ABSTRACT

TitleAbstract
A Method of Data Analysis Using 200Hz Ocean Acoustic Tomography System
   Ocean Acoustic Tomography (OAT) can observe both mesoscale and large-scale ocean structures, by under water sound techniques and inverse method. Japan Marine science and Technology Center (JAMSTEC) and Oki Electric Industry have been developing the OAT system which consists of 200Hz transceivers and the OAT data analysis soft ware since 1989. JAMSTEC has observed Kuroshio Extension by using five OAT transceivers from July 1997 to September 1997. The area of observation is from 28N to 36N and from143E to 154E. Perturbation of sound-speed dc is modeled by using some orthogonal functions in analysis of OAT. Here, perturbation of sound-speed dc is defined as difference between measured sound-speed and reference sound-speed. Reference sound-speed was made from the data base of the ocean area . In this analysis, we used trigonometric functions in horizontal direction and Empirical Orthogonal Functions (EOFs) in vertical direction to model dc, and used stochastic inverse method. It's concluded that we can monitor moving Kuroshio Extension during these 60 days using the OAT system.

1000km Square Scale Tomography Experiment in the Kuroshio Extension Area Using 200Hz Ocean Acoustic Tomography System
   Ocean acoustic tomography observation was carried out at JAMSTEC(Japan Marine Science and Technology Center) in the Kuroshio Extension region, east of Japan from July to September 1997. In this observation, Five transceivers (consist of sound source and hydrophone array) were used and arranged 28-36 N, 143-154 E aria. The longest distance between transceivers were about 1000km range and moored about 1100m depth. These transceivers were communicated at eight times (in August) or four times (in July and September) a day each other. and a time series of 1000km square propagation data in this observation area were recorded and analyzed. The time series of 3-D temperature fields shows the variability of Kuroshio Extension with high time resolution.
   Category most appropriate for this paper: Global Acoustics, Tomography

Spectrum Focusing of Acoustic Waves in the Ocean
   In this paper the results are presented of studies on the interference effects appearing in low frequency acoustic waves propagation along the ocean wave guide. The problems, associated with the forming of acoustic fields fine interference structure in the stationary laminar media are theoretically studied. The basic characteristics are shown of space and frequency distribution of acoustic fields in wave guides resulted from interference. Spectrum and correlation features are studied of two-dimension interference structures, which were found to
   have the remarkable property of discreteness of their two-dimensional spectrum. A new approach of experimental ocean acoustics is developed based on measuring integral (space and frequency) field characteristics, which, however, bring full information about field interference fine structure. The corresponding methods of direct full-scale acoustic experiments are observed. The methods of investigation of interbeam, intermodal (local) and space interference are delivered. The results of experimental researches in several ocean regions are presented. The experimental data on the interference structure of acoustic fields interference structure in wave guides of various types are given. The important result of researches is the establishment of the main regularity of the characteristics of interference space and frequency distribution of acoustic fields in real ocean wave guides, namely the discrete character of two-dimension spectrum of interference structure. The spectrum discrete features are associated with wave guide mode spectrum (or acoustic rays) and thus with acoustic environment parameters. This feature, known as sound spectrum focusing in wave guides gives new
   possibilities in promoting researches of the ocean and makes acoustic tools more effective. This work was supported by RFFI foundation (Grant 9705-65920).

Design Package for Parametric Transmitting Array
   This paper gives a brief description of a package for simulating the parametrically generated secondary sound fields from a parametric array in water. This is an User Interactive Package which simulates the secondary sound field for user selectable parameters like array shape and size, transmitting power, mean primary frequency and down shift ratios (ratio of difference (secondary) frequency to the mean central primary frequency).
   The parametric array characteristics are expressed in terms of the achievable acoustic source levels and beam patterns at secondary signal for both nearfield (at any range R) and farfield of rectangular arrays. These curves exhibit the optimal specifications (maximum secondary level vs. secondary beamwidth) that may be achieved with a difference frequency. These curves highlight the importance of the proper selection of primary central frequency level and primary acoustic source levels.
   The secondary field in the nearfield can be represented by a triple integral of an oscillating integrand. Numerical solution of the Three dimensional scattering Integral has been obtained by using the automatic global adaptive quadrature routines (D01FCF and the interpolation routines) present in NAG toolbox of the MATLAB for numerical computations.

