Travel-time Sensitivity Kernel Method For Ocean Acoustic Tomography

In 1979, W.Munk proposed the method of ocean acoustic tomography, using acoustic propagation to monitoring the meso-scale ocean. This method is aimed to solve the inverse media problem by estimating the sound speed and current in the observed area, which exploits acoustic measurements like travel time. In general, acoustic tomography is a multidimensional problem which associated with ocean survey, acoustic propagation model and acoustic inverse problem. Conventional acoustic tomography is based on ray theory, also perturbation method, academically. This method uses ray travel time between sources and receivers to inverse the sound speed profile under vertical slice. The linear integral relation between ray path and slowness (reciprocal of sound speed) forms the basic function of acoustic inverse problem, which can be solved by method like least square.However, when it comes to the situation of low frequency and deep sea, perturbation method may lead to low resolution and precision, due to the restriction of both resolution of travel time measurement and high frequency approximate of ray theory. Here, we propose the method of travel time sensitivity kernel to derive the three dimensional integral relation between sound speed perturbation and travel time upon wave theory and Born approximate, which can realize the estimation of sound speed perturbation in arbitrary position with a two nodes transmission. Further, travel time sensitivity kernel method uses peak arrivals as a substitute for multipath wave packet, which increases the information observed.In addition, travel time sensitivity kernel changes the integral space from one dimension on the ray path into three dimension in the space, which increases the spatial resolution.With regard to the travel time sensitivity kernel method tomography, we study the theoretical derivation of travel time sensitivity kernel, peak arrival detection and also solution to the inverse problem of acoustic tomography. First, we derive the three dimensional expression of travel time sensitivitity kernel using normal mode model for broadband signal. Then, we propose a waveform template matching method based on correlation, which can find out the peak arrival pairs effectively. At last, we try to solve the acoustic inverse problem under various ocean environment using methods like least square method with smoothing constraint, truncated singular value decomposition method and convex optimization method.In this paper, we present a series of simulations of ocean acoustic tomography based on perturbation method and travel time sensitivity kernel. Comparisons are conducted between this two methods, which manifest the high resolution and high accuracy of the proposed one. Further more, an ocean acoustic tomography software is developed according to the research above.