Adaptive Beamforming Analysis for Directionality Using Data from a Vertical Array in the Mediterranean
Abstract
Data were taken from a 5-sensor vertical array in the Ionian Basin. The data were reduced via sophisticated analytical processes to find the vertical directionality of the ambient noise field. The directionality from the measured data was compared to several noise fields simulated a priori. The measured directionality was found not to correspond well to any of the priori simulated fields. It was observed that the amount of uncorrelated (circuit) noise has a strong influence on the measured directionality. Large percentages of uncorrelated noise depress the ability of an adaptively processed array to resolve the field. A new simulated noise field was generated to include the same percentage of uncorrelated noise as was reported for the recording instrumentation (10%). This simulation was found to compare very closely with the actual array response. Since the high uncorrelated noise probably is a result of the recording rather than the actual field, the directionality was again computed after 10% uncorrelated noise was removed from the spectral cross power matrix form of the actual data. The resultant directionality was then compared with that from a simulated field without uncorrelated noise, which is presented as a candidate directionality model for the Mediterranean Sea ambient noise at 97 Hz. This report is presented as a prototype example for directionality analysis of data from multi-element arrays (VLAM, FLIP, ACODAC, MODANS, SONODIVER, ITASS, etc.). Adaptive beamforming (ABF) algorithms were employed as the analysis tools. Array gain versus frequency results are also shown for actual and simulated noise fields.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1971
- Accession Number
- AD0517696
Entities
People
- Craig G. Anderson
- Dorothy A. Anderson
- Gerald L. Kinnison