"Quasi-Blind" Calibration of an Array of Acoustic Vector-Sensors That Are Subject to Gain Errors/Mis-Location/Mis-Orientation

• Published in: IEEE transactions on signal processing Vol. 62; no. 9; pp. 2330 - 2344
• Main Authors: Yang Song, Wong, Kainam Thomas, Fangjiong Chen
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.05.2014; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Acoustic interferometry; acoustical signal processing; Acoustics; Algorithms; Applied sciences; array signal processing; Arrays; Calibration; Computational efficiency; Computer simulation; Detection, estimation, filtering, equalization, prediction; direction of arrival estimation; Effectiveness; Exact sciences and technology; Gain; Information, signal and communications theory; Manifolds; Miscellaneous; Narrowband; phased arrays; Signal and communications theory; Signal processing; Signal processing algorithms; Signal, noise; sonar arrays; sonar signal processing; Telecommunications and information theory; underwater acoustic arrays; Vectors; Direction-of-arrival estimation; Array signal processing; Acoustic interferometry; Signal estimation; Radiolocalisation; Sonar signal processing; Velocity distribution; Phased array antenna; Learning; phased arrays; direction of arrival estimation; Localization; Antenna array; acoustical signal processing; Interferometry; Monte Carlo method; Speed; Acoustic measurement; Sonar; Measurement sensor; Calibration; Algorithm; underwater acoustic arrays; sonar arrays; Arrival angle; Signal processing; Numerical simulation; Acoustic sensor; Maximum likelihood; Underwater acoustics
• ISSN: 1053-587X; 1941-0476
• DOI: 10.1109/TSP.2014.2307837

This paper advances a new "quasi-blind" calibration algorithm to calibrate a multi-array network (MAN) of acoustic-vector-sensors, whose component-sensors may have non-ideal gain/phase responses, incorrect orientations, and imprecise locations. This proposed calibration is "quasi-blind" in not requiring any prior knowledge/estimation of any training signal's arrival-angle. This proposed algorithm is computationally orders-of-magnitude more efficient than maximum-likelihood estimation. These advantages are achieved here by exploiting the acoustic vector-sensor's quintessential characters, to interplay between two complementary approaches of direction-finding: (1) customary interferometry between vector-sensors, and (2) "acoustic particle-velocity-field normalization" DOA-estimation within each individual vector-sensor. Monte Carlo simulations verify the proposed algorithm's efficacy in "quasi-blind" calibration and its aforementioned computational efficacy.