Face recognition with radial basis function (RBF) neural networks

A general and efficient design approach using a radial basis function (RBF) neural classifier to cope with small training sets of high dimension, which is a problem frequently encountered in face recognition, is presented. In order to avoid overfitting and reduce the computational burden, face features are first extracted by the principal component analysis (PCA) method. Then, the resulting features are further processed by the Fisher's linear discriminant (FLD) technique to acquire lower-dimensional discriminant patterns. A novel paradigm is proposed whereby data information is encapsulated in determining the structure and initial parameters of the RBF neural classifier before learning takes place. A hybrid learning algorithm is used to train the RBF neural networks so that the dimension of the search space is drastically reduced in the gradient paradigm. Simulation results conducted on the ORL database show that the system achieves excellent performance both in terms of error rates of classification and learning efficiency.     

The design and optimization of high-performance, double-polyself-aligned p-n-p technology

The device design and performance of double-poly self-aligned p-n-p technology, featuring a low-resistivity p⁺ subcollector, thin p-epi, and boron-doped poly-emitter are described. Device isolation is provided by deep and shallow trenches which reduce the collector-to-substrate capacitance while maintaining a high breakdown voltage (⩾40 V). By utilizing a shallow emitter process in conjunction with an optimized arsenic-base implant, devices with emitter-base junction depths as shallow as 20 nm and base widths of less than 100 nm were obtained. Cutoff frequencies of up to 27 GHz were obtained, and the AC performance was demonstrated by an NTL-gate delay of 36 ps and an active-pull-down (APD) ECL-gate delay of 20 ps. This high-performance p-n-p technology was developed to be compatible with existing double-poly n-p-n technologies. The matching speed of p-n-p devices opens up new opportunities for high-performance complementary bipolar circuits     

Molecular dynamics simulations of HLA-DR4 (DRB1*0405) complexed with analogue peptide: conformational changes in the putative T-cell receptor binding regions

The specific recognition of foreign peptide bound to the major histocompatibility complex (MHC) molecule by T-cell receptor (TCR) leads to T-cell activation. We found that analogue peptides containing single amino acid substitutions at the third amino acid position (p3), p5, p7 and p8 of the index peptide (YWALEAAAD) induced different response patterns of T cell clones specific for the index peptide in the context of the human MHC class II molecule HLA-DR4. Analogue peptides were classified into three types, agonists, antagonists or null peptides (non-agonistic and non-antagonistic peptides). A molecular basis for how these slight changes lead to such different consequences for T cells has not been described. To explore the mechanistic basis of these observations, molecular dynamics simulations at 300 K of 300 ps duration were carried out for the DR4-index peptide, DR4-agonist, and DR4-antagonist complexes. The simulations showed that the DR4-antagonist complexes were distinguished from the DR4-index peptide and DR4-agonist complexes by relatively higher deviations of C(alpha) atoms in proposed TCR-binding regions, suggesting that subtle changes of the exposed framework of the peptide binding groove by the antagonist peptides could induce the TCR antagonistic activities.     

Block preconditioners for symmetric indefinite linear systems

This paper presents a systematic theoretical and numerical evaluation of three common block preconditioners in a Krylov subspace method for solving symmetric indefinite linear systems. The focus is on large‐scale real world problems where block approximations are a practical necessity. The main illustration is the performance of the block diagonal, constrained, and lower triangular preconditioners over a range of block approximations for the symmetric indefinite system arising from large‐scale finite element discretization of Biot's consolidation equations. This system of equations is of fundamental importance to geomechanics. Numerical studies show that simple diagonal approximations to the (1,1) block K and inexpensive approximations to the Schur complement matrix S may not always produce the most spectacular time savings when K is explicitly available, but is able to deliver reasonably good results on a consistent basis. In addition, the block diagonal preconditioner with a negative (2,2) block appears to be reasonably competitive when compared to the more complicated ones. These observation are expected to remain valid for coefficient matrices whereby the (1,1) block is sparse, diagonally significant (a notion weaker than diagonal dominance), moderately well‐conditioned, and has a much larger block size than the (2,2) block. Copyright © 2004 John Wiley & Sons, Ltd.     

