https://www.mdu.se/

mdu.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Complex-Valued Convolutional Neural Network With Learnable Activation Function for Frequency-Domain Radar Signal Processing
Mälardalen University, Faculty of Engineering and Health Sciences, Department of Computer Science & Engineering.
Mälardalen University, Faculty of Engineering and Health Sciences, Department of Computer Science & Engineering. TalTech University, Department of Computer Systems, Tallinn, 19086, Estonia.
2026 (English)In: IEEE Transactions on Aerospace and Electronic Systems, ISSN 0018-9251, E-ISSN 1557-9603, Vol. 62, p. 3044-3059Article in journal (Refereed) Published
Abstract [en]

Recent advancements in deep learning and the availability of open-source datasets have enabled real-valued Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) to achieve high performance in Synthetic Aperture Radar (SAR) target recognition, SAR-based land use and land cover classification, radar micro-Doppler signature-based Human Activity Recognition (HAR), and Small Unmanned Aerial Vehicle (SUAV) target recognition. However, their high computational cost and resource requirements limit deployment in resource-constrained environments. Frequency-domain complex-valued CNNs have recently emerged as a promising alternative, leveraging the convolution theorem to perform efficient spectral convolutions, reducing computational complexity while preserving both amplitude and phase characteristics of SAR and Continuous-wave (CW) radar signals. Despite their potential, adoption remains constrained by the need for frequency-adaptive complex-valued layers, robust spectral complex-valued activation functions, and efficient parameter initialization methods. Traditional frequency-domain CVNNs often require frequent Fourier transform (FFT/IFFT) transitions (O}(n log n) complexity) for spatial-domain pooling and activations, increasing computational overhead. Additionally, many existing complex-valued CNNs employ real-valued activation functions on complex tensors in a split-type manner, which might destroy phase-magnitude relationships and reduce effectiveness for phase-sensitive tasks. Moreover, architectures that use complex-valued weights but rely on real-valued activation functions suffer from phase distortion, limited expressiveness, and mathematical inconsistency. Considering these limitations, we propose a frequency-adaptive complex-valued CNN with a complex-valued learnable activation function designed for SAR-based analysis, SUAV detection, and HAR. Our model operates entirely in the frequency domain and processes only complex-valued data. Extensive experiments on MSTAR-10, EuroSAT all bands, DIAT-μRadHAR, and DIAT-μSAT datasets exhibit performance comparable to existing real and complex-valued CNNs. Additionally, we show that our learnable activation function preserves phase-magnitude relationships and mitigates phase distortion, thereby enhancing accuracy compared to the C{ReLU} activation function. 

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE) , 2026. Vol. 62, p. 3044-3059
Keywords [en]
Deep learning, frequency-domain complex-valued convolutional neural networks, frequency-domain complex-valued learnable activation function, radar signal processing, synthetic aperture radar, Activation analysis, Antennas, Chemical activation, Classification (of information), Complex networks, Computational complexity, Continuous wave radar, Convolution, Convolutional neural networks, Deep neural networks, Frequency domain analysis, Land use, Open systems, Radar target recognition, Radioactivity logging, Tensors, Unmanned aerial vehicles (UAV), Activation functions, Complex-valued, Convolutional neural network, Domain complexes, Frequency domains, Frequency-domain complex-valued convolutional neural network, Radar signals, Signal-processing
National Category
Signal Processing
Identifiers
URN: urn:nbn:se:mdh:diva-75339DOI: 10.1109/TAES.2025.3646567ISI: 001700560400019Scopus ID: 2-s2.0-105025665939OAI: oai:DiVA.org:mdh-75339DiVA, id: diva2:2025503
Available from: 2026-01-07 Created: 2026-01-07 Last updated: 2026-03-11Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Chakraborty, MainakDaneshtalab, Masoud

Search in DiVA

By author/editor
Chakraborty, MainakDaneshtalab, Masoud
By organisation
Department of Computer Science & Engineering
In the same journal
IEEE Transactions on Aerospace and Electronic Systems
Signal Processing

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 332 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf