Multi-Signal Classification Using Deep Learning and Sparse Arrays

Shebert, Samuel R.; Amin, Moeness G.; Kirk, Benjamin H.; Buehrer, R. Michael

doi:10.1109/milcom55135.2022.10017861

Cited by 8 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is partially because adding more interfering signals increases the likelihood that signals have close angular spacing and makes weak signals more difficult to detect. In a previous work [3], we showed that using a sparse linear array increases classification accuracy in scenarios with many emitters, close angular spacing between emitters, or large power imbalances without increasing the number of array elements.…”

Section: Resultsmentioning

confidence: 92%

“…The authors acknowledge Advanced Research Computing at Virginia Tech for providing computational resources that have contributed to the results reported within this paper, URL: https://arc.vt.edu/. Portions of this paper were presented at GLOBECOM 2021 [1], MILCOM 2021 [2], and MILCOM 2022 [3]. classification problems and have high computational complexity.…”

Section: Introductionmentioning

confidence: 99%

“…The goal of an open set classifier is to detect data from "unknown" classes, while still accurately classifying data from the "known" classes [30]. In the context of deep learning classifiers, data from "known" classes are In this paper, we propose an open set CNN-based wireless standard classifier that uses a modified version of the unknown class detector originally developed for text classification by the authors in [32] [3] by introducing expert feature classifiers and considering signal detection and isolation. Specifically, the deep learning feature preprocessing method from [1], the open set unknown class detector from [2], and the co-frequency signal classification system from [3] were used in this paper.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Open Set Wireless Signal Classification: Augmenting Deep Learning with Expert Feature Classifiers

Shebert¹,

Kirk²,

Buehrer³

2023

Preprint

View full text Add to dashboard Cite

Section: Resultsmentioning

confidence: 92%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Open Set Wireless Signal Classification: Augmenting Deep Learning with Expert Feature Classifiers

Shebert¹,

Kirk²,

Buehrer³

2023

Preprint

View full text Add to dashboard Cite

“…A well known classical adaptive beamformer was proposed by Capon, named as minimum variance distortionless beamformer (MVDR) [2]. Various classical, as well as state of the art beamformers, have been mostly confined to the case of ULA, but to reduce the complexity of a system, non-uniform arrays and sparse arrays can be used instead of ULA [3], [4], [5]. Sparse arrays are created by choosing some antenna elements from the entire antenna array and only the chosen elements are used for further computation.…”

Section: Introductionmentioning

confidence: 99%

Improvements in Sparse Array Based Beamformer via Additional Constraints

Thakur,

Baghel,

Talluri

2023

IJCDS

View full text Add to dashboard Cite

This paper presents the improvements that can be achieved through the additional constraints in the design of highly sparse arrays beamforming. The design of a highly sparse array beamformer is an optimization problem of multivariable functions and its performance depends on the given conditions. For a uniform linear array (ULA), the Output Signal to Noise Ratio (OSINR) is a linear function of antenna elements of the array. However, this linearity is not consistent in various antenna array environments. This deviation can be minimized by selecting an optimal sparse array with a lesser number of elements than that are actually required by an ULA. In this paper, a two-step procedure has been presented to find an optimal solution for designing a highly sparse antenna array beamformer with sidelobe control. Furthermore, to avoid the grating lobes and minimize the mutual coupling between the elements an additional degree of freedom in the form of separation between antenna elements is added to the optimization problem. Simulations results provided in the paper reveals that sparse arrays with the additional conditions are better in comparison with the arrays without constraints in terms of sidelobe level, grating lobes and mutual coupling.

show abstract

“…A considerable number of studies have been conducted on antenna modelling, especially those related to WG modelling approach [12,13], since it can significantly reduce the computational cost. It is worth noting that investigations on sparse antennas (SA) are also relevant nowadays [14][15][16][17], since they can assist in optimizing the antenna structure and obtaining desirable characteristics. Therefore, combining the WG approach and the SA designing technology into one approach with a MoM core would be reasonable to take advantage of their benefits [18].…”

Section: Introductionmentioning

confidence: 99%

Modelling and Designing Wire-Grid Sparse Antennas Using MoM-based Approaches for Enhanced Performance and Reduced Cost

Alhaj Hasan,

Nguyen,

Gazizov

2023

Mikrotalasna revija = Microwave Review

View full text Add to dashboard Cite

In today's competitive market, it is crucial to continually improve antenna manufacturing technologies to keep up with the increasing demands. Researchers can utilize numerical methods to create novel and highly effective antenna designs, pushing the boundaries of antenna engineering. This paper considers two approaches to efficiently model and design sparse wire-grid antennas. The main idea is to create an optimal wire structure that closely approximates the current paths in the antenna while maintaining its integrity with minimal mass. This structure can be used in further simulations with controlled characteristic accuracy and less resources. In this study, these approaches were applied to modelling horn, conical horn, and reflector antennas. The verification was conducted by comparing their results with those obtained from numerical and experimental tests for the same antenna designs. The findings demonstrated good agreement between the results. In addition, we investigated the impact of using these approaches on the antenna characteristics and found that they can enhance the performance of certain characteristics while reducing costs in manufacturing and modelling.

show abstract

Multi-Signal Classification Using Deep Learning and Sparse Arrays

Cited by 8 publications

References 21 publications

Open Set Wireless Signal Classification: Augmenting Deep Learning with Expert Feature Classifiers

Open Set Wireless Signal Classification: Augmenting Deep Learning with Expert Feature Classifiers

Improvements in Sparse Array Based Beamformer via Additional Constraints

Modelling and Designing Wire-Grid Sparse Antennas Using MoM-based Approaches for Enhanced Performance and Reduced Cost

Contact Info

Product

Resources

About