Strategies to improve the performance of very low bit rate speech coders and application to a variable rate 1.2 kb/s codec

This paper presents several strategies to improve the performance of very low bit rate speech coders and describes a speech codec that incorporates these strategies and operates at an average bit rate of 1.2 kb/s. The encoding algorithm is based on several improvements in a mixed multiband excitation (MMBE) linear predictive coding (LPC) structure. A switched-predictive vector quantiser technique that outperforms previously reported schemes is adopted to encode the LSF parameters. Spectral and sound specific low rate models are used in order to achieve high quality speech at low rates. An MMBE approach with three sub-bands is employed to encode voiced frames, while fricatives and stops modelling and synthesis techniques are used for unvoiced frames. This strategy is shown to provide good quality synthesised speech, at a bit rate of only 0.4 kb/s for unvoiced frames. To reduce coding noise and improve decoded speech, spectral envelope restoration combined with noise reduction (SERNR) postfilter is used. The contributions of the techniques described in this paper are separately assessed and then combined in the design of a low bit rate codec that is evaluated against the North American Mixed Excitation Linear Prediction (MELP) coder. The performance assessment is carried out in terms of the spectral distortion of LSF quantisation, mean opinion score (MOS), A/B comparison tests and the ITU-T P.862 perceptual evaluation of speech quality (PESQ) standard. Assessment results show that the improved methods for LSF quantisation, sound specific modelling and synthesis and the new postfiltering approach can significantly outperform previously reported techniques. Further results also indicate that a system combining the proposed improvements and operating at 1.2 kb/s, is comparable (slightly outperforming) a MELP coder operating at 2.4 kb/s. For tandem connection situations, the proposed system is clearly superior to the MELP coder.     

Programmable analogue VLSI implementation for asymmetric sigmoid neural activation function and its derivative

A new CMOS VLSI implementation of an asymmetric programmable sigmoid neural activation function, as well as of its derivative, is presented. It consists of two coupled PMOS and NMOS differential pairs with different programmable bias currents that set the upper and lower limits of the sigmoid. The circuit works in the weak inversion region, for low power consumption and exponential envelope, or in strong inversion to achieve higher speeds. The results obtained from the theoretical transfer function, and from the simulations of the circuit implemented in AMI's 0.35 /spl mu/m technology, show a very good match.     

Dynamic vulcanisation of EPDM/PE-based thermoplastic vulcanisates studied along the extruder axis

Simple blending and dynamic vulcanisation of EPDM/PE blends using the resol/SnCl₂ system was studied in an extruder using a series of sampling devices. The melting of the PE pellets occurs just in front of and in the first kneading zone. Upon complete melting of the PE phase, the EPDM/PE blends reach very quickly their final morphology. Crosslinking of the EPDM phase to high levels occurs already when the PE phase is not yet fully molten. The higher the EPDM content, the higher the viscous dissipation, the higher the melt temperature and, consequently, the higher the crosslinking rate. For the EPDM/PE (50/50; w/w) TPV a transition from continuous via co-continuous to fully dispersed EPDM is observed, which is driven by crosslinking. Although the degree of crosslinking of the EPDM rubber is very high, this does not prevent phase inversion of the blend. The EPDM/PE interface is rather blurred, because PE crystalline lamellae are growing into the EPDM domains, probably because of the high compatibility between PE and EPDM (60 wt% of ethene). This study shows that dynamic vulcanisation in extruders proceeds quite differently from that in batch kneaders, where melting, mixing and crosslinking are separated in time. In extruders mass and heat transport, melting of the thermoplastic, morphology development (including dispersion and phase inversion), distribution and dissolution of (crosslinking) chemicals and crosslinking of the rubber do not occur as independent phenomena, but mutually and/or continuously interact.     

Integration of prior knowledge of measurement noise in kernel density classification

Samples can be measured with different precisions and reliabilities in different experiments, or even within the same experiment. These varying levels of measurement noise may deteriorate the performance of a pattern recognition system, if not treated with care. Here we seek to investigate the benefit of incorporating prior knowledge about measurement noise into system construction. We propose a kernel density classifier which integrates such prior knowledge. Instead of using an identical kernel for each sample, we transform the prior knowledge into a distinct kernel for each sample. The integration procedure is straightforward and easy to interpret. In addition, we show how to estimate the diverse measurement noise levels in a real world dataset. Compared to the basic methods, the new kernel density classifier can give a significantly better classification performance. As expected, this improvement is more obvious for small sample size datasets and large number of features.     

Life cycle of hydroprocessing catalysts and total catalyst management

Hydroprocessing catalysts based on Ni, Co, Mo and W are used in various refinery processing applications where several deactivation mechanisms become of importance (coke formation, active phase sintering, metals deposition, poisoning) in the catalyst's life cycle. The life cycle of commercial hydroprocessing catalysts is very complex and includes the catalyst production, sulfidation, use, oxidative regeneration followed by re-sulfidation and reuse or, if reuse is not possible, recycling or disposal. To understand the changes in catalyst properties taking place during a life cycle, the catalyst quality in the different stages can be best monitored by using advanced analytical techniques. The catalyst's life cycle is further complicated by numerous technical, environmental and organizational issues involved. In principle, different companies can be involved in each of the life cycle steps. Leading catalyst manufacturers, together with specialized firms, offer refineries a total catalyst management concept, starting with the purchase of the fresh catalyst and ending with its final recycling or disposal. Total catalyst management includes a broad range of services, ensuring optimal timing during the change-out process, reliable, smooth and safe operations, minimal downtime and maximum catalyst and unit performance.     

