1,721,166 research outputs found
Dolphin-inspired target detection for sonar and radar
No full textGas bubbles in the ocean are produced by breaking waves, rainfall, methane seeps, exsolution, and a range of biological processes including decomposition, photosynthesis, respiration and digestion. However one biological process that produces particularly dense clouds of large bubbles, is bubble netting. This is practiced by several species of cetacean. Given their propensity to use acoustics, and the powerful acoustical attenuation and scattering that bubbles can cause, the relationship between sound and bubble nets is intriguing. It has been postulated that humpback whales produce ‘walls of sound’ at audio frequencies in their bubble nets, trapping prey. Dolphins, on the other hand, use high frequency acoustics for echolocation. This begs the question of whether, in producing bubble nets, they are generating echolocation clutter that potentially helps prey avoid detection (as their bubble nets would do with man-made sonar), or whether they have developed sonar techniques to detect prey within such bubble nets and distinguish it from clutter. Possible sonar schemes that could detect targets in bubble clouds are proposed, and shown to work both in the laboratory and at sea. Following this, similar radar schemes are proposed for the detection of buried explosives and catastrophe victims, and successful laboratory tests are undertaken
Non-linear techniques for estimating non-stationary spatial spectra
No full textThe problem of estimating the distribution of energy arriving at an array as a function of bearing is central to many array processing applications. To form such spatial spectral estimates, data is collected over an interval of time. If the geometry between the source and receiver varies over that interval then the spatial spectral estimator will lose resolution. This work describes a methodology for mitigating this loss based on employing higher order correlation matrices that are robust to such non-stationarities
Adaptive signal processing and its application to infrared detector systems
No full textThis thesis deals with several general aspects of adaptive filtering as well as the application of adaptive techniques to specific problems associated with infrared detectors. Two infrared detection problems provide the motivation for the majority of the work considered herein. The first problem is that of reducing microphony, i.e. vibration induced signals, in a Pyroelectric detector. The approach proposed in this thesis uses a form of adaptive noise cancellation. The second is that of enhancing/detecting pulse-like periodic signals in broadband noise. Once again an adaptive solution is sought to this problem. With these objectives in mind a general study of adaptive filtering is presented. The classical Least Mean Squares (LMS) algorithm is discussed. Its performance in the microphony cancellation problem is limited by the large eigenvalue spread of the data, which results in slow convergence times. The alternative approaches of Gradient Adaptive Lattices (GALs) and Exact Least Squares (ELS) methods are considered. The ELS algorithms suffer from two problems; a large computational burden and numerical instability. The computational loading can be reduced by the use of `fast' ELS algorithms but this only serves to exacerbate the numerical instability. This thesis makes a minor study of this instability and proposes a new algorithm which solves this problem, only in part. The LMS algorithm in the line enhancement problem suffers from poor performance due to the noisy nature of the updates used. Several methods are proposed for reducing this noise, including the use of structures containing two adaptive filters. Two of these structuresare highlighted; the error filter routine and the balancedfilter. Attempts are made to analyse the performance of both of thesestructures but the majority of the theoretical results pertain to the balanced filter. It is shown that such a filter requires constraints and the consequence of applying these constraints is examined. The adaptive algorithm used to update the balanced filter is analysed and it is shown that convergence to the constrained optimum occurs. Unfortunately, this constrained optimum possesses some undesirable properties and as a result the balanced filter is of limited use. Finally, the application of these techniques to data from actual infrared systems is performed. The results from the off-linemicrophony simulations are encouraging. Although, ideally, one would like to apply an alternative algorithm, the pragmatic issues of available hardware and expense limit on-line applications to using the LMS algorithm. The off-line simulations for the line enhancement problem indicate that the balanced filter has good detection capabilities but the problems already mentioned make it unattractive. Once again the LMS algorithm is implemented in real time to perform this task.</p
Non-parametric techniques for the estimation of spatial spectra in non-stationary environments
No full textA novel approach to the problem of estimating a spatial spectrum in a non-stationary environment has been presented. The approach adopted relies on the same principle as that which underpins the Wigner distribution. The proposed methodology results in the construction of a new sufficient statistic that can be used in conjunction with a variety of existing array-processing techniques. The resulting schemes are shown to be robust to source motions. Some of the limitations of this approach have been discussed and how they can be overcome is demonstrated
Comparison of theories for acoustic wave propagation in gassy marine sediments
No full textMore than three decades ago, Anderson and Hampton [1, 2] (A&H) presented theories for wave propagation in gassy water, saturated sediments and gassy sediments in their two part review, which has been cited by many researchers in the geoacoustics and underwater acoustics areas. They gave an empirical formulation based on the theory of Spitzer [3] for the wave propagation in gassy water by adapting that for a viscoelastic, lossy medium. Following Leighton [4], this paper presents a theory based on non-stationary nonlinear dynamics of spherical gas bubbles and extends that 2007 paper to include liquid compressibility and thermal damping effects. The paper then shows how that nonlinear formulation can be reduced to the linear limit, and derives the expressions for the damping coefficients, the scattering cross section, the speed of sound and the attenuation, and compares these with the A&H theory. The current formulation has certain advantages over A&H theory such as implementing an energy conservation based nonlinear model for the gas pressure inside the bubble, having no sign ambiguity for the speed of sound formula (which is important when estimating the bubble void fraction) and correcting the ambiguity on the expression for scattering cross section, as identified in the recent work of Ainslie and Leighton [5]. Moreover, the theory presented here forms a basis for a nonlinear, time-dependent acoustic estimation model for gas bubble distributions in viscoelastic mediums since it avoids the commonly encountered assumptions on the bubble dynamics such as linearity, steady-state behaviour and monochromaticity
Estimation of spatial spectra in a non-stationary environment
No full textThe problem of computing the time-bearing representation for data from a sensor array is considered. A novel approach to dealing with situations where the sources are mobile is presented. The methodology adopted is based on ideas borrowed from the field of time-frequency analysis. Through simulations it is demonstrated that the new approach produces time-bearing plots with greater resolution that conventional methods. Further it is demonstrated that this approach yields bearing estimates with smaller variances than can be achieved using conventional techniques
Independent component analysis using Gaussian mixture models
No full textThis paper discusses a method for performing independent component analysis exploiting Gaussian mixture models (GMMs). Previously most techniques that combine these method shave used GMMs to model the source signals. This paper considers a parsimonious method for modelling the observed signals. The GMM is fitted to the observed data using a modified version of the expectation maximisation algorithm
- …
