Identifying Cognitive Radars and Inverse Tracking

Abstract

Cognitive radars are reconfigurable sensors that optimize their sensing mechanism and transmit functionalities. The concept of cognitive radar [36, 11, 32] has evolved over the last two decades and a common aspect is the sense-learn-adapt paradigm. A cognitive fully adaptive radar enables joint optimization of the transmit and receive functions by sensing (estimating) the radar channel that includes clutter and other interfering signals. This project addresses the next logical step, namely inverse cognitive radar. By observing the emissions of a radar, how can we identify if it is cognitive (rational utility maximizer) and how can we predict its future actions? The scientific challenges of this project involve extending Bayesian filtering, inverse reinforcement learning and stochastic optimization of dynamical systems to a data-driven adversarial setting. The research transcends classical statistical signal processing (sensing and estimation/detection theory) to address the deeper issue of how to infer strategy from sensing. The generative models, adversarial inference algorithms and associated mathematical analysis will lead to advances in understanding how sophisticated adaptive cognitive radars operate.

Document Details

Document Type
DoD Grant Award
Publication Date
Jan 21, 2022
Source ID
FA95502210016XX0

Entities

People

  • Vikram Krishnamurthy

Organizations

  • Air Force Office of Scientific Research
  • Cornell University
  • United States Air Force

Tags

Readers

  • Distributed Systems and Data Platform Development
  • Neural Network Machine Learning.
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms