Beampattern bounds for block errors in sonar arrays using interval arithmetic

When evaluating the imaging performance of sensor arrays, errors are typically modeled independently for each element. This is a reasonable assumption if the elements of the array are mounted onto a single structure. However, it may be desirable to construct the array from several modules, each cont...

Full description

Saved in:
Bibliographic Details
Published inOCEANS 2023 - Limerick pp. 1 - 7
Main Authors Arnestad, Havard Kjellmo, Gereb, Gabor, Lonmo, Tor Inge Birkenes, Kirkebo, Jan Egil
Format Conference Proceeding
LanguageEnglish
Published IEEE 05.06.2023
Subjects
Antenna arrays
Antenna radiation patterns
Beampattern
Block errors
Fourier series
Imaging
Interval analysis
Interval arithmetic
Microphone arrays
Modular arrays
Modular construction
Oceans
Synthetic aperture sonar
Tolerance analysis
Transducers
Ultrasonic transducer arrays
Online AccessGet full text

Cover

More Information
Summary:When evaluating the imaging performance of sensor arrays, errors are typically modeled independently for each element. This is a reasonable assumption if the elements of the array are mounted onto a single structure. However, it may be desirable to construct the array from several modules, each containing multiple elements. In this case, errors in module placement may dominate the error for each of the individual elements within the module, impacting the beampattern. In this study, we propose a method for modeling the worst-case imaging performance of modular sonar arrays subject to interval errors, where deviations from the design specifications of the array are bounded and have a block structure. We take into account modular position and orientation errors. The method is based on interval arithmetic, and is flexible with respect to array layout and design. It is found that modular errors can give characteristic sidelobe profiles due to the periodic error structure. We then show that for uniform linear arrays the error effects can be understood via Fourier analysis. Compared to independent element errors, worst-case modular errors are significantly more probable but occur in a more limited region.
DOI:10.1109/OCEANSLimerick52467.2023.10244467