Application of real data to sonar detection by neural networks

Abstract

Artificial neural networks are being studied for a variety of applications in ASW relating to the detection of target echoes by means of pattern recognition. This memorandum is a status report on an attempt to detect low-frequency submarine echoes in a set of real received time-domain beam outputs. The reason for studying the application of a neural network to the problem of detection stems from the fact that neural networks use cumulative a priori knowledge to perform a specific task rather than (as in the case of a simple peak detector) taking each signal as an isolated event.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 01, 1992
Accession Number
AD1114241

Entities

People

  • P. Nielsen

Organizations

  • SACLANT ASW Research Centre

Tags

DTIC Thesaurus Topics

  • Amplitude
  • Artificial Intelligence
  • Computer Programs
  • Computing System Architectures
  • Detection
  • Detectors
  • Echoes
  • False Alarms
  • Governments
  • Learning
  • Nato
  • Naval Warfare
  • Network Architecture
  • Neural Networks
  • Pattern Recognition
  • Probability
  • Recognition
  • Target Detection
  • Target Echoes
  • Test Sets
  • Warning Systems

Readers

  • Neural Network Machine Learning.
  • Radar Systems Engineering.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks