Narrowband Signal Processor

Abstract

A method to process narrowband signals includes dividing the input signal into segments of length N, where N optimizes the system's filter bandwidth, Fast Fourier Transform (FFT) size, processing capabilities and memory constraints. Each N-length segment is processed sequentially by filtering, a FFT and a peak detector that identifies the N-length segment's K largest spectral components. The frequency, bandwidth and power for the K largest spectral components are stored sequentially as N- processed data. After processing multiple N-length segments, processing intervals of length NxJ are obtained by sequentially reconstructing individual frequency spectrums for J continuous segments of the N-processed data, mapping the J reconstructed spectrums to a single composite spectrum, and applying the peak detector to the composite spectrum to separately store the composite spectrum's K largest frequencies, with respective powers and bandwidths, as (NxJ)-processed data. The N-length data is continually processed in groups of J until all N-length data is reprocessed. J may have multiple values, thereby generating multiple processed data sets.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 20, 2000
Accession Number
ADD020190

Entities

People

  • Scott D. Fisher

Organizations

  • Naval Undersea Warfare Center

Tags

DTIC Thesaurus Topics

  • Bandwidth
  • Composite Materials
  • Computational Complexity
  • Detectors
  • Filters
  • Filtration
  • Frequency
  • Frequency Domain
  • Governments
  • Intervals
  • Inventions
  • Narrowband
  • Power Spectra
  • Spectra
  • Undersea Warfare
  • United States
  • United States Government

Fields of Study

  • Engineering

Readers

  • Image Processing and Computer Vision.
  • Regression Analysis.