The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The Fano equality is joined with the data-processing inequality to develop a theory model for component level trade studies within radar signature exploitation systems. Entropy is used to represent propagating uncertainty within an information channel. Measures are developed to identify information flow bottlenecks within an information loss budget. The propagating effects of various sources of uncertainty...
Some of the wireless standards, e.g. IEEE 802.16-2004, allow successive transmission of a number of consecutive slots, where each of them contains a training (pilot) interval. If a number of neighbouring cells asynchronously transmit similar frames, then the resulting interference environment becomes similar to a distributed training scenario. In this paper, a distributed training scenario is addressed...
This paper outlines the use of an Evolutionary Algorithm (EA) to perform the Equalisation of a non minimum phase channel. Conventional techniques utilising first and second order approximations of the error surface, have been demonstrated to be ineffective in achieving an optimal solution in continuous simulations, and have proved incapable of dealing with the more difficult non minumum phase problems...
A method of incorporating implementation aspects in the algorithm-level design of nonlinear filters is proposed. As a case study, the trade-off between the visual properties and the complexity of soft morphological filters is studied using training-based optimization methods. Specifically, it is shown that the use of the complexity constraints can provide the filter designer valuable information on...
In this paper, we present several new robust isolated word speech recognition systems which employ FMQ/MQ as the spectral labelling process, followed by a Hidden Markov Model (HMM), or a HMM and Neural Network (HMM/MLP) classification technique. The ISWR systems provide selective input data to a neural network in response to speech signal to acoustic noise ratios to improve speech recognition system...
We investigate the identifiability conditions for blind and semi-blind FIR multichannel estimation in terms of channel characteristics, data length and input symbol excitation modes. Parameters are identifiable if they are determined uniquely by the probability distribution of the data. Two models are presented: in the deterministic model, both channel coefficients and input symbols are considered...
In previous works [5], [6], we studied some speech enhancement algorithms based on the iterative Wiener filtering method due to Lim-Oppenheim [2], where the AR spectral estimation of the speech is carried out using a second-order analysis. But in our algorithms we consider an AR estimation by means of cumulant analysis. This work extends some preceding papers due to the authors: a cumulant-based Wiener...
The three-class recognition problem of respiratory sounds based on multi-stage decisions is addressed. The method consists of dividing respiratory cycles of patients into phases, and classifying each phase with a separate multilayer perceptron, called the “phase expert”. Each phase information consists of several time segments and their parametric representation. Expert decisions on phase segments...
The purpose of this paper is to propose the design and the use of a Neural Network for model order selection The proposed neural network learns from real life situation by constructing an input/output mapping (for detection) which brings to mind the notion of non parametric statistical inference. Such a strategy can improve performances of traditional tests relying on linearity, stationarity and second...
The paper describes the use of associative models for integrating different sensors. Integrated associative structures are outlined and related to previous approaches; the enhanced robustness resulting from the integration of Associative Memories (AMs) and Neural Networks (NNs) is shown. Discussion then focuses on how different information sources can cooperate on associative visual recognition. Experimental...
A two-dimensional adaptive nonlinear filter, called 2-D FIR-PWL filter is introduced for noise cancellation from images. It is based on the cascade of a linear FIR filter and a piecewise-linear interpolating function. Experimental results show a very good behaviour of the filter, which outperforms in many application examples the Sigma filter both in terms of visual quality and numerical results.
A relatively underexplored question in fMRI is whether there are intrinsic differences in terms of signal composition patterns that can effectively characterize and differentiate task-based and resting state fMRI (tfMRI or rsfMRI) signals. In this paper, we propose a novel two-stage sparse representation framework to examine the fundamental difference between tfMRI and rsfMRI signals. In the first...
We explore techniques to improve the robustness of small-footprint keyword spotting models based on deep neural networks (DNNs) in the presence of background noise and in far-field conditions. We find that system performance can be improved significantly, with relative improvements up to 75% in far-field conditions, by employing a combination of multi-style training and a proposed novel formulation...
In recent years recurrent neural network language models (RNNLMs) have been successfully applied to a range of tasks including speech recognition. However, an important issue that limits the quantity of data used, and their possible application areas, is the computational cost in training. A signi??cant part of this cost is associated with the softmax function at the output layer, as this requires...
Pitch information is an important cue for speech separation. However, pitch estimation in noisy condition is also a task as challenging as speech separation. In this paper, we propose a supervised learning architecture which combines these two problems concisely. The proposed algorithm is based on deep stacking network (DSN) which provides a method of stacking simple processing modules in building...
Model-based single-channel source separation (SCSS) is an ill-posed problem requiring source-specific prior knowledge. In this paper, we use representation learning and compare general stochastic networks (GSNs), Gauss Bernoulli restricted Boltzmann machines (GBRBMs), conditional Gauss Bernoulli restricted Boltzmann machines (CGBRBMs), and higher order contractive autoencoders (HCAEs) for modeling...
Despite recent advancements in digital signal processing technology for cochlear implant (CI) devices, there still remains a significant gap between speech identification performance of CI users in reverberation compared to that in anechoic quiet conditions. Alternatively, automatic speech recognition (ASR) systems have seen significant improvements in recent years resulting in robust speech recognition...
In this work, we consider enhancing a target speech from a singlechannel noisy observation corrupted by non-stationary noises at low signal-to-noise ratios (SNRs). We take a classification-based approach, where the objective is to estimate an Ideal Binary Mask (IBM) that classifies each time-frequency (T-F) unit of the noisy observation into one of the two categories: speech-dominant unit or noise-dominant...
This paper presents an investigation into the detection and classification of drum sounds in polyphonic music and drum loops using non-negative matrix deconvolution (NMD) and the Itakura Saito divergence. The Itakura Saito divergence has recently been proposed as especially appropriate for decomposing audio spectra due to the fact that it is scale invariant, but it has not yet been widely adopted...
We introduce an unsupervised optimization method for optimal fusion of multiple classifiers in retrieval problems. The method is based on a ranking loss called the “clarity” index, which does not depend on the label of the test instances. The technique optimizes the weights with which individual classifier scores must be combined to maximize this clarity. Our method is instance-specific; the weights...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.