A New Algorithm for Multipath Time Delay Estimation in Low SNR Using Mile Method
   In this paper, we consider the effects of multipath on a white gaussian random process which might have originated from an acoustic source in the far field, this process is observed along with a white gaussian noise. Maximum likelihood technique (ML) is being used to estimate the time delays caused by a dispersive environment. Our formulation of the new approach is presented below:
    y(n)=x(n)+Sum(k=1,M)[gk x(n-nk)] + e(n),
    0<gM<...<g1<1, n1<n2<...<nM
   where y(n), x(n) and e(n) are the received signal, the source signal and the observation noise, respectively.
   In this paper, we prove that at low SNR, ML technique to estimate time delays corresponds to the technique of maximizing output autocorrelation. Then we present a new algorithm to obtain time delays using autocorrelation maximization.
   The autocorrelation maximization estimates the time delays if we maximize the following:
    s(n1,n2,...,nM) = Sum(k=1,M)[gk Ryy(nk)] + Sum(k=1,M-1)Sum(L=k+1,M)[gkgLRyy(nL-nk)]
    over : n1,n2,...nM

Acoustic Wave Scattering from Rough Surfaces in a Medium with Continuously Varying Density and Sound Speed
    The effect of a transition density layer on the wave propagation in an oceanic environment is studied. The reflection characteristics from a medium containing a density transition layer overlying on a uniform fluid/elastic layer such as seabed are first examined, followed by a study of surface noise propagation in a wave guide environment. The results have shown that the effect of density transition layer is most eminent if there exist no refraction in the medium. Moreover, between two layers enlarged. If the medium contains a sound speed contrast, then the effect of the density transition is minimized, indicating that the refraction largely controls the over effect of the meium stratification.

Shallow-Water Tracking in the Sea of Nazare
   In the summer of 1996, an experiment was conducted off the coast of Portugal to study the effects of internal tides on sound propagation. This experiment, called INTIMATE 96 (Internal Tide Investigation by Means of Acoustic Tomography Experiment), has provided a great deal of insight about the variability of pulse transmission over space and time. In contrast to the traditional view of shallow water propagation as extremely complicated and unpredictable, we find a steady pattern of echoes. The echo-pattern stretches and shrinks in a systematic way with the tides and allows us to infer the components of the first few oceanographic modes. In a separate phase the echo pattern was used to reliably track the source from morning to late evening as it traveled both across the continental shelf and down the continental slope during a period when isotherms in the ocean wavered by 30 m as a result of the tides. We will discuss these acoustic results with emphasis on the source tracking.

An Efficient Sampling Technique of Continuous Distributions for Ring and Elliptical Arrays Synthesis
    The synthesis of planar arrays to produce desired far field patterns is in general a mathematically difficult problem,specially if many constrains are placed on the required pattern such as sidelobe levels,beamwidth and directivity. On the other hand, there exist analytical solutions for continuous fed apertures that yield patterns with desired properties such as apertures with linear and parabolic tapered feeding and Taylor distributions. Thus, there exist a need to convert a continuous aperture distribution into a discrete equivalent. This is known as sampling technique for array synthesis.
    In general, sampling of rectangular aperture distributions is suitable to design planar arrays with certain desired patterns. On the other hand, continuous circular or elliptical aperture distributions can be sampled to design circular, concentric multi-ring or elliptical arrays. The analysis in the case of elliptical arrays can be carried out on the basis of a simple transformation between circular and elliptical apertures.
    In this paper, we introduce some new sampling techniques of continuous circular and elliptical aperture distributions. In these techniques,the aperture is discretized into angular sectors of equal area and each sector is divided into a given number of radial cells. The array elements are placed in the geometrical centers of the cells, resulting in concentric multi-circular, or multi-elliptical,i.e. ring arrays. The feeding coefficients of the corresponding array element is either taken as the actual value of the corresponding continuous aperture feeding (conventional sampling (CS)) or as the average value among the cell area (integrated sampling (IS)). The generally complicated synthesis problem is thus transformed to simply determining the array excitation coefficients. Both IS and CS sampling techniques are examined to obtain the array feeding.
    Comparative numerical results based on different aperture distributions such as, uniform, linear and parabolic tapers as well as the Taylor source distribution have been carried out. The proposed techniques are shown to result in better results for ring array synthesis for most of the considered aperture distributions.

