Loss Design for Single-carrier Joint Communication and Neural Network-based Sensing

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Muth, Charlotte, Geiger, Benedikt, Gaviria, Daniel Gil, Schmalen, Laurent
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910354271895552
author Muth, Charlotte
Geiger, Benedikt
Gaviria, Daniel Gil
Schmalen, Laurent
author_facet Muth, Charlotte
Geiger, Benedikt
Gaviria, Daniel Gil
Schmalen, Laurent
contents We evaluate the influence of multi-snapshot sensing and varying signal-to-noise ratio (SNR) on the overall performance of neural network (NN)-based joint communication and sensing (JCAS) systems. To enhance the training behavior, we decouple the loss functions from the respective SNR values and the number of sensing snapshots, using bounds of the sensing performance. Pre-processing is done through conventional sensing signal processing steps on the inputs to the sensing NN. The proposed method outperforms classical algorithms, such as a Neyman-Pearson-based power detector for object detection and ESPRIT for angle of arrival (AoA) estimation for quadrature amplitude modulation (QAM) at low SNRs.
format Preprint
id arxiv_https___arxiv_org_abs_2403_02929
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Loss Design for Single-carrier Joint Communication and Neural Network-based Sensing
Muth, Charlotte
Geiger, Benedikt
Gaviria, Daniel Gil
Schmalen, Laurent
Signal Processing
We evaluate the influence of multi-snapshot sensing and varying signal-to-noise ratio (SNR) on the overall performance of neural network (NN)-based joint communication and sensing (JCAS) systems. To enhance the training behavior, we decouple the loss functions from the respective SNR values and the number of sensing snapshots, using bounds of the sensing performance. Pre-processing is done through conventional sensing signal processing steps on the inputs to the sensing NN. The proposed method outperforms classical algorithms, such as a Neyman-Pearson-based power detector for object detection and ESPRIT for angle of arrival (AoA) estimation for quadrature amplitude modulation (QAM) at low SNRs.
title Loss Design for Single-carrier Joint Communication and Neural Network-based Sensing
topic Signal Processing
url https://arxiv.org/abs/2403.02929