|
DETI Broadband Flagship
Derry City was designated by the Department for Enterprise, Trade and Investment as the location for the Northern Ireland Flagship Project - the demonstration of Wireless Broadband Technology. The purpose of the project was to implement a wireless enabled working environment in the City for students, lecturers, tourists, City Councillors and local government officers and provide an experimental testbed for wireless technology. The project launched three innovative pilot initiatives: Wireless Campus: this saw the University of Ulster, Magee campus and North West Institute of Further and Higher Education becoming the most advanced wireless educational establishments at that time in Ireland. Over 200 students were provided with Tablet PCs in September 2005 and Sept 2006, allowing them to work wirelessly wherever they were located - lecture theatres, laboratories, library or student union or campus lawns. Wireless Council: this permitted the implementation of a wireless network in the Historic Guildhall in Derry. Elected officials now operate in a working environment with electonic sources of council documents and have access to a state of the art Virtual Private Network allowing access to the Council's network from a remote location. Wireless Walls: the historic walled area of Derry become a wireless network environment for businesses and tourists. Tourists are now able to experience the rich historical and cultural heritage of the city through hand held PDAs which bring alive the story of Derry through a diverse range of digital media including video footage, soundscapes and photographic images.
For more information see the WirelessDerry website at
http://www.wirelessderry.com/
Period: August 2004 - December 2006
Type: Research
Status: Completed
Funding Body: DETI
Funding to the ISRC: £800,000
Personnel Involved
Professor TM McGinnity, Principal Investigator
Mr MJ Callaghan, Investigator
Dr J Harkin, Investigator
Mr DN Woods, Investigator
INTERWAVE
The basic processing units within the brain are neurons interconnected in a highly complex pattern. Neurons transform their input signals into a sequence of electrical pulses/spikes that are transmitted along the axon to many other neurons, where it is thought that the exact timing of these spikes convey information. This realisation has stimulated significant research into the development of “engineering equivalent models”. However, current topologies used to model biological networks are proving difficult to implement in hardware, owing to the complexity of interconnect required. Interconnect currently uses metal tracks to convey information (spikes) between neurons but at the expense of silicon surface area. It is widely accepted that the silicon surface area consumed by metal tracks increases rapidly with the number of neurons and eventually limits the size of the networks, even if multiple metal layers are employed. For example, it has been reported that for a fully connected feed-forward topology the silicon surface area consumed by interconnect may approach 90% of the total silicon surface for networks of moderate complexity.
Under this project the ISRC investigated a completely radical approach to the implementation of high density spiking neural networks in VLSI. The project performed research into the use of EM ( electromagnetic waves) to propagate “temporal events” between multiple neurons on a chip ie the use of EM waves to propagate spike events between neurons. The project examined the use of customised neurons designed to have output/input antennas from which to output/receive EM waves transmitted/received as the result of a spike events. Many challenging research issues emerged and whioe theoreticlaly promising it is clear that a number of technical and IC fabrication issues have to be resolved before the potential of this transmission medium can be realised. However, the potential of this approach is enormous as it completely eradicates the interconnect problem and thus allows for the first time ever the possibility of unlimited scaling of neural networks in hardware.
Period: October 2003 - March 2007
Type: Research
Status: Completed
Funding Body: North South Programme for Collaborative Research
Funding Value: Euro 282,000
Personnel Involved
Professor TM McGinnity, Grant Holder
Mr PM Kelly, Grant Holder
Dr LJ McDaid, Grant Holder
Dr JA Santos, Grant Holder
Sense Maker
A Multi-sensory, Task-specific Adaptable Perception System.
Details of this project can be found here:
Sense Maker Project Overview
Recent news articles about the Sensemaker project:
SenseMaker in the News
Funded by Information Society Technologies Programme as part of the Life- Like Perception Systems Website.
 
Period: June 2002 - October 2005
Type: Research
Status: Completed
Funding Body: EC
Funding Value: 1.5 MEuros
Personnel Involved
Professor TM McGinnity, Grant Holder
Dr LP Maguire, Grant Holder
ProculTech - INVESTNI Proof of Concept
The project aimed to develop a software and hardware environment to provide a remote-access technical facility for embedded computational systems on a commercial basis, enabling learners to access and control specialised equipment and instrumentation over the Internet and perform experiments on embedded systems.The project built on the highly successful preliminary work developed during an EPSRC funded project (DIESEL) addressing this issue in a university context. The remote technical facility provides a suite of embedded systems experiments to supplement advanced lecture material presented within current academic and continuing professional training courses. Learners can access the remote technical facility for lecture notes on courses and in addition, perform the necessary experiments over the Internet. This provides distance-learning users with the capability to gain knowledge and training in embedded systems with the added value of essential practical experience.
