JPEG2000 is a recently standardized image compression system that provides substantial improvements over the existing JPEG compression scheme. This improvement in performance comes with an associated cost in increased implementation complexity, such that a purely software implementation is inefficient. This work identifies the arithmetic coder as a bottleneck in efficient hardware implementations, and explores various design options to improve arithmetic coder speed and size. The designs produced improve the critical path of the existing arithmetic coder designs, and then extend the coder throughput to 2 or more symbols per clock cycle. Subsequent work examines more system level implementation issues. This work examines the communication between hardware blocks and utilizes certain modes of operation to add flexibility to buffering solutions. It becomes possible to significantly reduce the amount of intermediate buffering between blocks, whilst maintaining a loose synchronization. Full hardware implementations of the standard are necessarily limited in the amount of features that they can offer, in order to constrain complexity and cost. To circumvent this, a hardware / software codesign is produced using the Altera NIOS II softcore processor. By keeping the majority of the standard implemented in software and using hardware to accelerate those time consuming functions, generality of implementation can be retained, whilst implementation speed is improved. In addition to this, there is the opportunity to explore parallelism, by providing multiple identical hardware blocks to code multiple data units simultaneously.

The aim of this work was to develop devices for elite athletes to record performance related parameters during their training. A device was initially designed and built for rowing to record the motion of the boat. This was to gain understanding of motion signals in a one dimensional plane. The device uses a iPAQ handheld computer for recording and display of data to the user. Using the knowledge obtained from the accelerometer data of the rowing system an initial prototype device was designed and constructed for use in swimming. This device was required to be wearable whilst the swimmer was training, thus it had to record the data onboard. A second version of the swimming device was constructed to improve the usability of the device. The swimming device has fully sealed electronics, wireless charging and infrared communications. The device records three dimensional acceleration patterns at 150Hz, and can store over 6 hours of data using the internal memory. The device can operate for greater than 12 hours before needing to be recharged. The data collected from the swimming device was used to develop processing algorithms to extract when the swimmers push off from the wall, the type of stroke they are swimming, and for freestyle the stroke count. The results of the wall push off algorithm were compared against manual hand timing with 90% algorithm results being with ±1 second of the hand timing data. The stroke type identification algorithm determines which stroke is being swum and presently has an accuracy of 95%. The results of the freestyle stroke count algorithm were compared against manual stroke counts from raw accelerometers data and underwater video. Of the 164 data sets analysed over 90% of the algorithm results were within ±1 strokes of the manual recorded stroke counts.

This study examines the effects of acceleration on fluid forces acting on the hand in swimming and presents a new pressure method to predict fluid forces acting on the accelerating hand. Swimmers and coaches require accurate information about swimmers� fluid forces, propulsion and drag, in order to improve performance. In general, swimmers are likely to generate propulsion mainly with their hands in the front crawl stroke, butterfly and back crawl stroke. Researchers have attempted to estimate the fluid forces on the hands with various techniques including a cinematographic method (a "quasi-static" approach), a pressure method and a numerical method of computational fluid dynamics. However, the effect of accelerations on fluid forces acting on the hand has not yet been well quantified. Understanding of the effect of acceleration on fluid forces on the hand can provide useful information to enhance swimming performance. Also the developments of a method to predict fluid forces acting on the accelerating hand in swimming can be used to evaluate swimming performance more accurately. The present study used a hand model attached to a load cell to measure forces in three orthogonal directions and pressure sensors to measure pressures on the hand model rotated in the flume. The model position was measured by a potentiometer fixed to the axis of the model rotation. The quantification of the effect of acceleration was based on a simple theoretical understanding for fluid mechanics, using the inertia coefficients and the coefficients of fluid forces, that is widely accepted in other disciplines. The quantification was focused mainly in the direction tangential to the model rotation because the magnitude of the velocity changed in this direction. The overall effect of acceleration on fluid forces on the hand model was that the inertia coefficients increased rapidly in the early phase of the model movement, then in the final deceleration phase of model movement the inertia coefficients decreased to a negative value and then became small. The inertia coefficient increased in the impulsive start of the hand model, indicating that fluid forces acting on the hand increased as accelerations of the hand increased. This result was consistent with the simple theoretical understanding to induce additional fluid forces on the hand, that is, fluid forces on the hand increased as accelerations increased. However, the inertia coefficient decreased and reached large negative values in the late phase of the model movement involving decelerated motion, indicating that fluid forces on the hand increased as acceleration of the hand decreased to negative values (decelerations). That result was not consistent with the simple theoretical understanding to induce additional fluid forces on the hand because the simple theoretical understanding cannot take account of the preceding history of the fluid motion around the hand model associated with the formation of vortices. Thus, more sophisticated theory is needed. The dynamic pressure measured by the pressure sensors implied that the induced fluid forces might be due to large attached vortices behind the hand model. The hand was considered as a blunt body when the angle of attack was large (maximum = 90�) and an aerofoil shape when the angle of attack was small. The inertia coefficients became large when the hand model was set at the large angles of attack, indicating that the effect of accelerations on the hand model increased when the hand surface was directed to the on-coming flows (blunt body). For the development of the new pressure method, a regression analysis was used to build a single best-fit equation to predict fluid forces acting on the accelerating hand model. The single best-fit equation was acquired for various orientations of the hand model. The accuracy of prediction of fluid forces acting on the accelerating hand model was checked by a root mean square (RMS) difference. The RMS difference by the pressure method was approximately half of the RMS difference by the "quasi-static" approach that has been a major method to predict fluid forces exerted by the hand in swimming. The present study has quantified the effect of acceleration on fluid forces acting on the hand in swimming and developed a new pressure method including acceleration effects to predict fluid forces acting on the accelerating hand.

