Venugopalan Pallayil

Seabed Characterization Using Acoustic Communication Signals on an Autonomous Underwater Vehicle With a Thin-Line Towed Array

Sediment classification was demonstrated using the self-noise of an autonomous underwater vehicle (AUV) received on a short towed array. The adopted approach was to separate the direct path and the surface- and bottom-reflected signals. Electrical interference from the source was used to verify source receiver separation. The amplitude ratio of the bottom reflected to the direct path signal levels, after compensating for the differences in absorption, spreading losses, and beam patterns, yields the bottom-reflection loss, at the applicable grazing angle. The latter is calculated from the travel time difference between the direct path and bottom-reflected signals. The method is self-calibrating, requiring absolute calibration of neither sound source nor receivers. The definitive isolation of the reflected and direct path signals and the self-calibrating property make this approach robust. The reflection loss may be compared to known seabed models to estimate sediment type.

Pressure compensated fiber laser hydrophone: Modeling and experimentation

A pressure compensated metal diaphragm based fiber laser hydrophone configuration that can provide good sensitivity, large bandwidth, and sea state zero noise floor is proposed in this paper. A simplified theoretical model of the proposed sensor configuration is developed in which the acoustic elements of the sensor configuration are modeled using a four-pole acoustic transfer matrix and the structural elements are modeled as second order single degree of freedom elements. This model is then used to optimize the design parameters of the sensor system to achieve the performance objectives. An axisymmetric finite element analysis of the sensor configuration is also carried out to validate the results from the simplified theoretical model. Prototype sensors were fabricated and hydrostatic testing in a pressure vessel validated the static pressure compensation performance of the sensor. Frequency dependent sensitivity of the sensor system was measured through acoustic testing in a water tank. The prototype sensor gave a flat frequency response up to 5 kHz and experimental results compared well with theoretical predictions. The sensor has an acceleration rejection figure on the order of 0 dB ref 1 m/s(2) Pa and the pressure compensation approach worked reasonably well up to a hydrostatic pressures equivalent to a depth of 50 m.

Development of a Second-Generation Underwater Acoustic Ambient Noise Imaging Camera

A nominally circular 2-D broadband acoustic array of 1.3-m diameter, comprising 508 sensors and associated electronics, was designed, built, and tested for ambient noise imaging (ANI) potential in Singapore waters. The system, named Remotely Operated Mobile Ambient Noise Imaging System (ROMANIS), operates over 25-85 kHz, streaming real-time data at 1.6 Gb/s over a fiber optic link. By using sensors that are much larger than half-wavelength at the highest frequency of interest, so with some directionality, good beamforming performance is obtained with a small number of sensors compared to a conventional half-wavelength-spaced array. A data acquisition system consisting of eight single-board computers enables synchronous data collection from all 508 sensors. A dry-coupled neoprene cover is used to encapsulate the ceramic elements as an alternative to potting or oil filling, for easier maintenance. Beamforming is performed in real-time using parallel computing on a graphics processing unit (GPU). Experiments conducted in Singapore waters yielded images of underwater objects at much larger ranges and with better resolution than any previous ANI system. Although ROMANIS was designed for ANI, the array may be valuable in many other applications requiring a broadband underwater acoustic receiving array.

A PGC demodulation based on differential-cross-multiplying (DCM) and arctangent (ATAN) algorithm with low harmonic distortion and high stability

A novel phase generated carrier (PGC) demodulation algorithm based on differential-cross-multiplying (DCM) and arctangent function (ATAN) is proposed in this paper. We investigate the stability with light intensity disturbance (LID) and the harmonic distortion due to nonlinearity both theoretically and experimentally. The nonlinearity of the proposed PGC demodulation algorithm has been analyzed. Compared with the traditional PGC-arctan and PGC-DCM algorithm, this PGC algorithm inherently has much lower total harmonic distortion (THD) as well as high stability with LID.

Ambient Noise Imaging through joint source localization

Underwater Ambient Noise Imaging (ANI) systems rely on the acoustic illumination produced by natural noise sources to image an object of interest. Snapping shrimp are a dominant natural source of illumination in tropical waters and their snaps occur randomly. Hence incoherent energy detection methods, which require no knowledge of the source locations, are usually employed to form images of the objects. This approach, although simple, only produces images when the anisotropy in ambient noise is conducive. Even in anisotropic noise, some sources ‘illuminate’ the target while others obscure it. In this paper we describe a different approach. We use a snap detection algorithm to estimate the locations of the noise sources on the sea-bottom and then use the sound from these sources to passively range and form images of the objects. By using only the noise from sources that provide us good illumination and rejecting undesirable sources, we improve the image quality. The feasibility of this approach has been experimentally demonstrated with the data collected during a recent deployment of ROMANIS, an ANI camera developed in Singapore.

A Digital Thin Line Towed Array for Small Autonomous Underwater Platforms

Conventional towed arrays have been built with large diameters to allow separation of the boundary layer from the array elements so that the effect of turbulent flow has less impact on the signal to noise ratio performance of the array. These arrays, therefore, tend to be bulky and heavy. The successful development of AUV and USV required the focus of towed array development to much smaller diameter and light-weight arrays. The Acoustic Research Laboratory (ARL) of the Tropical Marine Science Institute, National University of Singapore, arising from a recent requirement, has developed a light weight digital thin line towed array (DTLTA) for underwater sensing applications. The 10 mm diameter and 12 m long array weighing not more than 2 kg (excluding the tow cable) is a promising sensor platform for use from small autonomous assets such as AUV and USV and is believed to be the smallest diameter digital towed array ever developed. The digital output enables easy interfacing of the array to any micro-controller or PC-based data acquisition platform. A set of diagnostic and beamforming software tools were developed along with the DTLTA to help test the array. These tools are based on conventional beamforming techniques and Ronald Wagstaffs towed array diagnostics. In this paper we are presenting the details of the design and construction of the array, special features of the software tools developed and results from a tow test conducted using an AUV platform in local waters. Limitations of the current design and future development plans to improve upon them will also be discussed.

