Wasserstein Mechanism for Hiding Cognition in Metacognitive Radars: A Distribution Privacy Framework

In this paper, we consider a cognitive radar electronic warfare scenario consisting of an inverse learning adversary - the adversary aims to learn the cognitive capabilities of the radar. The cognitive radar, as an electronic countercountermeasure, hides its cognitive capabilities during normal operation, a capability present in metacognitive radars. We formulate the problem of hiding cognition in the distribution privacy framework, a framework from machine learning that guarantees privacy from learning. The main contributions of the paper are: (i) Proposed an algorithm, based on Wasserstein mechanism, that guarantees distribution privacy, i.e. hiding cognitive capabilities of the radar, (ii) Characterizing the loss in utility while hiding cognition, (iii) Numerical results, for a beam allocation strategy, demonstrates that the proposed algorithm outperforms existing methodology in terms of utility loss and privacy guarantees, for a given privacy budget. In comparison to existing methodologies, the proposed algorithm requires minimal changes to the cognitive radar controller.

• Publication year

  2025

• Venue

  International Radar Conference

• Publication date

  2025-10-04

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1109/RadarConf2559087.2025.11204857
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Bayesian Inverse Learning and Online Changepoint Detection of Cognitive Radar Strategies

  2025influential reference
• Bayesian Non-Parametric Active Learning for Information Asymmetric ECCM in Cognitive Radar Electronic Warfare

  2025cited by this paper
• Radar Jamming Decision-Making in Cognitive Electronic Warfare: A Review

  2023cited by this paper
• Metacognitive Radar: Masking Cognition From an Inverse Reinforcement Learner

  2023influential reference
• Adaptive Eccm for Mitigating Smart Jammers

  2022cited by this paper
• Protecting Global Properties of Datasets with Distribution Privacy Mechanisms

  2022influential reference
• Principal–Agent Problem as a Principled Approach to Electronic Counter-Countermeasures in Radar

  2021cited by this paper
• Identifying Cognitive Radars - Inverse Reinforcement Learning Using Revealed Preferences

  2019cited by this paper
• Deep Q-Network based Anti-Jamming Strategy Design for Frequency Agile Radar

  2019cited by this paper
• How to Calibrate Your Adversary's Capabilities? Inverse Filtering for Counter-Autonomous Systems

  2019cited by this paper
• Local Obfuscation Mechanisms for Hiding Probability Distributions

  2018cited by this paper
• Pufferfish Privacy Mechanisms for Correlated Data

  2016cited by this paper
• Cognitive Radar Framework for Target Detection and Tracking

  2015cited by this paper
• Equilibrium between a statistical MIMO radar and a jammer

  2015cited by this paper
• The Algorithmic Foundations of Differential Privacy

  2014cited by this paper
• RADAR SIGNALS

  2013cited by this paper
• Principles of Modern Radar: Basic Principles

  2013cited by this paper
• The MIMO Radar and Jammer Games

  2012cited by this paper
• Revealed Preference and its Applications

  2012cited by this paper
• Directional Statistics, I

  2011cited by this paper
• Cognitive radar: a way of the future

  2006cited by this paper
• Differential Privacy

  2006cited by this paper
• Apprenticeship learning via inverse reinforcement learning

  2004cited by this paper
• Algorithms for Inverse Reinforcement Learning

  2000cited by this paper

• No citing papers are available for this paper.