A Simultaneous M-ary Channel Hypothesis Test with Least-Mean-Square Signal Amplitude Estimation

Abstract

This paper presents a new strategy for detecting and estimating multiple linearly independent signals immersed in additive random noise. This new approach is a Maximum Log-likelihood Ratio Test (MLRT) and determines the presence of one or more signals through a single filter operating on a received input data vector. In addition, it performs a signal-amplitude estimation when a positive response occurs. This test is a Uniformly Most Powerful Invariant Test and reduces the computationally intensive work to only those situations where important data exists. The estimation technique determines the signal amplitude of all possible input signals in a least-mean-square sense with a variance inversely proportional to the inherent signal-to-noise ratio of each linearly independent library waveform. Computer simulations verify the theory and demonstrate near-optimum performance of the MLRT for small signal libraries. The MLRT is close to optimum for larger signal sets, and is clearly optimum for complex multiwaveform signals. Keywords include: Multivariate Gaussian nature, Maximum likelihood ratio test(MLRT), and Signal-to-noise ratio.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 1987
Accession Number
ADA186526

Entities

People

  • I. S. Reed
  • L. B. Stotts

Tags

Communities of Interest

  • C4I
  • Energy and Power Technologies
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Computational Science
  • Computer Programs
  • Computer Simulations
  • Computers
  • Data Processing
  • Detection
  • Equations
  • Image Processing
  • Probability
  • Probability Density Functions
  • Probability Distributions
  • Random Variables
  • Sequences
  • Signal Detection
  • Signal Processing
  • Statistics
  • Waveforms

Fields of Study

  • Engineering

Readers

  • Computational Modeling and Simulation
  • Radio communications and signal processing.
  • Statistical inference.