Fibre channel storage area network design for an acoustic camera system with 1.6 Gbits/s bandwidth

Driven by Gigabit-rate fibre channel (FC) networking capability, storage area networks (SAN) have emerged as a high performance interface that can interconnect multiple servers and storage devices with high bandwidth, high availability and high fault tolerance. Most FC SAN designs employ fabric switches to retime and manage traffic. This is a relatively expensive approach but provides the robustness demanded by workstation network users. We present the application of FC-AL (arbitrated loop)-based SAN technology to a novel problem of constructing a data management system for a broadband acoustic camera demanding 1.6 Gbit/s data flow rate. The needs of such a dedicated SAN system are quite different from that of a generic computing network. Our SAN consists of Intel embedded processors running Windows NT4.0 and Seagate FC hard drives. In order to improve the end user throughput, the mechanisms of FC I/O systems are examined and major factors affecting FC-AL performance have been analysed. Furthermore, we have observed the application performance of FC-AL systems employing different types of host bus adapters (HBA) and CPUs. Finally, test results are presented and evaluated.

Broadband acoustic reflectivity and its application to the characterisation of materials

Underwater acoustic imaging and classification sonars are being progressively extended into broadband and interest is shifting to higher frequencies for use in shallow waters with particular emphasis on object classification for mine counter-measures. These shifts raise the opportunity to use the frequency-dependent scattering properties of different materials to characterise target composition. Many mechanisms may play a role in the total backscattered signal, including specular reflections, scattering from surface irregularities and multiple resonances (often modified by internal structures). These contain a great deal of unexploited information that can tell us a lot about the nature of the object. The acoustic backscattering properties of an object depend to a large extent on the materials used on the object surface and on its structure, yet little is documented about the broadband reflective properties of common marine materials such as rubber foams, Aluminium, Steel, etc., in open literature. In this paper, we explore the reflective properties of these materials over a range of frequencies and incident angles through direct measurement via ensonification by a broadband source in an acoustic tank. An experimental set up is described that can record the reflected energy over a several discrete angles simultaneously. Compact source pulses are used to permit multiple reflections from tank walls and the free surface to be separated out

Detecting the direction of arrival and time of arrival of impulsive transient signals

The ability to detect the direction of arrival and time of arrival (DoA-ToA) of propagating signals generated by underwater sources using array of sensors is crucial for passive sensing. This is a challenging task especially for transient and impulsive signals which have similar acoustic signature. We consider these signals, received by each sensor in a sensor array over an observation period. We assume that only a small number of these signals exist in the observation window, and that they have distinct direction of arrival (DoA) and time of arrival (DoA). The sensor array data of these signals can be transformed into a DoA-ToA space and we should expect this to be sparse. We show that this transformation can be written as an underdetermined linear system. We suggest robust methods to recover the DoA-ToA of the signals by enforcing sparsity in the DoA-ToA space. Through receiver operating characteristic (ROC) analysis, we show in numerical simulations that our methods outperform conventional practices such as cross-correlation-based time difference of arrival (TDoA) and beamforming. We present a scatter plot of the detected DoA-ToA based on the acoustic pressure sensor array recording in Singapore waters to show that the sparse DoA-ToA technique is robust and produces imaging results that match the known underwater man-made structures at the experimental site.

Analysis of effective signal design for active sensing of undersea objects/bottoms in tropical shallow waters

Precise understanding of the sea bottom characteristics and accurate identification of objects on the seabed are critical in many underwater applications. The severe multi-path along with the time varying and random reflections from the surface and the bottom of shallow waters, present an interesting and complex signal processing problem. A thin line hydrophone array, deployed from small Autonomous Underwater Vehicles (AUV) provides significant operational advantage in terms of accessibility in shallow coastal areas and does open up substantial application possibilities. The shallow tropical waters in general present high ambient noise levels due to high density snapping shrimp beds and shipping. This often requires that for getting a reasonably good Signal to Noise Ratio (SNR) the acoustic transmissions have to be powerful to combat the noise level. However, this is not a good option for many battery powered platforms like an AUV as it heavily limits their operational endurance. An alternative is to design specific sonar signals which have high correlation properties and can provide good detection performance under severe noise conditions. Although the sonar waveform design is not new, such application specific signal processing attempt is not reported in the literature. The work will present a two-stage, Non-Linear Frequency Modulated (NLFM) signal design for active undersea sensing using a Digital Thin Line Array (DTLA). The signal design has been undertaken for the frequency band 3-10 kHz, with variations in center frequency and bandwidths. The different chirp rates in a two-stage LFM signal and non-LFM were evaluated for their effectiveness in suppressing the side lobes. The bottom reflected signal has been correlated with the direct path signal, to allow minimal difference between the received signal and the receive filter characteristics, due to the underwater channel. The well-known shallow water propagation loss model proposed by Rogers (1974) was used to simulate the underwater channel conditions in shallow water conditions. The performance metric was based on the ability to minimize the integrated side lobe level metric (ISL) of the normalized correlation function. Bottom types comprising sand and silt have been used in the simulated channel model to compare the performance.