Underwater acoustics

Background Structure Functions (BSFs) are wavefront distortion metrics, functions of Sound Speed Profiles (SSPs) that are functions of depth. Use of these BSFs is a synthesis form of Matched Field Processing (MFP) that detects signals that are otherwise lost to receivers. Underwater Acoustics (UWA) can use these models to forecast communication and imaging performance and to reduce power radiated into the sea. This reduction of Transmission Loss (TL) occurs because the commercial wavefront control has an input format that accepts BSFs. The BSF plots represent the purely statistical distortion for communications and remote sensing. Another source of TL reduction comes from the enclosed BSF-based phase and phase variance forecasting that protects equalizers from losing phase-lock. Protecting the equalizers protects the Signal To Noise (SNR) ratios. This dissertation derives the UWA version of these metrics and applies them to the following locations of our SSPs: The BSFs use measured, corrected, and verified SSP groups for 132 different locations in the Atlantic Ocean and the Gulf of Mexico from a Navy Ocean Atlas, as well as 64 SSPs in two areas in the littorals, Port Everglades, and Saint Andrew Bay, plus tidal variations. Since BSFs digitize the propagation into one or more segments, our purely statistical phase screen model uses only 3 or 4 degrees of freedom (DOFs) per segment compared to many dozen DOFs for conventional structure functions. The BSFs forecast communications and imaging performance, including range, in locations where acoustic measurements are not available, but SSPs are. A separate algorithm forecasts Gouy phase anomalies from background SSPs, which otherwise requires a priori knowledge of anomaly location and use of Catastrophe theory due to ray theory failure at focuses. Avoiding these anomalies and loss of Phase-Locked Loops (PLLs) also helps maintain SNR and lowers transmission power requirements. Combining with phase parameters and performance forecasts improves UWA propagation efficiency using the background (SSPs). In a spatial version of delay equalization, BSF analysis also produces the enclosed Shear Distortion Ratios (SDRs) for the same locations mentioned above, to allow optimum selection of image enhancement algorithms that mitigate image shear distortion.

A detailed study of a novel method for high-speed acoustic communications in ports and
shallow water is presented. A series of field experiments, coupled with simulated results
using an acoustic channel model have been conducted to outline the optimal modulation
schemes for use in the highly reverberant and Doppler dominated shallow water
acoustic channel. Field experiments were conducted in the vicinity of the SeaTech
marina and the Port Everglades turning basin in water depths of 2 to 15 meters and
ranges of between 25 and 75 meters. An automated FAU acoustic modem transmitted
BPSK and QPSK modulated messages centered at 300 kHz, with a source level of 173
dB re 1pPa and a symbol bandwidth of 25, 50 or 75 kHz. The coded rate varied from
25000 to 150000 bits per second. These high data rates are made possible using a high
resolution Decision Feedback Equalizer with an efficient Doppler compensation process.
The results of this study demonstrate the ability of such a system to transmit video
images in a shallow water environment.

Options for tracking dynamic underwater targets using optical methods is currently limited. This thesis examines optical reflectance intensities utilizing Lambert’s Reflection Model and based on a proposed underwater laser tracking system. Numerical analysis is performed through simulation to determine the detectable light intensities based on relationships between varying inputs such as angle of illumination and target position. Attenuation, noise, and laser beam spreading are included in the analysis. Simulation results suggest optical tracking exhibits complex relationships based on target location and illumination angle. Signal to Noise Ratios are a better indicator of system capabilities than received intensities. Signal reception does not necessarily confirm target capture in a multi-sensor network.

Ocean is home to a large population of marine mammals such as dolphins and whales and concerns over anthropogenic activities in the regions close to their habitants have been
increased. Therefore the ability to detect the presence of these species in the field, to
analyze and classify their vocalization patterns for signs of distress and distortion of their
communication calls will prove to be invaluable in protecting these species. The objective of this research is to investigate methods that automatically detect and classify vocalization patterns of marine mammals. The first work performed is the classification of bottlenose dolphin calls by type. The extraction of salient and distinguishing features from recordings is a major part of this endeavor. To this end, two strategies are evaluated with real datasets provided by Woods Hole Oceanographic Institution: The first strategy is to use contour-based features such as Time-Frequency Parameters and Fourier Descriptors and the second is to employ texture-based features such as Local Binary Patterns (LBP) and Gabor Wavelets. Once dolphin whistle features
are extracted for spectrograms, selection of classification procedures is crucial to the success of the process. For this purpose, the performances of classifiers such as K-Nearest Neighbor, Support Vector Machine, and Sparse Representation Classifier (SRC) are assessed thoroughly, together with those of the underlined feature extractors.