Mobile multicasting in wireless ATM networks

Multicasting is a desired function in both wired and wireless networks. Currently, multicasting‐based applications have pervasive presence and influence in our Internet. Wireless ATM aims to support different traffic types over a high speed wireless network. In particular, traditional approaches towards multicasting in wired networks cannot be directly applied to mobile ATM networks. This includes the mechanism proposed to handle mobility in wired networks. In this paper, we address the issues behind multicasting in wireless ATM networks and the underlying mechanisms to support handoffs of both unicast and multicast connections. Through simulation, we evaluated the performance of optimal core‐based tree (CBT), late‐joined CBT, optimal source‐based tree (SBT), and late‐joined SBT during changing multicast host membership conditions. In addition, we evaluated the impact of multicast receivers' migration on the total link cost of the resultant multicast tree. Simulation results revealed that source‐based trees incur more total link costs under both dynamic host membership and receivers' migration scenarios. However, it generates less control messages than the distributed CBT and SBT approaches. This revised version was published online in June 2006 with corrections to the Cover Date.     

Instability of plastic flow in mild steel

Plastic flow localization in the yielding of mild steel is theoretically studied in terms of the instability of the uniform mode of yielding. The instability is judged by the small perturbation theory in which the development of non-uniform modes superimposed upon the basic uniform mode is examined. The constitutive equation of rate-dependent type and the criterion of yield initiation at the plastic-zone front are employed as the constitutive relations of mild steel. The results of the analysis show the relation between the instability and the characteristic nature of mild steel.     

Evolution and stability of 2-mercaptobenzimidazole inhibitor film upon Al alloy 6061

Journal of Applied Electrochemistry - Organic corrosion inhibitors have become competent alternatives to hazardous chrome conversion coatings due to their rapid adsorption over metal surfaces in...     

Thermal challenges in next generation electronic systems - summary of panel presentations and discussions

The presentations made, as well as the discussions, in the panels at the workshop, Thermal Challenges in Next Generation Electronic Systems (THERMES), are summarized in this paper. The panels dealt with diverse topics including thermal management roadmaps, microscale cooling systems, numerical modeling from the component to system levels, hardware for future high performance and Internet computing architectures, and transport issues in the manufacturing of electronic packages. The focus of the panels was to identify barriers to further progress in each area that require the attention of the research community.     

A Flexi-PEGDA Upconversion Implant for Wireless Brain Photodynamic Therapy

Near-infrared (NIR) activatable upconversion nanoparticles (UCNPs) enable wireless-based phototherapies by converting deep-tissue-penetrating NIR to visible light. UCNPs are therefore ideal as wireless transducers for photodynamic therapy (PDT) of deep-sited tumors. However, the retention of unsequestered UCNPs in tissue with minimal options for removal limits their clinical translation. To address this shortcoming, biocompatible UCNPs implants are developed to deliver upconversion photonic properties in a flexible, optical guide design. To enhance its translatability, the UCNPs implant is constructed with an FDA-approved poly(ethylene glycol) diacrylate (PEGDA) core clad with fluorinated ethylene propylene (FEP). The emission spectrum of the UCNPs implant can be tuned to overlap with the absorption spectra of the clinically relevant photosensitizer, 5-aminolevulinic acid (5-ALA). The UCNPs implant can wirelessly transmit upconverted visible light till 8 cm in length and in a bendable manner even when implanted underneath the skin or scalp. With this system, it is demonstrated that NIR-based chronic PDT is achievable in an untethered and noninvasive manner in a mouse xenograft glioblastoma multiforme (GBM) model. It is postulated that such encapsulated UCNPs implants represent a translational shift for wireless deep-tissue phototherapy by enabling sequestration of UCNPs without compromising wireless deep-tissue light delivery.