We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Networking and Internet Architecture

Title: Narrowband Interference Detection via Deep Learning

Abstract: Due to the increased usage of spectrum caused by the exponential growth of wireless devices, detecting and avoiding interference has become an increasingly relevant problem to ensure uninterrupted wireless communications. In this paper, we focus our interest on detecting narrowband interference caused by signals that despite occupying a small portion of the spectrum only can cause significant harm to wireless systems, for example, in the case of interference with pilots and other signals that are used to equalize the effect of the channel or attain synchronization. Due to the small sizes of these signals, detection can be difficult due to their low energy footprint, while greatly impacting (or denying completely in some cases) network communications. We present a novel narrowband interference detection solution that utilizes convolutional neural networks (CNNs) to detect and locate these signals with high accuracy. To demonstrate the effectiveness of our solution, we have built a prototype that has been tested and validated on a real-world over-the-air large-scale wireless testbed. Our experimental results show that our solution is capable of detecting narrowband jamming attacks with an accuracy of up to 99%. Moreover, it is also able to detect multiple attacks affecting several frequencies at the same time even in the case of previously unseen attack patterns. Not only can our solution achieve a detection accuracy between 92% and 99%, but it does so by only adding an inference latency of 0.093ms.
Comments: 6 pages, 10 figures, 1 table. ICC 2023 - IEEE International Conference on Communications, Rome, Italy, May 2023
Subjects: Networking and Internet Architecture (cs.NI); Signal Processing (eess.SP)
Cite as: arXiv:2301.09607 [cs.NI]
  (or arXiv:2301.09607v1 [cs.NI] for this version)

Submission history

From: Clifton Robinson [view email]
[v1] Mon, 23 Jan 2023 18:10:58 GMT (4322kb)

Link back to: arXiv, form interface, contact.