Improving Channel Charting with Representation-Constrained Autoencoders

Pengzhi Huang; Oscar Castaneda; Emre Gonultas; Sa Id Medjkouh; Olav Tirkkonen; Tom Goldstein; Christoph Studer

doi:10.1109/SPAWC.2019.8815478

Improving Channel Charting with Representation-Constrained Autoencoders

Pengzhi Huang, Oscar Castaneda, Emre Gonultas, Sa Id Medjkouh, Olav Tirkkonen, Tom Goldstein, Christoph Studer

Communications Theory

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

31 Citations (Scopus)

Abstract

Channel charting (CC) has been proposed recently to enable logical positioning of user equipments (UEs) in the neighborhood of a multi-antenna base-station solely from channel-state information (CSI). CC relies on dimensionality reduction of high-dimensional CSI features in order to construct a channel chart that captures spatial and radio geometries so that UEs close in space are close in the channel chart. In this paper, we demonstrate that autoencoder (AE)-based CC can be augmented with side information that is obtained during the CSI acquisition process. More specifically, we propose to include pairwise representation constraints into AEs with the goal of improving the quality of the learned channel charts. We show that such representation-constrained AEs recover the global geometry of the learned channel charts, which enables CC to perform approximate positioning without global navigation satellite systems or supervised learning methods that rely on extensive and expensive measurement campaigns.

Original language	English
Title of host publication	2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019
Publisher	IEEE
ISBN (Electronic)	978-1-5386-6528-2
ISBN (Print)	978-1-5386-6529-9
DOIs	https://doi.org/10.1109/SPAWC.2019.8815478
Publication status	Published - 1 Jul 2019
MoE publication type	A4 Conference publication
Event	IEEE International Workshop on Signal Processing Advances in Wireless Communications - Cannes, France Duration: 2 Jul 2019 → 5 Jul 2019 Conference number: 20

Publication series

Name	IEEE International Workshop on Signal Processing Advances in Wireless Communications
ISSN (Print)	2325-3789
ISSN (Electronic)	1948-3252

Workshop

Workshop	IEEE International Workshop on Signal Processing Advances in Wireless Communications
Abbreviated title	SPAWC
Country/Territory	France
City	Cannes
Period	02/07/2019 → 05/07/2019

Access to Document

10.1109/SPAWC.2019.8815478

https://arxiv.org/abs/1908.02878

Cite this

Huang, P., Castaneda, O., Gonultas, E., Medjkouh, S. I., Tirkkonen, O., Goldstein, T., & Studer, C. (2019). Improving Channel Charting with Representation-Constrained Autoencoders. In 2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019 Article 8815478 (IEEE International Workshop on Signal Processing Advances in Wireless Communications). IEEE. https://doi.org/10.1109/SPAWC.2019.8815478

@inproceedings{104378e7881e46ba9ba3e2e0312b861d,

title = "Improving Channel Charting with Representation-Constrained Autoencoders",

abstract = "Channel charting (CC) has been proposed recently to enable logical positioning of user equipments (UEs) in the neighborhood of a multi-antenna base-station solely from channel-state information (CSI). CC relies on dimensionality reduction of high-dimensional CSI features in order to construct a channel chart that captures spatial and radio geometries so that UEs close in space are close in the channel chart. In this paper, we demonstrate that autoencoder (AE)-based CC can be augmented with side information that is obtained during the CSI acquisition process. More specifically, we propose to include pairwise representation constraints into AEs with the goal of improving the quality of the learned channel charts. We show that such representation-constrained AEs recover the global geometry of the learned channel charts, which enables CC to perform approximate positioning without global navigation satellite systems or supervised learning methods that rely on extensive and expensive measurement campaigns.",

author = "Pengzhi Huang and Oscar Castaneda and Emre Gonultas and Medjkouh, {Sa Id} and Olav Tirkkonen and Tom Goldstein and Christoph Studer",

year = "2019",

month = jul,

day = "1",

doi = "10.1109/SPAWC.2019.8815478",

language = "English",

isbn = "978-1-5386-6529-9",

series = "IEEE International Workshop on Signal Processing Advances in Wireless Communications",

publisher = "IEEE",

booktitle = "2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019",

address = "United States",

note = "IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC ; Conference date: 02-07-2019 Through 05-07-2019",

}

Huang, P, Castaneda, O, Gonultas, E, Medjkouh, SI, Tirkkonen, O, Goldstein, T & Studer, C 2019, Improving Channel Charting with Representation-Constrained Autoencoders. in 2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019., 8815478, IEEE International Workshop on Signal Processing Advances in Wireless Communications, IEEE, IEEE International Workshop on Signal Processing Advances in Wireless Communications, Cannes, France, 02/07/2019. https://doi.org/10.1109/SPAWC.2019.8815478

Improving Channel Charting with Representation-Constrained Autoencoders. / Huang, Pengzhi; Castaneda, Oscar; Gonultas, Emre et al.
2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019. IEEE, 2019. 8815478 (IEEE International Workshop on Signal Processing Advances in Wireless Communications).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Improving Channel Charting with Representation-Constrained Autoencoders

AU - Huang, Pengzhi

AU - Castaneda, Oscar

AU - Gonultas, Emre

AU - Medjkouh, Sa Id

AU - Tirkkonen, Olav

AU - Goldstein, Tom

AU - Studer, Christoph

N1 - Conference code: 20

PY - 2019/7/1

Y1 - 2019/7/1

N2 - Channel charting (CC) has been proposed recently to enable logical positioning of user equipments (UEs) in the neighborhood of a multi-antenna base-station solely from channel-state information (CSI). CC relies on dimensionality reduction of high-dimensional CSI features in order to construct a channel chart that captures spatial and radio geometries so that UEs close in space are close in the channel chart. In this paper, we demonstrate that autoencoder (AE)-based CC can be augmented with side information that is obtained during the CSI acquisition process. More specifically, we propose to include pairwise representation constraints into AEs with the goal of improving the quality of the learned channel charts. We show that such representation-constrained AEs recover the global geometry of the learned channel charts, which enables CC to perform approximate positioning without global navigation satellite systems or supervised learning methods that rely on extensive and expensive measurement campaigns.

AB - Channel charting (CC) has been proposed recently to enable logical positioning of user equipments (UEs) in the neighborhood of a multi-antenna base-station solely from channel-state information (CSI). CC relies on dimensionality reduction of high-dimensional CSI features in order to construct a channel chart that captures spatial and radio geometries so that UEs close in space are close in the channel chart. In this paper, we demonstrate that autoencoder (AE)-based CC can be augmented with side information that is obtained during the CSI acquisition process. More specifically, we propose to include pairwise representation constraints into AEs with the goal of improving the quality of the learned channel charts. We show that such representation-constrained AEs recover the global geometry of the learned channel charts, which enables CC to perform approximate positioning without global navigation satellite systems or supervised learning methods that rely on extensive and expensive measurement campaigns.

UR - http://www.scopus.com/inward/record.url?scp=85072331155&partnerID=8YFLogxK

U2 - 10.1109/SPAWC.2019.8815478

DO - 10.1109/SPAWC.2019.8815478

M3 - Conference article in proceedings

AN - SCOPUS:85072331155

SN - 978-1-5386-6529-9

T3 - IEEE International Workshop on Signal Processing Advances in Wireless Communications

BT - 2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019

PB - IEEE

T2 - IEEE International Workshop on Signal Processing Advances in Wireless Communications

Y2 - 2 July 2019 through 5 July 2019

ER -

Huang P, Castaneda O, Gonultas E, Medjkouh SI, Tirkkonen O, Goldstein T et al. Improving Channel Charting with Representation-Constrained Autoencoders. In 2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2019. IEEE. 2019. 8815478. (IEEE International Workshop on Signal Processing Advances in Wireless Communications). doi: 10.1109/SPAWC.2019.8815478

Improving Channel Charting with Representation-Constrained Autoencoders

Abstract

Publication series

Workshop

Access to Document

Other files and links

Fingerprint

-: Radio Network Optimization for Heterogeneous Machine Connectivity

Cite this

Improving Channel Charting with Representation-Constrained Autoencoders

Abstract

Publication series

Workshop

Access to Document

Other files and links

Fingerprint

Projects

-: Radio Network Optimization for Heterogeneous Machine Connectivity

Cite this