By Topic

Application of Artificial Intelligence Techniques to Signal Processing , IEE Colloquium on

Date 15 Mar 1989

Filter Results

Displaying Results 1 - 9 of 9
  • A review of applications of artificial intelligence techniques to naval ESM signal processing

    Page(s): 5/1 - 5/5
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (232 KB)  

    The Admiralty Research Establishment (ARE) is supporting research into ESM (electronic support measures) digital signal processing techniques to provide the functions of automatic detection, surveillance and identification of radar emitters for application in future naval ESM systems. The use of artificial intelligence techniques is believed to be the solution to current ESM signal processing problems. Initial ARE work has confirmed this belief and more extensive projects are now being pursued. The value of this software work to the development of operational ESM system is, however, highly dependent on the success of parallel research work to implement such software on real time hardware View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • IEE Colloquium on `The Application of Artificial Intelligence Techniques to Signal Processing' (Digest No.42)

    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (16 KB)  

    The following topics were dealt with: artificial intelligence; signal processing; connectionist techniques; biomedical image processing; and expert systems View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Learning and mathematical reasoning using adaptive signal processing techniques

    Page(s): 2/1 - 2/4
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (160 KB)  

    The MLS (machine learning system) approach offers a more generic solution to the problem of fault diagnosis and adjustments in electronic devices and systems than the classical ESs. The MLS is capable of learning the relationships between the inputs and the outputs of a system by looking at a number of examples which include features and corresponding desired action. Therefore the performance of the MLS depends on how the features have been selected. The authors present a technique for finding the correlation between the selected features, the significance of each individual feature and the interaction between the outcomes. This enables one to modify the feature set and also the training sequence. The main limitations of this approach are (a) the combiners are a linear structure, therefore they can not be used to model nonlinear relationships and (b) the decisions of the DAP are based on the estimates of the correlation matrices. In order to resolve the first problem it is intended to look at nonlinear structures such as neural networks for the implementation of the MLS. The second problem may be solved by either using a large number of training examples or introducing some form of memory into the structure of the DAP so that the present decisions are not only based on the present correlation matrices but also on the previous decisions View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The knowledge-based detection, segmentation, and classification of foetal heart sounds

    Page(s): 6/1 - 6/4
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (172 KB)  

    Presents the implementation of an expert system for the identification of the principal heart sounds in the foetal phonocardiogram. One of the strengths of the system is that there is feedback from the classification process to the lower levels of data abstraction. This allows the weight of contextual information to be brought to bear on areas where local processing was not sufficient to resolve ambiguities. To increase the reliability of the system, particularly when there is a high noise level, work is commencing on integrating information from a different type of transducer. By using two transducers each with a different perspective on the same problem the resulting redundancy of information can be used to combat noise View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Intelligent enhancement of EEG signals

    Page(s): 4/1 - 4/9
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (320 KB)  

    A novel method of removing ocular artefacts from the human electroencephalogram (EEG) in real-time, which makes use of an adaptive signal processing method, has been recently developed. While this method represents a significant improvement on previous methods, it has a number of deficiencies. For example, it does not distinguish between different ocular artefacts or differentiate between them and certain pathological waves of interest. The authors develop an intelligent real-time adaptive system capable of removing a range of different artefacts from biomedical signals, the EEG signals in particular, with minimal distortion of pathological signals. This would allow a more reliable diagnosis of abnormalities and accurate automatic analysis. Artificial intelligence (AI) techniques are being developed to recognise and subsequently classify the pathological waves and artefacts, based on a knowledge of their characteristics, in order to remove the artefacts from the EEG signal when necessary View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Connectionist techniques for signal processing

    Page(s): 3/1 - 3/3
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (116 KB)  

    Neural network or connectionist techniques are currently generating considerable excitement in areas as diverse as financial analysis and cognitive psychology. The author presents an introduction to neural networks and their possible application to signal processing tasks. A neural network is an interconnected structure of many simple nonlinear processing elements or neurons, which learn from examples to form an internal representation of a problem. These processors are often analogue in operation, and by the standards of modern digital circuitry may be very slow. Layers within a network represent degrees of abstraction, and very complicated representations can be constructed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Expert system for patient realignment in MRI [magnetic resonance imaging]

    Page(s): 8/1 - 8/3
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (100 KB)  

    Describes an approach for carrying out automatic patient realignment using a completely separate knowledge based system which calls on a series of low level image/signal analysis tools to extract low level features from the images. These features are compared using the knowledge based expert system to measure and then correct the positional inaccuracies. Higher level features such as the labyrinth and individual sulci can then be extracted and used to fine tune the realignment. While the feature extraction and image analysis operations are implemented in a standard numerical programming language, the symbolic reasoning is carried out using an expert system shell. The authors discuss some of the reasons for adopting this type of approach and describe how breaking the problem up as an intelligent overseer driving a series of low level image processing algorithms, has led and guided the work View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Analysis of linear predictive data such as speech by a class of single-layer connectionist models

    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (24 KB)  

    Summary form only given: A class of single-layer connectionist models is analysed where the nonlinearity includes the logistic function commonly used in error back propagation analysis. It is shown that when the input data can be modelled by a linear predictive or autoregressive process, a solution exists for the weights which minimises the cost function and hence an output error cost function. This establishes the weight sets for single-layer connectionist models whose input data are linear predictive processes, such as speech. Comparisons are made with the equivalent error back propagation analysis and an alternative network structure is described. The method also suggests a form of connectionist vector quantisation (CVQ) for the analysis of speech. A number of experimental results are given View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An overview of AI applied to signal processing: a perspective on coupled systems

    Page(s): 1/1 - 1/4
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (196 KB)  

    Addresses the marriage between two important scientific disciplines namely, artificial intelligence (AI) and signal processing. The authors define AI as the science of building computational models that emulate intelligent problem solving behaviour and signal processing as the science of manipulating numeric signals or data to make explicit information that is contained within the signal. The authors identify, rather than review, the benefits that result from the cooperative use of these two disciplines from a signal processing perspective. In particular, the field of coupled systems is investigated View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.