The Department of Computing Science and Mathematics presents the following seminars. Unless otherwise stated, seminars will take place in Room 4B94 of the Cottrell Building, University of Stirling from 15.00 to 16.00.
If you would like to give a seminar to the department in future or if you need more information, please contact the seminar organiser Carron Shankland (Phone 01786-467-444, Email ces@cs.stir.ac.uk).
Prof. Ken Turner, University of Stirling
Formal methods tend to get bad press because they are perceived as being abstract, obscure, dry, and basically not very interesting except for the mathematically inclined. This talk aims to show that they can be fun, by using LOTOS on a variety of rather unconventional applications for a formal method.
It will be seen how the classical n-body problem in mechanics can be generalised and formalised in LOTOS. A number of variants are produced by instantiation of the common specification framework. These include a pendulum, Newton's cradle, Brownian motion, gas motion, the `game of life' and an orrery. It will be shown how these are derived from the generic framework using a configuration tool. The resulting LOTOS specifications are simulated automatically to graphically animate the system behaviour. Live demonstrations will be given of the approach.
Prof. Ken Turner, University of Stirling
It is shown how Dill (Digital Logic in Lotos) can be used to specify, verify and test asynchronous (unclocked) hardware designs. New relations for (strong) conformance are defined for assessing a circuit implementation against its specification. An algorithm is also presented for generating and applying implementation tests based on a specification. Tools have been developed for automated verification of conformance and generation of tests. The approach is illustrated with three case studies that explore speed independence, delay insensitivity and testing of sample asynchronous circuits.
Prof. Evan Magill, University of Stirling
The convergence of computing and telecommunications is well documented. One aspect is the concept of programming networks, that is programming networks to carry out value-added services. In a sense the existing telephony networks are programmable. Services such as 1471 and call waiting have been programmed into the network nodes. But it is a clumsy, costly, and certainly slow affair.
Attempts such as the Intelligent Network attempt to alleviate this, but this still shows a rate of service introduction many orders of magnitudes less than that of internet applications. The approach is steeped in the traditions of telephony.
In contrast concepts such as active networks, programmable networks, and mobile agents reflect the computing and IP world. Radically different approaches are advocated that promise a much more rapid introduction of services. This seminar will describe these various approaches and will address the impact on the nature of future telecommunication services engineering.
Dr. Paul Jay, University of Ottawa (Canada)
Over the next few years, the progressive advancements of Moore's Law start to encounter increasingly challenging hurdles within the confines of conventional semiconductor technology. These hurdles are both technological and economic, and place silicon in the position of being vulnerable to a `disruptive technology' upstart that could assume at least part of the conventional semiconductor empire. The talk will consider the technology factors, the economic context, and the changing demands of technology users. With these in mind, the talk will look at the suite of alternative technologies from which a next generation computing platform could select. The speaker wonders whether neural technology would be ready in time for this challenge, or if it might prefer to wait out this dance and try again in about 10 years time!
Dr. Richard Lai, La Trobe University, Melbourne (Australia)
One of the main problems in industrial testing is the enormous number of test cases that can be derived from any complex communication protocol. Due to budget constraints and tight schedule, the number of test cases has to be within a certain limit. However, having a limit raises the following issues. What criteria should be used for selecting the test cases? How can we ensure that important test cases have not been excluded? A good test case is one that has a high probability of finding undiscovered errors. One of the selection criteria is by means of determining the value of a test case. By assigning a value to each of the test cases of a test suite, the relative importance of each of the test cases can be ranked and an optimal test suite can then be designed. The value of a test case should be measured in economic terms, which could be based on the probability that a particular case will occur and the consequence if it fails. In this seminar, a method for assigning a value to each of the test cases of a communication protocol test suite will be presented.
Dr. Sharon Curtis, University of Stirling
Quilting (including the patchwork which the typical quilter indulges in) is primarily a leisure activity, one that seems to have little to do with computing or mathematics. However, quilting yields many varied combinatorial problems concerning patchwork (e.g. aesthetic or practical layout arrangements) and social arrangements (e.g. how to organise a group project to make everything run smoothly). Two examples of such problems will be presented together with their solutions, which were found using such tools as functional programming and latin squares.
David Marples, Telcordia, Morristown (USA)
Networked Appliances are popularly viewed as one of the next major Internet growth areas. An example appliance is an alarm clock that can adjust its wake-up time based on your calendar, current weather and traffic conditions. Another is an Internet-enabled home security system that allows you to see the people approaching your home when you are in the office. Another example, seen in a recent United States TV advertisement, is a refrigerator that reports to a service station when it needs maintenance, without ever needing to inform the owner. The application of Internet technology to appliance devices opens up whole new vistas of exploitation, the extent of which we can only guess at today.
Within this talk, we discuss the current state of the art in Networked Appliances and examine some of the challenges that need to be addressed before they are suitable for the mass market.
Dr. Marwan Al-Akaidi, De Montfort University
Computer speech recognition is an important subject that has been studied for many years. Until relatively recently, classical mathematics and signal processing techniques have played a major role in the development of speech recognition systems. This includes the use of frequency-time analysis, the Wigner transform, applications of wavelets, and a wide range of Artificial Neural Network paradigms. Relatively little attention has been paid to the applications of random scaling fractals for speech recognition. The fractal characterisation of speech waveforms was first reported by Pickover and Al Khorasani who investigated the self-affinity and fractal dimension for human speech in general. They found a fractal dimension of 1.66 using Hurst analysis. In this talk we investigate the use of fractal dimension segmentation for feature extraction and recognition of isolated words. We will discuss a few preliminaries which relate to Speech Recognition techniques, Sampling and Data Extraction. This will include the magnitude of the data and the number of zero crossings in the data. The talk will then consider the use of fractal geometry for segmenting digital signals and images. A method of texture segmentation is introduced which is based on the Fractal Dimension. Using this approach, variations in texture across a signal or image can be characterised in terms of variations in the fractal dimension. By analyzing the spatial fluctuations in fractal dimension obtained using a conventional moving window approach, a digital signal or image can be texture segmented. This is the principle of Fractal Dimension Segmentation (FDS).
To
Computing Science and Mathematics Seminars (Current Series)
To
Computing Science and Mathematics Research
To
Computing Science and Mathematics Home Page