The ANS (Autonomic Networked System) is a ubiquitous computing managementtool which is designed to mimic the ANS (Autonomic Nervous System) of living creatures. The organic ANS is the part of the nervous system controlling many organs and muscles within thebody. It is flexible, constantly in operation and that it happens in the background without our interference or knowledge of its mechanism. This metaphore is being adapted to support ubiquitous computing environments, especially in the application of the intelligent home and medical applications where constant technical supportis impossible. Such a system will provide the intelligence to optimise its operation through constant monitoring and tuning to achieve its goal.
Southampton Universities Microfabrication Facility is the national centre for silicon based microelectronics with a 600m2 cleanroom with full lithographic, deposition, etching and characterisation capability. Currently the facility does not have standardised solar cell processes. This workpackage will establish facility processes for the fabrication of crystalline and polycrystalline solar cells with efficiencies of ~20% and ~14% respectively. This will directly be of benefit to the wider UK photovoltaic research community who will then have access to these baseline capabilities via EPSRC facility requests. This work will also provide important first steps towards our longer term goals from this baseline we will be able to readily attempt new structures, devices and processes.
A visible wavelength linear photosensor featuring a pixel size of 3 mm has been designed for fabrication using commercial 0.25 mm CMOS technology. For the photo-sensing element, the design uses a special deep N-well in P-epi diode offered by the foundry for imaging devices. Pixel reset is via an adjacent p-FET, thus allowing high reset voltages for a wide pixel voltage swing. The pixel voltage is buffered using a voltage-follower op-amp and a sampling scheme is used to allow correlated double sampling (CDS) for removal of reset noise. Reset and signal levels are buffered through a 16:1 multiplexer to a switched capacitor amplifier which performs the CDS function. Incorporated in the CDS circuit is a programmable gain of 1:8 for increased signal-to-noise ratio at low signal levels. Data output is via 4 analogue output drivers for off-chip conversion. Each driver supplies a differential output voltage with a 71V swing for improved power supply noise rejection. The readout circuitry is designed for 12 bit accuracy at frame rates of up to 6.25 kHz. This gives a peak data rate at each output driver of 10M samples/s. The device will operate on a 3.3V supply and will dissipate approximately 950mW. Simulations indicate an equivalent noise charge at the pixel of 66.3 for a full well capacity of 255,000, giving a dynamic range of 71.7 dB.
Planar waveguides and photonic crystal structures are being intensively investigated as primary solutions for integrated photonic devices. However, there may be an alternative approach to the manufacture of highly integrated optical devices with structural elements smaller than the wavelength, which nevertheless enable strong guidance and manipulation of light - the use of metallic and metallodielectric nanostructures and propagating plasmonÂpolanton waves. This approach is now branded as "the next big thing" in nanotechnology. Here we propose, for the first time, a research programme to investigate the basic principles of active nanoscale functional elements operating with surface plasmon-polanton signals. We will employ an original and revolutionary active control concept that will use structural transformations in the waveguide material to control the signals. The proposed solution takes advantage of the most characteristic features of surface plasmon-polantons, namely their localization in nanometer thick surface layers of metal, and the fact that their propagation is strongly dependent on the metal's dielectric properties. If successful, our approach could provide an alternative means of developing active integrated photonic (logic) circuits.
As the range and scope of thermal imaging and sensing applications expand quantum well infrared photodetectors (QWIPs) are emerging as a new and important technology. The best known and most widely discussed are the n-type devices based on GaAs/AlGaAs. The combined promise of normal incidence detection, a route to integration and the potential for far-infrared ( >20 micron) detection, with Si/SiGe quantum well and quantum dot infrared photodetectors (QDIPS), provides the motivation behind this challenging projects that will develop SiGe epitaxy still further.
Recently a new group of layered planar and quasi-planar metamaterials has emerged which promise unique electromagnetic properties. Layered metallic microstructures could play a special role in future technology, as they can be manufactured on a sub-optical wavelength scale and can be fabricated using established microelectronics technologies. We have begun a predominantly experimental study of planar and quasi-planar metallic microstructures, a new generation of metamaterials for optical applications. We will concentrate on various chiral planar metamaterials fabricated on the optical wavelength scale.
This project is concerned with the development and application of optimisation methods for machine learning algorithms. Many modern machine learning algorithms can be viewed as optimising bounds on the generalisation error derived in learning theory. Modern tools from mathematical programming such as second-order cone and semi-definite programs will be adapted to the optimisation problems arising in machine learning. The resulting methods will be tested on benchmark data and - whenever possible - on suitable real-world data sets.
Learning systems based on kernels are a powerful class of algorithms that includes Support Vector Machines and Gaussian Processes. These systems have become a major part of current research into and applications of adaptive systems. Despite this fact very little is known about when we can expect these systems to perform well. There has even been the assumption made that they provide a universal learning methodology. The proposed project will address this in order to:
This project invetsigates the application of adaptive numerical modelling techniques for the process optimisation of aluminium alloys.