High-Resolution Beamforming by Wigner-Ville Distribution Method
    The Wigner-Ville distribution (WVD) function was originally proposed by E.Wigner in quantum mechanics and J.Ville applied it for a signal analysis. This method has made it possible to represent signal's power density spectrum in time-frequency domain as a natural extension of the Fourier transformation method (FTM). Recently it has attracted great interest for its validity to analyze time-varying signals accomplished by the development of high speed digital signal processing and it is used for analyzing non-stationary signals. Conventionally, sonar beamformer is constructed by using delay lines, but the development of the high speed processor has made it possible to apply the FTM for sonar beamforming. However the resolution of the beamformer is not enough for discriminating small underwater objects on the sea bottom by this method. To solve this problem, we aim to apply the WVD method, which can represent finer structure of signals as a natural extension of the FTM, for sonar beamforming to obtain sharper beam pattern than that of the beamforming method by FTM. Simulation results by the computational calculations clarify that resolution by WVD method presented in this paper becomes about twice as high as by conventional FTM.
   

68 Element Projector - Receiver Array for Parametric Sonar
   A 68 element transmit-receiver array operating around 80 kHz has been designed and fabricated for parametric sonar applications. The projector array consists of sixteen staves of four elements each provide a beam-width of 3.6 degrees in bearing and 15 degrees in azimuth. The elements are made with PZT-6 material to handle large powers. The array transmits dual frequency signals simultaneously in the same direction at high powers. These signals interact in the medium and generates the difference and the sum frequency signals. the difference frequency signal which has the same directivity is now used to detect or visualise the underwater objects. The received signals which range between 1kHz to 5 kHz are picked up on a low frequency broad bandwidth hydrophones.
   The array also has a seventeenth stave of four elements which is used for receiving the signals at primary frequencies. This is required to be used when the array is being used in a conventional or a sector scan sonar thus eliminating the need for a separate receiving array. To have good receiving sensitivity, PZT-5 material is used for these four elements.
   The array has been characterised for its transmitting response and receiving sensitivity. The other array characteristics like impedance, susceptance, conductance have also been measured and reported.

Design Considerations of Parametric Arrays
   This paper is concerned with the design, testing and performance evaluation of Parametric Sonar Transmitting Array to be used in Sub-Bottom Applications. The chief advantage of Parametric Sonar Array for its application as a sub-bottom profiler lies in its ability to reduce absorption in the sea-bed by operating at low frequencies from small transmitting array whilst retaining good signal/reverberation ratio by virtue of the narrow beams and short pulse lengths. The basic disadvantage of the Parametric Array is well known, and it is its low conversion efficiency, which is normally less than 1%, that is equivalent to at least 20 dB power loss. This paper discusses the design considerations in the designing of a Optimum Parametric Transmitter to be used for Sub-Bottom applications. The main factors to be considered while designing a Parametric Sonar are the a.) choice of primary frequency, which basically governs the virtual array length formation, cavitation threshold and the maximum achievable source level; b.) transmit array size, which is a compromise between directivity and maximum transmitted powers and maximum scanning sectors; c.) no of staves, governed by steerability requirements and system complex ability; d.) maximum transmit power, which is decided by the array. Based on these design criteria a Compact Parametric Sonar operating around 40 kHz and capable of generating difference frequencies between 500 Hz to 10 kHz has been designed and developed at IIT Delhi employing a flexible sixteen stave transmitting system. For carrying out the Performance evaluation of the system designed, a comprehensive series of tests were carried out in a fresh water lake to measure the Source levels and the Beam Patterns at the difference frequencies. The system performed satisfactorily to its designed specifications and the results are being presented here.

Optimizing the Directivity Diagram of Underwater Acoustic Wide Band Antennas
    LAMI has been working on the question of echo sounding in continental waters for some years. This work, carried out in cooperation with the hydrobiology laboratory of Toulouse University, has concerned portable and water-tight instrumentation. Place on-board a small boa and coupled to a GPS location system, the instrument allows us to study the aquatic environment by assessing the fish population and localization and making precise bathymetric measurements.
    We are now interested in determining and classifying the nature of lake bottoms for which this system, based on a single frequency method, is not convenient. To obtain much more information about lake bottoms it is necessary to use a wide-band instrument capable of acquiring the corresponding impulse response and then the transfer function. To do this, the instrument needs to operate with a wide-band underwater acoustic antenna and poses the problem of the directivity diagram, which varies with frequency.
    In this article we describe the development of multisensor antennas in a specified band of frequencies using beamforming and processing techniques. This method is based on the spatial positioning of omnidirectional transducers and the use of specific processing in order to maintain the directivity diagram constant over the whole frequency band of interest. The digital signal processing algorithms associated with each transducer are based on RIF filters. The corresponding coefficients of the filters are determined by optimizing the desired antenna directivity diagram. We shall first present the method used, then the theoretical results together with those obtained during the experiment campaign. Finally we shall present some possibilities for making the hardware part of a real-time signal processing system suitable for such an antenna.

