ranging Error Detection and Mitigation using Machine Learning on Embedded Ultra-Wideband Devices. Poster session presented at 19th International Conference on Embedded Wireless Systems and Networks, Linz, Austria. https://ewsn2022.pro2future.at/paper/posters+demos/ewsn22posters-final24.pdf

Poster: Towards NLOS Ranging Error Detection and Mitigation using Machine Learning on Embedded Ultra-Wideband Devices. / Gallacher, Markus ; Boano, Carlo Alberto ; Stocker, Michael et al.
2022. Poster session presented at 19th International Conference on Embedded Wireless Systems and Networks, Linz, Austria.

Research output: Contribution to conference › Poster › peer-review

@conference{592a5b916d7843dfa9663fd089153769,

title = "Poster: Towards NLOS Ranging Error Detection and Mitigation using Machine Learning on Embedded Ultra-Wideband Devices",

abstract = "We study the ranging error classification and mitigation capabilities of machine learning models used in ultrawideband systems. This is relevant, as distance estimates in non-line-of-sight (NLOS) conditions can be off by severalmeters, which may severely compromise the performance of applications that require location awareness. Our ultimate goal is to optimize the size of a convolutional neural network (CNN) used for classifying and mitigating ranging errors such that it can run on constrained embedded devices without affecting its performance. To this end, we present an optimized CNN implementation that, in contrast to resourcehungry machine learning models requiring hundreds of kBof memory, can classify and mitigate NLOS conditions with 12 kB of RAM and 75 kB of ROM.",

author = "Markus Gallacher and Boano, {Carlo Alberto} and Michael Stocker and R{\"o}mer, {Kay Uwe}",

year = "2022",

month = oct,

language = "English",

note = "19th International Conference on Embedded Wireless Systems and Networks : EWSN 2022, EWSN 2022 ; Conference date: 03-10-2022 Through 05-10-2022",

url = "https://ewsn2022.jku.at/",

}

TY - CONF

T1 - Poster: Towards NLOS Ranging Error Detection and Mitigation using Machine Learning on Embedded Ultra-Wideband Devices

AU - Gallacher, Markus

AU - Boano, Carlo Alberto

AU - Stocker, Michael

AU - Römer, Kay Uwe

PY - 2022/10

Y1 - 2022/10

N2 - We study the ranging error classification and mitigation capabilities of machine learning models used in ultrawideband systems. This is relevant, as distance estimates in non-line-of-sight (NLOS) conditions can be off by severalmeters, which may severely compromise the performance of applications that require location awareness. Our ultimate goal is to optimize the size of a convolutional neural network (CNN) used for classifying and mitigating ranging errors such that it can run on constrained embedded devices without affecting its performance. To this end, we present an optimized CNN implementation that, in contrast to resourcehungry machine learning models requiring hundreds of kBof memory, can classify and mitigate NLOS conditions with 12 kB of RAM and 75 kB of ROM.

AB - We study the ranging error classification and mitigation capabilities of machine learning models used in ultrawideband systems. This is relevant, as distance estimates in non-line-of-sight (NLOS) conditions can be off by severalmeters, which may severely compromise the performance of applications that require location awareness. Our ultimate goal is to optimize the size of a convolutional neural network (CNN) used for classifying and mitigating ranging errors such that it can run on constrained embedded devices without affecting its performance. To this end, we present an optimized CNN implementation that, in contrast to resourcehungry machine learning models requiring hundreds of kBof memory, can classify and mitigate NLOS conditions with 12 kB of RAM and 75 kB of ROM.

M3 - Poster

T2 - 19th International Conference on Embedded Wireless Systems and Networks

Y2 - 3 October 2022 through 5 October 2022

ER -

Poster: Towards NLOS Ranging Error Detection and Mitigation using Machine Learning on Embedded Ultra-Wideband Devices

Abstract

Conference

Fields of Expertise

Access to Document

Fingerprint

Projects

Intelligent & Networked Embedded Systems

Activities

Poster: Towards NLOS Ranging Error Detection and Mitigation using Machine Learning on Embedded Ultra-Wideband Devices

Cite this