Mémoire de ma thèse d'Habilitation, soutenue le 11 mai 2005 à Orsay.

In this thesis the following problems are studied: 1-Relativistic self-focusing and channelling of intense laser pulses have been studied in underdense plasma using 2D PIC simulations, for different laser powers and plasma densities. Analytical solutions for the stationary evacuated channels have been recovered in PIC simulations. It is shown that otherwise stable channels can accelerate electrons due to surface waves on the walls of the channels. Relativistic filaments with finite electron density are unstable to transverse modulations which lead in the nonlinear stage to the break-up of laser pulses into independent filaments. 2-Although 3D simulations are limited, they are more realistic. Azimuthal stability of the laser pulses in interaction with underdense plasma can only be studied in 3D geometry. Relativistic self-focusing and channelling of intense laser pulses have been studied in underdense plasma using 3D PIC simulations, for different laser powers and plasma densities. Analytical solutions for the stationary evacuated channels and ring structure have been recovered in PIC simulations. The stability of ring structure due to azimuthal perturbations has been studied both in theory and in simulations. The gain length of such instability is smaller at higher densities $(>0.1n_{cr})$. It is shown that the azimuthal perturbation can break up the azimuthal symmetry of the laser pulse. 3-Working with circularly polarized laser pulses, gave us a motivation to study Inverse Faraday Effect in interaction of circularly polarized laser pulses with plasma. Axial magnetic field generation by intense circularly polarized laser beams in underdense plasmas has been studied with 3D particle-in-cell (PIC) simulations and by means of theoretical analysis. The source of azimuthal nonlinear currents and of the axial magnetic field depends on the transverse inhomogeneities of the electron density and laser intensity. The fields reach maximum strength of several tens of MG for laser pulses undergoing relativistic self-focusing and channelling in moderately relativistic regime. 4-Electron wakefield acceleration was studied in support of the experiment which was carried on using 7 TW laser beam at Canadian Advanced Laser Light source facility. 2D simulations were performed to study this problem. The energy the electrons gained in the process was peaked at 20-30 Mev close to the experimental results.

In this thesis electron generation is studied via laser plasma interaction known as laser wakefield acceleration in two regimes of weakly relativistic and highly relativistic laser intensity regimes. The plasma targets consisted of gas jets photonionized by rising edge of the laser pulse to densities as high as 10^20cm3. In the weakly relativistic regime, 210 mJ at 33 fs were focused to intensities of up to 310^18 Wcm2 on the gas targets of 2.4 mm length. In the highly relativistic regime, 3 J of energy compressed in 30 fs were delivered at intensity as high as 6.5 10^18 Wcm2 on targets of 2.4, 5, and 10 mm. Monoenergetic electrons in tens of MeV were observed in weakly relativistic regime, while electron energies as high as 300 MeV were observed in highly relativistic regime. Higher input laser intensity and prepulse levels were found to enhance electron production. Scaling of energy and stability of electron generation were also studied. / Photonics and Plasmas

This thesis deals with the reliability and life-time ofelectronic components and ways to determine these factors.Plastic encapsulated and open test circuits were assessed atdifferent humidity and temperature conditions. From the resultsan acceleration factor could be derived using the Arrheniusrelation. This factor is used to determine failure rates atdifferent drift conditions under accelerated test conditions. Aformula for the factor containing both relative humidity andtemperature could be established and was found to hold also formeasurements published by others. Electrostatic discharge (ESD) transients were studiedexperimentally and by simulation with good agreement. A verysensitive method to detect latent failures of two kinds wasintroduced by nonlinearity measurements utilizing the thirdharmonic of a test signal. The ESD-susceptibility dependence ondesign and technology is shown and can be used to improvebuilt-in reliability. Influences in the performance of semiconductor devices fromdefects like fixed charges and ions were interpreted for thefirst time by simulation using a 2D- finite element componentprogram. Significant results gave an application to a MOSFETdevice showing parameter derating, especially the change of thethreshold value. A short description of later development insimulation methods with new, more powerful tools improvingcomponent performance and reliability is given. Charged thin films of Teflon, so calledelectrets, are used as microphone membranes. Theelectret voltage is a suitable reliability factor. Fromexperimental results a mathematical relation including thetemperature was established for the rate of decay of theelectret voltage with time. A method to charge the electretswith radioactive sources is outlined and described in apatent. Finally an attempt was done to analyze the reliability ofthin film circuits by mathematical methods. Bell LabsintroducedRC-feedback filters realized in tantalum thin filmtechnology. The phase shift of the filter is about π or180°. A mathematical apparatus was developed to calculatethe change in frequency and attenuation from small componentvariations in resistors and capacitors. First and higher ordercorrections were derived, using expansion by the Taylor seriesfor the higher order. <b>Keywords:</b>reliability, failure mechanism, accelerationtests, ESD, latent failure, plastic encapsulation, electret,thin film

