Model-Based Assisted Deep Learning for Adaptive, Resilient Target Tracking and Identification of Electromagnetic Interference (EMI)

Abstract

This report summarizes the preliminary research effort on target tracking and identification of spoofing and jamming attacks on satellites from an Electromagnetic Interference (EMI) source. We hypothesize that it is possible to identify and track both active and passive EMI sources using the Radio Frequency (RF) signatures of different sources. The RF signatures are input into to a model-based deep neural network (DNN) that classifies and tracks different objects. To test our hypothesis, we study two settings: a) QAM modulation classification using a 2D convolutional neural network and b) people counting system using a MIMO radar with a 3D convolutional neural network.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 07, 2022
Accession Number
AD1200993

Entities

People

  • Thinh Nguyen

Organizations

  • Oregon State University

Tags

Communities of Interest

  • Energy and Power Technologies
  • Sensors
  • Space
  • Weapons Technologies

DTIC Thesaurus Topics

  • Air Force
  • Air Force Research Laboratories
  • Artificial Intelligence
  • Artificial Intelligence Software
  • Artificial Satellites
  • Bayesian Networks
  • Computational Science
  • Computer Vision
  • Convolutional Neural Networks
  • Deep Learning
  • Detection
  • Engineering
  • Identification
  • Kalman Filters
  • Machine Learning
  • Mimo Radar
  • Modulation
  • Multiple Input Multiple Output
  • Neural Networks
  • Radar
  • Three Dimensional

Readers

  • Neural Network Machine Learning.
  • Radar Systems Engineering.
  • Radio communications and signal processing.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks
  • Space
  • Space - Space Objects