Robust Constant Curvature Curve Communications with Complex and Quaternion Neural Networks

dc.contributor.authorBuvarp, Anders M.en
dc.contributor.authorMili, Lamine M.en
dc.contributor.authorZaghloul, Amir I.en
dc.date.accessioned2024-12-13T13:59:47Zen
dc.date.available2024-12-13T13:59:47Zen
dc.date.issued2024-06-25en
dc.description.abstractThe concept of Digital Twin has recently emerged, which requires the transmission of a massive amount of sensor data with low latency and high reliability. Analog error correction is an attractive method for low-latency communications; hence, in this paper, we propose the use of complex-valued neural networks and Quaternionic Neural Networks (QNNs) to decode analog codes. Furthermore, we propose mapping our codes to the baseband of the frequency domain to enable easy time and frequency synchronization as well as to mitigate frequency-selective fading using robust estimation theory. This is accomplished by applying inverse Discrete Fourier Transform (DFT) modulation, which achieves a significant reduction in hardware complexity, power, and cost as compared to our previously proposed analog coding scheme. Additionally, we introduce a scaled version of our previous analog codes that enables statistical signal processing, something we have not been able to achieve until now. This achieves significant noise immunity with drastic performance improvements at low Signal-to-Noise Ratios (SNR) and a small loss at high SNR.en
dc.description.versionAccepted versionen
dc.format.mimetypeapplication/pdfen
dc.identifier.doihttps://doi.org/10.1109/TCOMM.2024.3418915en
dc.identifier.eissn1558-0857en
dc.identifier.issn0090-6778en
dc.identifier.orcidMili, Lamine [0000-0001-6134-3945]en
dc.identifier.urihttps://hdl.handle.net/10919/123791en
dc.language.isoenen
dc.publisherIEEEen
dc.rightsIn Copyrighten
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/en
dc.titleRobust Constant Curvature Curve Communications with Complex and Quaternion Neural Networksen
dc.title.serialIEEE Transactions on Communicationsen
dc.typeArticle - Refereeden
dc.type.dcmitypeTexten
dc.type.otherJournal Articleen
pubs.organisational-groupVirginia Techen
pubs.organisational-groupVirginia Tech/Engineeringen
pubs.organisational-groupVirginia Tech/Engineering/Electrical and Computer Engineeringen
pubs.organisational-groupVirginia Tech/All T&R Facultyen
pubs.organisational-groupVirginia Tech/Engineering/COE T&R Facultyen

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Robust_Constant_Curvature_Curve_Communications_with_Complex_and_Quaternion_Neural_Networks.pdf
Size:
1.99 MB
Format:
Adobe Portable Document Format
Description:
Accepted version
License bundle
Now showing 1 - 1 of 1
Name:
license.txt
Size:
1.5 KB
Format:
Plain Text
Description: