Marine Physical Laboratory, Scripps Institution of Oceanography, University of California, San Diego, 9500 Gilman Drive, La Jolla, California 92093-0238, USA.
J Acoust Soc Am. 2010 Sep;128(3):989-1003. doi: 10.1121/1.3466858.
Pressure sensitivity kernels were recently applied to time-reversal acoustics in an attempt to explain the enhanced stability of the time-reversal focal spot [Raghukumar et al., J. Acoust. Soc. Am. 124, 98-112 (2008)]. The theoretical framework developed was also used to derive optimized source functions, closely related to the inverse filter. The use of these optimized source functions results in an inverse filter-like focal spot which is more robust to medium sound speed fluctuations than both time-reversal and the inverse filter. In this paper the theory is applied to experimental data gathered during the Focused Acoustic Fields experiment, conducted in 2005, north of Elba Island in Italy. Sensitivity kernels are calculated using a range-independent sound-speed profile, for a geometry identical to that used in the experiment, and path sensitivities are identified with observed arrivals. The validity of the kernels in tracking time-evolving Green's functions is studied, along with limitations that result from a linearized analysis. An internal wave model is used to generate an ensemble of sound speed profiles, which are then used along with the calculated sensitivity kernels to derive optimized source functions. Focal spots obtained using the observed Green's functions with these optimized source functions are then compared to those obtained using time-reversal and the inverse-filter. It is shown that these functions are able to provide a focal spot superior to time-reversal while being more robust to sound speed fluctuations than the inverse filter or time-reversal.
压力灵敏度核最近被应用于时间反转声学，试图解释时间反转焦斑的增强稳定性[Raghukumar 等人，J. Acoust. Soc. Am. 124, 98-112（2008）]。开发的理论框架还被用于推导优化的源函数，这些源函数与逆滤波器密切相关。使用这些优化的源函数会产生类似于逆滤波器的焦斑，它比时间反转和逆滤波器对介质声速波动更稳健。在本文中，该理论被应用于 2005 年在意大利厄尔巴岛北部进行的聚焦声场实验中收集的实验数据。使用与实验中使用的几何形状相同的无距离相关声速分布来计算灵敏度核，并通过观察到的到达来识别路径灵敏度。研究了核在跟踪时变格林函数方面的有效性，以及线性化分析带来的限制。使用内部波模型生成一组声速分布，然后使用计算出的灵敏度核和这些优化的源函数来推导优化的源函数。然后，将使用这些优化的源函数获得的观察到的格林函数的焦斑与使用时间反转和逆滤波器获得的焦斑进行比较。结果表明，这些函数能够提供比时间反转更好的焦斑，同时比逆滤波器或时间反转对声速波动更稳健。
