Accession Number:

ADA270608

Title:

State-Space Models for Multichannel Detection

Descriptive Note:

Final technical rept. Apr 93-Jan 93,

Corporate Author:

SCIENTIFIC STUDIES CORP PALM BEACH GARDENS FL

Personal Author(s):

Report Date:

1993-07-01

Pagination or Media Count:

138.0

Abstract:

In multichannel identification and detection or model-based multichannel detection problems the parameters of a model are identified from the observed channel process and the identified model is used to facilitate the detection of a signal in the observe process. A model-based multichannel detection algorithm was developed in the an innovations-based detection algorithm IBDA formulation for surveillance radar system applications. The state space model class was adopted to model the vector channel process because it is more general than the time series model class used in most analyses to date. An IBDA methodology was developed based on the canonical correlations algorithm which for state-space model identification offers performance advantages over alternative techniques. A computer simulation was developed to validate the methodology and the algorithm, and to carry out performance assessments. Simulation results indicate that the algorithm is capable of discriminating between the null hypothesis clutter plus noise and the alternative hypothesis signal plus clutter plus noise. In summary, the applicability of the approach to radar system problems has been established.

Subject Categories:

  • Active and Passive Radar Detection and Equipment

Distribution Statement:

APPROVED FOR PUBLIC RELEASE