Accuracy of the standard-target sonar-calibration method, and criteria for accepting the results of a calibration

The standard-target method of calibrating active sonars used to measure or image backscattering aspires to an accuracy of ±0.1 dB. This is an order-of-magnitude better than that of conventional, analytical calibrations in which the sonar is viewed as a collection of parts and the performance of each part is measured separately, often using standard hydrophones or standard acoustic sources requiring their own calibrations, with inevitable compounding of errors. In contrast, the holistic approach of the standard-target method, which views the sonar as a black box with input and output channels, enables both precise and accurate calibration through the fiducial of a standard target. This is typically a compact object whose acoustic scattering properties are known a priori or a posteriori. The claim of sonar measurement accuracy is supported by reference to methods establishing the target strength of the standard target, accurate to within ±0.01 dB, and repeated sonar measurements with different standard targets performed during the same calibration exercise, observed to be accurate to within ±0.1 dB. Criteria for accepting calibration results are more tolerant, but this reflects the complexity of the sonar as a collection of mechanical and electrical parts, need to suspend the standard target at known positions in the transmit and receive beams, and environmental variability over time and place. Nonetheless, time series of calibration results support criteria based on observed variability in gain parameters ranging from ±0.1dB at 38 kHz to ±0.5 dB at 200 kHz.