A study of impulsively started flow over cylindrical objects is made using the particle image velocimetry (PIV) technique for Reynolds numbers of Re = 200, 500 and 1000 in an X-Y towing tank. The cylindrical objects studied were a circular cylinder of diameter, D = 25.4 mm, and square and diamond cylinders each with side length, D = 25.4 mm. The aspect ratio, AR (= L/D) of the cylinders was 28 and therefore they were considered infinite. The development of the recirculation zone up to a dimensionless time of t* = 4 following the start of the motion was examined. The impulsive start was approximated using a dimensionless acceleration parameter, a*, and in this research, the experiments were conducted for five acceleration parameters, a* = 0.5, 1, 3, 5 and 10. The study showed that conditions similar to impulsively started motion were attained once a* ¡Ý 3.<p> A recirculation zone was formed immediately after the start of motion as a result of flow separation at the surface of the cylinder. It contained a pair of primary eddies, which in the initial stages (like in this case) were symmetrical and rotating in opposite directions. The recirculation zone was quantified by looking at the length of the zone, LR, the vortex development, both in terms of the streamwise location and the cross-stream spacing of the vortex centers, a and b, respectively, as well as the circulation (strength) of the primary vortices, ¦£.<p> For all types of cylinders examined, the length of the recirculation zone, the streamwise location of the primary eddies and the circulation of the primary eddies increase as time advances from the start of the impulsive motion. They also increase with an increase in the acceleration parameter, a*, until a* = 3, beyond which there is no more change, since the conditions similar to impulsively started conditions have been achieved. The cross-stream spacing of the primary vortices is relatively independent of Re, a* and t* but was different for different cylinders.<p> Irrespective of the type of cylinder, the growth of the recirculation zone at Re = 500 and 1000 is smaller than at Re = 200. The recirculation zone of a diamond cylinder is much larger than for both square and circular cylinders. The square and diamond cylinders have sharp edges which act as fixed separation points. Therefore, the cross-stream spacing of the primary vortex centers are independent of Re, unlike the circular cylinder which shows some slight variation with changes in Reynolds number.<p> The growth of the recirculation is more dependent on the distance moved following the start of the impulsive motion; that is why for all types of cylinders, the LR/D, a/D and ¦£/UD profiles collapse onto common curves when plotted against the distance moved from the start of the motion.

The purpose of this study was to compare the shoulder resultant joint moment (RJM) during a shoulder internal rotator exercise using elastic resistance employing four different movement strategies and two different starting elastic strains. Methods: Ten subjects aged 27.4 ± 2.6 yr (5 female and 5 male) with no previous shoulder pathology performed four sets of six repetitions of shoulder rotation though 180° using elastic resistance (Thera-Band® elastic band, blue) during two acceleration (medium and low) and two cadence (2s:2s, <1s:1s) strategies at 0% elastic starting strain. The acceleration movement strategies were also performed with starting strain of 30%. A mathematical model using Newtonian mechanics was used to compute the RJM. Elastic band recoil force was measured with a force transducer. Forearm acceleration was determined by a miniature uniaxial accelerometer secured at the wrist. Electrogoniometer data were collected to determine the range of motion (ROM) as well as the angle between the forearm and band which was used to determine elastic moment arm. Paired t-tests were used to identify joint angle specific RJM differences between conditions. Results: Angle specific comparisons revealed that RJM in the moderate acceleration movement strategy was significantly different from RJM in the low acceleration movement strategy through 150° (83%) of range of motion (p<0.05). Shoulder RJM was up to 111% higher in the moderate acceleration strategy (P < 0.01). Angle specific comparisons revealed RJM in the <1:1 cadence strategy was significantly different from RJM in the 2:2 cadence through 108° (60%) of the range of motion (p<0.05). RJM was up to 47% higher in the <1:1 cadence (p<0.01). RJM in the low acceleration strategy was significantly greater with 30% elastic start strain relative to 0% elastic start strain through 180º of angular excursion (p<0.001). The pattern and magnitude of neuromuscular loading was significantly different in higher acceleration movement strategies (moderate acceleration and fast cadence). Conclusions: These findings indicate that differential limb acceleration as a result of movement strategy significantly affects shoulder load during elastic resistance exercise. The pattern and magnitude of load was different in each movement strategy and could result in differential neuromuscular adaptation through training. Clinicians and exercise professionals should consider movement strategy/acceleration as an important factor when prescribing elastic resistance exercise for safety and efficacy. / October 2006

No description available.