LPNN-based approach for LASSO problem via a sequence of regularized minimizations

In compressive sampling theory, the least absolute shrinkage and selection operator (LASSO) is a representative problem. Nevertheless, the non-differentiable constraint impedes the use of Lagrange programming neural networks (LPNNs). We present in this article the -LPNN model, a novel algorithm that tackles the LASSO minimization together with the underlying theory support. First, we design a sequence of smooth constrained optimization problems, by introducing a convenient differentiable approximation to the non-differentiable -norm constraint. Next, we prove that the optimal solutions of the regularized intermediate problems converge to the optimal sparse signal for the LASSO. Then, for every regularized problem from the sequence, the -LPNN dynamic model is derived, and the asymptotic stability of its equilibrium state is established as well. Finally, numerical simulations are carried out to compare the performance of the proposed -LPNN algorithm with both the LASSO-LPNN model and a standard digital method.     

Immobilization and activation of vanadium(III) and titanium(III) single‐site catalysts for ethylene polymerization using MgCl2‐based supports

Significant increases in the activity of vanadium(III) amidinate catalysts for ethylene polymerization have been obtained by immobilization on a MgCl₂‐based support prepared by reaction of AlEt₃ with a MgCl₂/ethanol adduct. Catalyst immobilization and activation on this type of support prevents the rapid decay in activity observed under homogeneous polymerization conditions with unsupported catalysts. Stable polymerization activity is also observed with analogous titanium(III) complexes. Polyethylene with narrow molecular weight distribution and spherical particle morphology is obtained without reactor fouling. Copyright © 2005 Society of Chemical Industry     

In situ polymerization and characterization of polyester‐based polyurethane/nano‐silica composites

Polyester‐based polyurethane/nano‐silica composites were obtained via in situ polymerization and investigated by Fourier‐transform infrared spectroscopy (FTIR), or FTIR coupled with attenuated total reflectance (FTIR‐ATR), Transmission electron microscopy (TEM), atomic force microscopy (AFM), an Instron testing machine, dynamic mechanical analysis (DMA) and ultraviolet‐visible spectrophotometry (UV‐vis). FTIR analysis showed that in situ polymerization provoked some chemical reactions between polyester molecules and nano‐silica particles. FTIR‐ATR, TEM and AFM analyses showed that both surface and interface contained nano‐silica particles. Instron testing and DMA data showed that introducing nano‐silica particles into polyurethane enhanced the hardness, glass temperature and adhesion strength of polyurethane to the substrate, but also increased the resin viscosity. UV‐vis spectrophotometry showed that nano‐silica obtained by the fumed method did not shield UV radiation in polyurethane films. Copyright © 2003 Society of Chemical Industry     

ROI-based Watermarking Scheme for JPEG 2000

A new region of interest (ROI)-based watermarking method for JPEG 2000 is presented. The watermark is embedded into the host image based on the characteristics of the ROI to protect rights to the images. This scheme integrates the watermarking process with JPEG 2000 compression procedures. Experimental results have demonstrated that the proposed watermark technique successfully survives JPEG 2000 compression, progressive transmission, and principal attacks.     

Impact of QOS requirements on video coding for ATM networks

The broadband integrated services digital networks (B-ISDN) based on asynchronous transfer mode (ATM) technology can support a wide range of applications such as voice, video, still images, and data. Compression techniques increase the effective bandwidth utilization, but the bursty and asynchronous nature of the traffic can still lead to congestion in the network, and degradation of image quality and quality of service (QOS). Some of the features to provide better coding schemes for ATM networks are layered coding, resynchronization, buffering, interleaved schemes, constrained bit rate due to buffers, encapsulation with the RTP or AAL1 for clock recovery, lapped transforms, motion compensation, and optimal bit allocation for coders based on wavelet transforms. We review various techniques forimage and video coding such as transforms, motion compensation, vector quantization, and subband coding. We outline the impact of the cell loss ratio (CLR), delay and cell delay variation (CDV) on video coding: blocking effects, loss of frame synchronization, motion vectors, and vector quantization codewords. The open problems include tuning coding parameters to the available QOS provided by the network.     

Series resonant ZCS‐PFM DC‐DC power converter with high‐frequency high‐voltage transformer link for high‐power magnetron drive

This paper presents a single lossless inductive snubber‐assisted ZCS‐PFM series resonant DC‐DC power converter with a high‐frequency high‐voltage transformer link for industrial‐use high‐power magnetron drive. The current flowing through the active power switches rises gradually at a turned‐on transient state with the aid of a single lossless snubber inductor, and ZCS turn‐on commutation based on overlapping current can be achieved via the wide range pulse frequency modulation control scheme. The high‐frequency high‐voltage transformer primary side resonant current always becomes continuous operation mode, by electromagnetic loose coupling design of the high‐frequency high‐voltage transformer and the magnetizing inductance of the high‐frequency high‐voltage transformer. As a result, this high‐voltage power converter circuit for the magnetron can achieve a complete zero current soft switching under the condition of broad width gate voltage signals. Furthermore, this high‐voltage DC‐DC power converter circuit can regulate the output power from zero to full over audible frequency range via the two resonant frequency circuit design. Its operating performances are evaluated and discussed on the basis of the power loss analysis simulation and the experimental results from a practical point of view. © 2005 Wiley Periodicals, Inc. Electr Eng Jpn, 153(3): 79–87, 2005; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.20126     

SuperDense Quantum Teleportation

Recent advances in experimental technique make SuperDense Teleportation (SDT) possible only now, ten years after my first proposal at an ISI Torino summer conference on Quantum Computing. The effect uses remote state preparation to send more state-specifying parameters per bit than ordinary quantum teleportation (QT) can transmit. The SDT uses a maximally entangled state to teleport the relative phases of an n-dimensional state with equal amplitudes on every standard basis vector. For n greater than or equal to 3, the SDT sends more of these state-specifying parameters than QT. In the limit of large n the ratio is 2 to 1, hence the nomenclature by analogy with Super Dense Coding. Alice’s measurements and Bob’s transformations are far simpler than their corresponding operations in QT. The roles of Charles who chooses the state and Diana who deploys it are different than in QT. My discussion includes a brief review of the progress and possibilities of realization for several different experimental approaches around the world. This paper is the write-up of my remarks at the Festschrift conference for Anton Zeilinger, for many years a close collaborator in the Hampshire College NSF grant continuing our work with Mike Horne and Danny Greenberger started under Cliff Shull at MIT in the late 20th century.     

Site occupations and ordering behaviour of O phase in Ti2AlNb—based intermetallics

1　ＩＮＴＲＯＤＵＣＴＩＯＮＴｈｅｏｒｔｈｏｒｈｏｍｂｉｃａｌｌｏｙｓａｒｅｏｆｇｒｅａｔｔｅｃｈ ｎｏｌｏｇ ｉｃａｌｉｎｔｅｒｅｓｔｉｎｈｉｇｈｔｅｍｐｅｒａｔｕｒｅａｐｐｌｉｃａｔｉｏｎｓｓｉｎｃｅｔｈｅｙｈａｖｅａｇｏｏｄｂａｌａｎｃｅａｍｏｎｇｓｔ     

基于线性动态系统稀疏编码的异常事件检测

线性动态系统模型结合稀疏编码实现异常事件检测。线性动态系统可有效地捕捉动态纹理在时间和空间的转移信息,描述视频的时空小块。然而,线性动态系统属于非欧氏空间,无法直接用传统的稀疏编码进行异常检测。基于约束凸优化公式,将相似性变换与稀疏编码结合,可实现线性动态系统稀疏编码的优化求解。实验表明,所提出的方法具有更好的性能。