Synthesis and characterization of functional gradient copolymers of glycidyl methacrylate and butyl acrylate

Epoxy-functional spontaneous gradient copolymers of glycidyl methacrylate (G) and n-butyl acrylate (B) were synthesized via atom transfer radical polymerization (ATRP). The copolymerization reactions were carried out in toluene solution at 70 °C, using methyl 2-bromopropionate (MBrP) as initiator and copper chloride with N,N,N′,N′′,N′′-pentamethyldiethylenetriamine (PMDETA) as the catalyst system. The kinetic behaviour of the statistical copolymerizations was studied in a wide composition interval with molar fractions of G ranging from 0.10 to 0.75. The synthesized copolymers were characterized by size exclusion chromatography (SEC) and nuclear magnetic resonance (NMR) spectroscopy. ¹H NMR was employed to determine the copolymer composition, demonstrating the gradient character of the copolymers along the main chain in the whole monomer conversion interval. Apart from this, the sequence distribution and stereoregularity were analyzed. These microstructural experimental data agreed well with those calculated from Mayo-Lewis terminal model (MLTM) and a Bernoullian statistic with an isotacticity parameter of σ_G = 0.28 and a coisotacticity parameter of σ = 0.30.     

Separating n-alkane mixtures by exploiting differences in the adsorption capacity within cages of CHA, AFX and ERI zeolites

The saturation capacity of n-alkanes in CHA, AFX and ERI zeolites, that consist of cages separated by windows, decreases with increasing carbon number. The major aim of the present communication is to demonstrate the possibility of separating n-alkane mixtures relying on differences in saturation capacities. To investigate this possibility, Configurational-Bias Monte Carlo simulations for adsorption of C3–nC6, nC4–nC6, and nC5–nC6 mixtures in CHA, AFX and ERI were carried out for equimolar bulk fluid phase. These mixture simulations show that for operation at fluid phase fugacities below about 1 MPa, the adsorbed phase in equilibrium with the bulk vapor phase is predominantly the alkane with the longer chain length, i.e. nC6. However, for operation at pressures in excess of 1 MPa, the adsorbed phase in equilibrium with the bulk liquid phase is richer in the component with the smaller chain length. In some cases, the nC6 is practically excluded from the zeolite.     

Validity of the Parabolic Effective Mass Approximation in Silicon and Germanium n-MOSFETs With Different Crystal Orientations

This paper investigates the validity of the parabolic effective mass approximation (EMA), which is almost universally used to describe the size and bias-induced quantization in n-MOSFETs. In particular, we compare the EMA results with a full-band quantization approach based on the linear combination of bulk bands (LCBB) and study the most relevant quantities for the modeling of the mobility and of the on-current of the devices, namely, the minima of the 2-D subbands, the transport masses, and the electron density of states. Our study deals with both silicon and germanium n-MOSFETs with different crystal orientations and shows that, in most cases, the validity of the EMA is quite satisfactory. The LCBB approach is then used to calculate the values of the effective masses that help improve the EMA accuracy. There are crystal orientations, however, where the 2-D energy dispersion obtained by the LCBB method exhibits features that are difficult to reproduce with the EMA model.     

Prediction of necking of high strength steel tubes during hydroforming—multi-axial loading

This article studies tubular hydroforming of high strength low alloy (HSLA) and dual phase (DP600) straight tubes under the action of end feeding loads. Experiments demonstrate that higher end feed loads enhance the formability of the tubes and increase the internal fluid pressure for onset of necking and bursting. Because of the action of the internal pressure and the axial compressive load, the onset of localization (necking) is due to a complex three-dimensional state of stress. Using free expansion experiments, approximate upper and lower bound strain-based forming limit curves are determined for the tube materials. These limit curves, in turn, are used to derive upper and lower bound extended stress-based forming limit curves [Simha et al., Prediction of necking in tubular hydroforming using an extended stress-based FLC. Transactions of the ASME Journal of Engineering Materials and Technology 2007;129(1): 36-47]. In conjunction with finite element computations that use solid elements to model the tube, these stress-based limit curves are used to predict upper and lower bound necking pressures under the action of end feed loading. These predictions of necking pressures, when an appropriate coefficient of tube-die friction is used, are found to bracket the experimentally measured necking pressures. Computations using plane stress shell elements to model the tubes are shown to give erroneous results, since the plane stress approximation is not valid when tubes are hydroformed in a die.     

Driving Mechanisms of Delamination Related Reliability Problems in Exposed Pad Packages

Exposed pad packages were introduced in the late 1980s and early 1990s because of their excellent thermal and electrical performance. Despite these advantages, the exposed pad packages experience a lot of thermo-hygro-mechanical related reliability problems during qualification and testing. Examples are die lift, which occurs predominantly after moisture sensitivity level conditions, and die-attach to leadframe delamination leading to downbond stitch breaks during temperature cycling. In this chapter, nonlinear finite element (FE) models using fracture mechanics based J-integral calculations are used to assess the reliability problems of the exposed pad package family. Using the parametric FE models any geometrical and material effects can be explored to their impact on the occurrence diepad delamination, and dielift. For instance the impact of diepad size is found to be of much less importance as the impact of die thickness is. Using the fracture mechanics approach, the starting location for the delamination from thermo-hygro-mechanical point of view is deducted. The results indicate that when diepad delamination is present, cracks are likely to grow beneath the die and dielift will occur. The interaction between dielift and other failure modes, such as lifted ball bonds, are not found to be very significant. The FE models are combined with simulation-based optimization methods to deduct design guidelines for optimal reliability of the exposed pad family.