Eigenanalysis techniques for canceling narrowband interferences / Alexander M. Haimovich.

Format:

Book

Manuscript

Microformat

Thesis/Dissertation

Author/Creator:

Haimovich, Alexander M., 1954-

Contributor:

Bar-Ness, Yeheskel, advisor.

University of Pennsylvania.

Language:

English

Subjects (All):

Penn dissertations--Systems engineering.

Systems engineering--Penn dissertations.

Local Subjects:

Penn dissertations--Systems engineering.

Systems engineering--Penn dissertations.

Physical Description:

xii, 160 leaves ; 29 cm

Production:

1989.

Summary:

Signal subspace methods are applied to interference cancellation problems. While with common methods of interference cancellation, the rejection of undesired signals is effected by global optimization procedures that include the interferences as well as the background noise, the technique developed in this dissertation focuses on the interferences only, resulting in superior cancellation performance. Although the signal to noise ratio is optimized in all cases, existing cancellation methods performance is dependent on the interferences power to background noise ratio and their spatial distribution. Those methods typically provide accurate pattern zeros for multiple interference sources spaced wider than a beamwidth. However, for sources spaced closer than a beamwidth the pattern zeros will be generally displaced from the true source positions, furthermore the array will not necessarily allocate a zero to each source. For some applications it may be desirable to maximize the elimination of directional interferences. For these applications, we suggest a canceler that handles the interferences independent of the background noise. First, the interference subspace is derived from the eigenanalysis of the array correlation matrix, then an optimal weight vector is computed in the subspace orthogonal to the interference subspace. For infinite observation time, this restriction of the weight vector will result in zero response to interferences, yet in lower overall signal-to-noise ratio. However, our results show that not only does the method have superior interference canceling capabilities, but its robustness for short observation times results in signal to noise ratios that significantly exceed those of the conventional methods.

Notes:

Adviser: Yeheskel Bar-Ness.

Thesis (Ph.D. in Systems Engineering) -- Graduate School of Arts and Sciences, University of Pennsylvania, 1989.

Includes bibliography.

Local Notes:

University Microfilms order no.: 89-22510.

OCLC:

244967673