Period: September 2004 - December 2005
Type: Research
Status: Completed
Funding Body: InvestNI
Funding Value: £145K
Personnel Involved
Professor TM McGinnity, Grant Holder
Dr J Harkin, Grant Holder
Mr MJ Callaghan, Grant Holder
Dr LP Maguire, Grant Holder
More details on this project can be found here.
The Remote laboratory website can be found here.
If you wish to access the online laboratory or require more information please email here.
Funded by Invest Northern Ireland. Website
Perfecseal Project #1
An Investigation into Innovation and Knowledge Development of Medical Adhesive Components.
This was a START Project (ST202) funded by Invest Northern Ireland (INI). The project was a joint development between Perfecseal Ltd. and the Faculties of Engineering and Informatics of University of Ulster. Perfecseal Ltd. is part of a multi-national company, Bemis Inc. USA, whose main business is the manufacture and sale of flexible plastic packaging materials. Located in Londonderry, the core business of Perfecseal is the production of healthcare packaging material. Using data engineering and intelligent approaches, two main points being investigated in this project were:
- Alternative medical packaging adhesive components that provide enhanced performance with reduced component and process cost.
- Interrelationships between adhesive formulation, substrate type, processing route and performance.
Funded by Invest Northern Ireland.
Type: Research and Development
Status: Completed
Funding Body: InvestNI
Perfecseal Project #2
Intelligent inferential modelling and control for optimising bacteria barrier packaging material production
This was a project sponsored by Perfecseal Ltd and DEL under the Collaborative Award in Science & Technology (CAST) program. The production of medical packaging material in Perfecseal Ltd. consists of coating a paper or plastic (i.e. Tyvek®) and drying the wet coated material (web), in a furnace.The dried web is required to have anchorage within a fixed range. If the anchorage is outside the desirable range, corrective action is initiated by making changes in the production process. In order to minimise wastage, the anchorage needs be estimated on-line and the constraints be placed on its variation while controlling the drying process. Using intelligent modelling techniques and production process data, this project developed an inferential model describing the relationship between the process parameters and the anchorage of the dried web.
Researcher:
Karl Warne
Status: Completed
Investigators:
Dr. G. Prasad, Dr. NH Siddique, Dr. LP Maguire
Perfecseal Website
QUDOS
Quantum Tunnelling Device Technology on Silicon
The constant demand for ever-increasing computing power can only be satisfied if the semiconductor industry can keep increasing the speed, miniaturisation and component density of the devices they manufacture. At the component level, this requires transistors to be continually reduced in size as each new generation of processors is developed. Unfortunately, as sub-micron dimensions of transistors fall below 30-50 nanometres, unwanted quantum effects begin to dominate rather than normal transistor behaviour, thus limiting the potential of existing silicon technology. This has created interest in alternative technologies that actually make use of quantum effect behaviour in nano-scale devices.
The QUDOS project set out to investigate the use of resonant tunnelling diodes (RTDs), manufactured on silicon, as fundamental components in new high-speed digital circuitry. RTDs have historically been manufactured on III/V materials such as GaAs/InP; these materials are expensive and unattractive to the semiconductor industry. Developing the devices on silicon would allow mainstream electronics manufacturers to consider their use as a means of meeting the demand for increased speed and processing power beyond the limits of existing silicon technology.
The devices operate by harnessing the tunnelling-effect through a quantum well, which results in a characteristic behaviour displaying negative differential resistance. This important feature allows high-speed operation with potential switching speeds approaching 100 GHz. Designs making use of the characteristic behaviour of the RTD also tend to reduce the number of components required for fundamental binary circuits. Higher speed, increased device density and reduced circuit complexity on silicon are the potential benefits offered by this technology.
Partners in the project included the Cavendish Laboratories of Cambridge University, the Max Planck Institute, the University of Braunschweig and the University of Duisburg who co-ordinatated the project.
