CFARnet: Deep learning for target detection with constant false alarm rate

Tzvi Diskin*, Yiftach Beer, Uri Okun, Ami Wiesel

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


We consider the problem of target detection with a constant false alarm rate (CFAR). This constraint is crucial in many practical applications and is a standard requirement in classical composite hypothesis testing. In settings where classical approaches are computationally expensive or where only data samples are given, machine learning methodologies are advantageous. CFAR is less understood in these settings. To close this gap, we introduce a framework of CFAR constrained detectors. Theoretically, we prove that a CFAR constrained Bayes optimal detector is asymptotically equivalent to the classical generalized likelihood ratio test (GLRT). Practically, we develop a deep learning framework for fitting neural networks that approximate it. Experiments of target detection in different settings demonstrate that the proposed CFARnet allows a flexible tradeoff between CFAR and accuracy.

Original languageAmerican English
Article number109543
JournalSignal Processing
StatePublished - Oct 2024

Bibliographical note

Publisher Copyright:
© 2024 Elsevier B.V.


  • CFAR
  • Machine learning


Dive into the research topics of 'CFARnet: Deep learning for target detection with constant false alarm rate'. Together they form a unique fingerprint.

Cite this