Matches in SemOpenAlex for { <https://semopenalex.org/work/W4310509065> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W4310509065 abstract "Recent work has advocated for the use of deep learning to perform power allocation in the downlink of massive MIMO (maMIMO) networks. Yet, such deep learning models are vulnerable to adversarial attacks. In the context of maMIMO power allocation, adversarial attacks refer to the injection of subtle perturbations into the deep learning model's input, during inference (i.e., the adversarial perturbation is injected into inputs during deployment after the model has been trained) that are specifically crafted to force the trained regression model to output an infeasible power allocation solution. In this work, we develop an autoencoder-based mitigation technique, which allows deep learning-based power allocation models to operate in the presence of adversaries without requiring retraining. Specifically, we develop a denoising autoencoder (DAE), which learns a mapping between potentially perturbed data and its corresponding unperturbed input. We test our defense across multiple attacks and in multiple threat models and demonstrate its ability to (i) mitigate the effects of adversarial attacks on power allocation networks using two common precoding schemes, (ii) outperform previously proposed benchmarks for mitigating regression-based adversarial attacks on maMIMO networks, (iii) retain accurate performance in the absence of an attack, and (iv) operate with low computational overhead." @default.
- W4310509065 created "2022-12-11" @default.
- W4310509065 creator A5001355478 @default.
- W4310509065 creator A5020399355 @default.
- W4310509065 creator A5044519984 @default.
- W4310509065 creator A5084897218 @default.
- W4310509065 date "2022-11-28" @default.
- W4310509065 modified "2023-09-27" @default.
- W4310509065 title "Defending Adversarial Attacks on Deep Learning Based Power Allocation in Massive MIMO Using Denoising Autoencoders" @default.
- W4310509065 doi "https://doi.org/10.48550/arxiv.2211.15365" @default.
- W4310509065 hasPublicationYear "2022" @default.
- W4310509065 type Work @default.
- W4310509065 citedByCount "0" @default.
- W4310509065 crossrefType "posted-content" @default.
- W4310509065 hasAuthorship W4310509065A5001355478 @default.
- W4310509065 hasAuthorship W4310509065A5020399355 @default.
- W4310509065 hasAuthorship W4310509065A5044519984 @default.
- W4310509065 hasAuthorship W4310509065A5084897218 @default.
- W4310509065 hasBestOaLocation W43105090651 @default.
- W4310509065 hasConcept C101738243 @default.
- W4310509065 hasConcept C105339364 @default.
- W4310509065 hasConcept C108583219 @default.
- W4310509065 hasConcept C111919701 @default.
- W4310509065 hasConcept C119857082 @default.
- W4310509065 hasConcept C151730666 @default.
- W4310509065 hasConcept C154945302 @default.
- W4310509065 hasConcept C19247436 @default.
- W4310509065 hasConcept C207987634 @default.
- W4310509065 hasConcept C2776214188 @default.
- W4310509065 hasConcept C2779343474 @default.
- W4310509065 hasConcept C2779960059 @default.
- W4310509065 hasConcept C2780909371 @default.
- W4310509065 hasConcept C37736160 @default.
- W4310509065 hasConcept C41008148 @default.
- W4310509065 hasConcept C50644808 @default.
- W4310509065 hasConcept C54197355 @default.
- W4310509065 hasConcept C555944384 @default.
- W4310509065 hasConcept C76155785 @default.
- W4310509065 hasConcept C86803240 @default.
- W4310509065 hasConceptScore W4310509065C101738243 @default.
- W4310509065 hasConceptScore W4310509065C105339364 @default.
- W4310509065 hasConceptScore W4310509065C108583219 @default.
- W4310509065 hasConceptScore W4310509065C111919701 @default.
- W4310509065 hasConceptScore W4310509065C119857082 @default.
- W4310509065 hasConceptScore W4310509065C151730666 @default.
- W4310509065 hasConceptScore W4310509065C154945302 @default.
- W4310509065 hasConceptScore W4310509065C19247436 @default.
- W4310509065 hasConceptScore W4310509065C207987634 @default.
- W4310509065 hasConceptScore W4310509065C2776214188 @default.
- W4310509065 hasConceptScore W4310509065C2779343474 @default.
- W4310509065 hasConceptScore W4310509065C2779960059 @default.
- W4310509065 hasConceptScore W4310509065C2780909371 @default.
- W4310509065 hasConceptScore W4310509065C37736160 @default.
- W4310509065 hasConceptScore W4310509065C41008148 @default.
- W4310509065 hasConceptScore W4310509065C50644808 @default.
- W4310509065 hasConceptScore W4310509065C54197355 @default.
- W4310509065 hasConceptScore W4310509065C555944384 @default.
- W4310509065 hasConceptScore W4310509065C76155785 @default.
- W4310509065 hasConceptScore W4310509065C86803240 @default.
- W4310509065 hasLocation W43105090651 @default.
- W4310509065 hasOpenAccess W4310509065 @default.
- W4310509065 hasPrimaryLocation W43105090651 @default.
- W4310509065 hasRelatedWork W2404595106 @default.
- W4310509065 hasRelatedWork W2511279186 @default.
- W4310509065 hasRelatedWork W2567271240 @default.
- W4310509065 hasRelatedWork W2788487394 @default.
- W4310509065 hasRelatedWork W2922457425 @default.
- W4310509065 hasRelatedWork W2989980351 @default.
- W4310509065 hasRelatedWork W3044458868 @default.
- W4310509065 hasRelatedWork W4213225422 @default.
- W4310509065 hasRelatedWork W4250304930 @default.
- W4310509065 hasRelatedWork W4289656111 @default.
- W4310509065 isParatext "false" @default.
- W4310509065 isRetracted "false" @default.
- W4310509065 workType "article" @default.