Evaluation of a Synthetic Aperture Sonar with a Multi-Aperture Transducer
    This paper presents a Synthetic Aperture Sonar (SAS) with a multi-aperture transducer. SAS is a kind of side-looking sonar which has very fine azimuth resolution by constructing a virtual long line array. Pulse Repetition Frequency (PRF), however, restrict operation of SAS. The multi-aperture SAS can solve this problem, which has a transducer composed of multiple sub-apertures, one of which transmits signals and receive echoes, and the others of which receive echoes. The proposed multi-apertures SAS automatically modifies PRF so that necessary conditions for both maximum search range and azimuth resolution are simultaneously satisfied. The PRF modification makes the ratio of two aperture lengths an integer, each of which is obtained for the above two conditions. Consequently, unnecessary areas on the transducer always appear at the same position, which can be eliminated for a shorter transducer. The phase errors in range curvature compensation can also be reduced by precise adjustment of the round-trip time on each sub-aperture of the transducer. Simulation results indicate that the proposed SAS can eliminate over 40% of the original aperture length, and that it can also reduce ghost targets due to phase errors. Experimental results show that ghost targets appear in sonar images without the proposed method, while only a target echo can be observed using with the proposed method.

Chiral Composites as Underwater Acoustic Attenuators
    The plane wave propagating in effectively chiral composites (ECC) had been asymptotically investigated (Yang et al.: Jpn. J. Appl. Phys. 36 1997, 1201-1208). The analysis represented the reflected and transmitted phenomena of an incident P-wave from an elastic solid against the effective chiral composite. Consequently, an interesting and potentially important phenomenon from the computations was that the P-, SV-, and SH-waves are coupled together in the reflected field. In general, the ECC can be fabricated by embedding the helical arrangement of microstructures in the soft matrix. Therefore, two nondispersive longitudinal wavenumbers and four dispersive circularly polarized transverse wavenumbers can be found from the dispersion equation. The latter also indicated that two transition frequencies divide the frequency spectrum of the transverse wavenumber into three different groups and the four transverse modes can only be distinguished in a specified frequency range. In this work, we illustrate the reflected and transmitted fields of an acoustic plane wave propagating at the fluid-chiral-elastic interfaces. It is observed that, due to the mode conversion of the chirality in the chiral medium, the chiral material should instigate a reducible reflected plane wave, and may be used as an acoustic attenuator to "absorb" sound in underwater.

The Solution of Direct and Inverse Problems of Acoustics in the Sea Wedge
   The purpose of this article is the deduction of the function of Green for the sea wedge, limited by two impedance planes (that is, bottom and air). Currently, no strict solutions exist for the Helmholtz equations for the wedge with the arbitrary distribution of the environment's density and acoustic wave propagation speed. Existing approximate analytic and numeric models don't allow at the same time to correctly solve direct and inverse acoustic problems for the sea wedge. In this work the following strict approach is proposed for the determination of the function of Green for the wedge. The correct and stable algorithm, allowing us to form several "directivity patterns of an antenna" from a single point, has been proposed. It allows dividing the whole angular surveillance sector, (360 degrees) to three linearly independent fields, thus separating the variables for the confined space. Then for each field the problem of diffraction on reflection and refraction of the spherical wave on impedance plane is solved. Summing the three linearly independent solutions, we get the final strict solution of diffraction for the confined space, limited by three planes. Evidently, the function of Green found that way strictly solves both direct and inverse acoustic problem for the arbitrary impedance near waveguide
   surfaces. Numerical computations for the ideal wave-guides fully agree with known literary data.

Acoustic Fields in the Ocean Bottom
    In this paper we attempt to provide some grounding in theory of sound velocity reducing effect for waves propagation in the ocean bottom. It has been found experimentally that when acoustic waves are incident on the bottom at angles greater than critical ones, sound velocity in the bottom reduces to values less than such in the water. Here we derive rigorous generalized formulas for refractive index for spherical and cylindrical waves.These formulas are sufficiently simple. The computation results are sufficient to allow definite conclusions. Unlike plane waves, spherical and Cylindrical waves incident on a plane interface may have two critical angles.Sound waves incident on a bottom at angles placed between the two critical ones are not refracted and penetrate into the bottom. The energy of these waves is 30-35 per cent of such incident wave. Sound velocity in the ocean bottom is similar to such on the waveguide. This provides an explanation for the reducing sound velocity in the ocean bottom.