The Intelligent Systems Research Centre in the Faculty of Computing and Engineering, University of Ulster at Magee, developed simulation models for the new devices and investigated circuit design techniques required to make use of the technology. The staff involved were Professor Martin Mc Ginnity, Dr. Peter Kelly and Professor Liam Maguire. The overall project budget was approximately 1 million Euros.
QUDOS was funded by the EU under the Future and Emerging Technologies area of the IST Programme and has a duration of three years. Website.
Period: January 2002 - December 2004
Type: Research
Status: Completed
Funding Body: EC
Funding Value: 1MEuros
Personnel Involved
Professor TM McGinnity, Grant Holder
Mr PM Kelly, Grant Holder
Dr LP Maguire, Grant Holder
EpiCentre: Electronics Production and Innovation Centre
Primarily the EPICentre aimed to assist the development and deployment of technology within the existing industrial base within the region. This industrial base addresses those sectors which inherently uses technology; ranging from the mass production of low cost products (e.g. the textile sector) to the development of low volume high cost products such as in the software tools development domain. The EPICentre also aimed to act as a focal point for technology in the region and this additionality afforded to existing companies also serves as an attractor for new industry to further bolster the economic vibrancy of the region.
The EPICentre project aimed to build infrastructure for an innovation and technology transfer capability through the creation of an environment that assists in developing a diverse and robust industrial base in the Northwest region. This centre focused upon the development of a strong industrial research base through the integration of existing capabilities in UUM, NWIFHE and LYIT with local incubation and established industries. The centre was distributed across the three sites yet had one Centre Manager to ensure integration and maintain consistent service across the region and queries were directed to the appropriate expertise at each location.
The Epicentre funding period has completed but the project is being sustained and is continuing to support local industry.
Full details of this project can be found here.
Period: December 2004 - September 2008
Type: Research
Status: Completed
Funding Body: European Union via SEUPB
Funding Value: Euro 3,000,000
Direct Range Image Processing
Edge detection or, more commonly, feature extraction can be readily performed on intensity images where the pixels are regularly placed like squares on a chessboard. More recently, the world of imaging and computer vision has moved towards the use of range data, obtained using a range camera or sensor. This range data is not regularly spaced, but instead is slightly randomly spaced and some of the required information may be missing. In order to perform any type of image processing, such as feature extraction or segmentation, on such irregularly spaced data, the data must be re-aligned mathematically on to a regular lattice, and in some cases the missing data is reconstructed. It not only takes time to perform these calculations, but such calculations can introduce approximation errors and data mis-representations. To avoid this unnecessary computation and error introduction, this project proposes a technique based on the use of finite element methods (FEMs) that will enable feature extraction operators to be generated that can be applied directly to the range image without any such pre-processing, thus proving to be more appropiate for real-time vision with the application of moving robots. This project will initially develop and implement the operators for use on irregular data and evaluate them in comparison to other existing techniques available. The project will then address the issue of finding out what type of feature has been found in the range image. The reason for this is that range image contain various types of edges: roof, jump, crease and smooth edge. Each of these features has different characteristics that must be found in order to determine the type of feature in the image. This is an important aspect of this project as most existing research focuses only on finding crease and jump edges in range images and not roof or smooth edges. Many object recognition systems used today are based on segmentation algorithms. Rather than a robot being able to determine precisely all the detail of a scene, it segments the scene into recognisable regions or objects and tries to match them with objects that it has seen before. On perfecting the finite element based feature extraction techniques and evaluating their ability to accurately characterise the features found in the range images, this technique will be used for segmentation of range data. The technique, combined with a simple edge-linking algorithm, should provide enclosed regions and reduce over or under segmentation. In order for this research to be appropriate for real-time imaging and hence useful for developing robot vision systems, it is required that the programs are coded in the C++ programming language, where the programmer has control of garbage collection and the time that it takes the program to run can be easily measured. Overall, this project aims to provide feature extraction operators that can be applied directly to range data for range image processing without the pre-processing steps that are currently essential to other techniques. This will reduce the mathematical computation required and thus enable improved real-time vision that can be useful for developing robot vision systems.
Period: February 2006 - July 2008
Type: Research
Status: Completed
Funding Body: EPSRC
Funding Value: £125,952
Personnel Involved
Dr SA Coleman, Principal Investigator
|