The Federal Highway Administration's development of a new highway noise prediction model (TNM) necessitated the collection of equivalent source height data on moving highway vehicles as a function of frequency. An original method developed by Glegg and Yoon was used in the initial collection of this data. Analysis of this data indicated the measured source height was overestimated at frequencies below 500 Hz. In order to improve the equivalent source height estimates below this frequency two alternative methods were investigated. The first method made use of the coherence function to remove noise from the autospectral density estimate for an array element through the cross spectra of two other elements. This method is called the noise extraction method and was found to be ineffective for this application. However, the second method used matched-field processing, and a significant improvement in the estimated equivalent source heights was achieved for frequencies below 500 Hz.

This thesis presents an experimental analysis of the acoustic signature of an Ocean Explorer class AUV. The experimental analysis consists of three parts. The first part reports the measurements performed in an open water environment at NSWC in Lake Pend Oreille, Idaho. The second part reports on measurements performed at the FAU test tank on a mock model of the AUV and the third part reports the measurements also in the FAU test tank of the AUV under typical operating conditions. The model measurement results were also used to verify the prediction capabilities of a numerical FE model of the AUV using the reciprocity method. The measurements in the FAU tank considered different operating conditions and different mounting of the podule inside the AUV. The podule contains the main mechanical components of the AUV, which are the propulsion motor and the control surface motors. Also considered in these measurements is the influence of the propeller and the influence of covering the aft section of the AUV with a compliant layer. The results of this analysis show that the type of mounting of the podule is not very significant and that significant energy is transferred through the water trapped in between the podule and the hull. Furthermore, the propeller has a significant influence on the acoustic signature since it generates distinct tones. These tones were also observed in the results of the open water measurements.

Experiments have been performed which determined the fatigue crack growth rate (FCGR) of short cracks (a > 0.1mm) for five high strength steels (yield stress 370-570 MPa) in air and in natural seawater with and without cathodic protection. Attention was focused upon Regions I and Il of the classical FCGR-stress intensity range(Delta K) curve with particular consideration of the near-threshold behavior for short cracks. Single edge notch (SEN) three-point bend specimens and a direct current potential drop (DCPD) crack monitoring system were employed, and test parameters simulated offshore structure conditions. The results indicated enhanced FCGR for short cracks compared to macrocracks by 3-20 times in air and 2-6 in seawater free-corroding(FC). Also, the Delta Kth for short cracks was apparently lower than for long ones in both environments. The transition from short to long crack behavior occurred at constant $\Delta$K in each environment (15.6 MPa m in air and 10.0 MPa m in seawater(FC)) irrespective of initial Delta K (Delta K(0)). The transition crack length ranged from 0.25 to 1.6 mm and was inversely proportional to $\Delta$K(0). Scanning electron microscope fractography showed that the mechanism of enhanced crack growth rate was associated with secondary crack (SC) formation in air and SC or inter-granular cracking (or both) in seawater (FC). The enhanced FCGR for short cracks was minimized by polarization to -950 mV(SCE). Through an elastic-plastic fracture mechanics analysis using the J-integral parameter it was found that the influence of plastic deformation at the crack tip was approximately independent of crack length (short versus long), and the linear-elastic fracture mechanics analysis gave a realistic representation for fatigue behavior.

Currently, our acoustic modems are used to communicate underwater to Autonomous Underwater Vehicles AUVs. These modems have only one sensor and can transmit at low data rates (from 200 to 1200 bits per second) using Frequency Shift Keying (FSK) modulation. A two-dimensional array receiver (MillsCross) has been developed to receive underwater signals with more reliability, at a higher data rate (about 30,000 bits per second). This array has been designed to operate with Phase Shift Keying modulated signals. The purpose of this thesis is to design and implement a signal processing software to demodulate and decode FSK signals acquired by the MillsCross. By taking advantage of the spatial gain of the MillsCross receiver array, higher reliability and longer ranges are expected using FSK, in addition to achieving compatibility between the two systems. This software includes a robust synchronization scheme, a spatial and an equalizing filter, a time-window self-adjusting process and the error control decoding.

Autonomous Underwater Vehicles (AUV) rely on acoustics for a number of mission functions such as communications (Acoustic Modem) and vision (Forward and Side Looking Sonars). The AUV acoustic signature (self-noise and vibration) can thus interfere with AUV operations. Additionally, underwater measurements such as turbulence measurements can be contaminated by interference between the AUV generated acoustics pressures and the low pressures of the turbulence. In this thesis a Finite Element and Boundary Element approach is developed to characterize the self-noise (vibration and radiated sound pressure) of a simplified FAU Ocean Explorer AUV. Mechanical excitation from the "podule", which contains the motors for the propulsion and motion control, is assumed in the analysis. The low frequency (less than 1Khz) results are dominated by two types of modes. One type associated with the motion of the "podule" as a rigid body on the vibration isolation supports that connects it to the rest of the AUV structure. The second type is associated with local structural deformations of the "podule", support frame, and AUV hull. Modifying the stiffness of the supports reduces the frequency of the rigid body modes of the "podule", but does not influence the frequencies of the local structural deformations of the "podule" and the rest of the AUV. Decreasing the stiffness of the supports should result in a reduced AUV acoustic signature.