Alternative Representation of Information for Acoustics

Abstract

The description of this effort can be located in R2 sub-activity Mine/obstacle detection in PE 0602782N. Funds are provided to Penn" State University for applied research in acoustic scattering to enhance capabilities in mine/obstacle detection for ONR s applied research efforts that investigates and creates new solutions to enable naval forces to conduct more rapid and effective mine detection. The research will include investigation of multiple perspectives and approaches to understanding the information embedded in the data collected by acoustic interrogation of the environment and of man-made objects. This includes signalprocessing that will lead to simple and reliable feature extraction for improved detection and classification of targets: First is a physics-based, intuition-driven approach aimed at interpretability of features and mindful of physical sonar configurations. Second is a principled application of well-understood and robust signal processing methods to the MCM problem with an emphasis on investigation of parameter trade-offs. Third is an approach investigating hierarchical knowledge representation inspired by cognitive psychology, vision and auditory science, with tools from information theory. This research will focus on understanding the acoustic information structure, methods" for extracting latent information about target and environment, andrepresentation trade-offs.

Document Details

Document Type
DoD Grant Award
Publication Date
Apr 24, 2019
Source ID
N000141912221

Entities

People

  • Joonho Park

Organizations

  • Office of Naval Research
  • Pennsylvania State University
  • United States Navy

Tags

Readers

  • Acoustical Oceanography.
  • Neural Network Machine Learning.
  • Systems Analysis and Design

Technology Areas

  • AI & ML