Degree-of-polarization and eye-closure penalty associated with optical signals with orthogonal polarizations

We perform a systematic measurement of the degree-of-polarization (DOP) and eye-closure penalty for optical signals with orthogonal polarizations. We find that the symmetry of DOP is maintained for the orthogonal polarizations under both first and higher order polarization-mode dispersion (PMD), whereas the symmetry of eye-closure penalty is broken under second-order PMD. An orthogonal polarization pair can have large disparity of eye-closure penalty despite an identical DOP. We also demonstrate a novel approach to estimate the maximum eye-closure penalty asymmetry with three orthogonal polarizations on the Poincare/spl acute/ sphere.     

多带抗噪声语音识别算法研究

根据Flether等人的研究,基于感知独立性假设的子带识别方法被用于抗噪声鲁棒语音识别。本文拓展子带方法,采用基于噪声污染假定的多带框架来减少噪声影响。论文不仅从理论上分析了噪声污染假定多带框架在识别性能上的潜在优势,而且提出了多带环境下的鲁棒语音识别算法。研究表明：多带框架不仅回避了独立感知假设要求,而且与子带方法相比,多带方法能更好的减少噪声影响,提高系统识别性能。     

蜕变——中国气象局华风影视大楼

华风气象影视大楼总面积4万平方米，根据空间功能不同分为办公大楼和商务写字楼两部分，办公大楼是集演播室、制作机房、办公室为一体的九层综合性建筑。设计师杨宇在刚接触到这个项目的时候，面对的是一个既成事实的建筑。     

地震勘探测量数据处理的全面解决方案

本文针对近年来地震勘探技术的发展对测量工作提出的新要求，分析了SSOffiee软件解决问题的方案和技术特色，最后提出了SSOffice软件的发展方向。     

Defocusing image to pattern contact holes using attenuated phase shift masks

The patterning of contact holes by selecting out-of-focus image plane (defocus) using attenuated phase shift masks (APSM) has been studied. Defocus is found to enhance the image modulation at low partial coherence for contact holes with negative local average of mask function. Semi-dense holes up to 130 nm in 8% APSM have been printed by 0.5 μm defocus at a partial coherence of 0.31 using KrF scanner with highest numerical aperture of 0.68. However, these holes were closed with in-focus imaging. Defocus is also found to be beneficial for patterning the pitches that have extensive side lobes with in-focus imaging.     

Exposed Die-Top Encapsulation Molding for an Improved High- Performance Flip Chip BGA Package

The recent advancement in high- performance semiconductor packages has been driven by the need for higher pin count and superior heat dissipation. A one-piece cavity lid flip chip ball grid array (BGA) package with high pin count and targeted reliability has emerged as a popular choice. The flip chip technology can accommodate an I/O count of more than five hundreds500, and the die junction temperature can be reduced to a minimum level by a metal heat spreader attachment. None the less, greater expectations on these high-performance packages arose such as better substrate real estate utilization for multiple chips, ease in handling for thinner core substrates, and improved board- level solder joint reliability. A new design of the flip chip BGA package has been looked into for meeting such requirements. By encapsulating the flip chip with molding compound leaving the die top exposed, a planar top surface can be formed. A, and a flat lid can then be mounted on the planar mold/die top surface. In this manner the direct interaction of the metal lid with the substrate can be removed. The new package is thus less rigid under thermal loading and solder joint reliability enhancement is expected. This paper discusses the process development of the new package and its advantages for improved solder joint fatigue life, and being a multichip package and thin core substrate options. Finite-element simulations have been employed for the study of its structural integrity, thermal, and electrical performances. Detailed package and board-level reliability test results will also be reported     

宽带DDS跳频源设计

直接数字合成（DDS）简单可靠、控制方便，具有很高的频率分辨率，高速转换，非常适合快速跳频的要求。在对DDS基本原理进行了简要介绍和分析后，提出宽带跳频源设计方案。     

On the convergence of the inverses of Toeplitz matrices and its applications

Many issues in signal processing involve the inverses of Toeplitz matrices. One widely used technique is to replace Toeplitz matrices with their associated circulant matrices, based on the well-known fact that Toeplitz matrices asymptotically converge to their associated circulant matrices in the weak sense. This often leads to considerable simplification. However, it is well known that such a weak convergence cannot be strengthened into strong convergence. It is this fact that severely limits the usefulness of the close relation between Toeplitz matrices and circulant matrices. Observing that communication receiver design often needs to seek optimality in regard to a data sequence transmitted within finite duration, we define the finite-term strong convergence regarding two families of matrices. We present a condition under which the inverses of a Toeplitz matrix converges in the strong sense to a circulant matrix for finite-term quadratic forms. This builds a critical link in the application of the convergence theorems for the inverses of Toeplitz matrices since the weak convergence generally finds its usefulness in issues associated with minimum mean squared error and the finite-term strong convergence is useful in issues associated with the maximum-likelihood or maximum a posteriori principles.     

Mining predecessor–successor rules from DAG data

Data mining extracts implicit, previously unknown, and potentially useful information from databases. Many approaches have been proposed to extract information, and one of the most important ones is finding association rules. Although a large amount of research has been devoted to this subject, none of it finds association rules from directed acyclic graph (DAG) data. Without such a mining method, the hidden knowledge, if any, cannot be discovered from the databases storing DAG data such as family genealogy profiles, product structures, XML documents, task precedence relations, and course structures. In this article, we define a new kind of association rule in DAG databases called the predecessor–successor rule, where a node x is a predecessor of another node y if we can find a path in DAG where x appears before y. The predecessor–successor rules enable us to observe how the characteristics of the predecessors influence the successors. An approach containing four stages is proposed to discover the predecessor–successor rules. © 2006 Wiley Periodicals, Inc. Int J Int Syst 21: 621–637, 2006.     

一种改进的MC-CDMA系统结构

MC-CDMA在频域扩频时实际上是一种频率分集,对于频率分集重数的要求通常小于子载波的数目N,故扩频因子可以小于子载波数目。改进了这种MC-CDMA系统的结构,通过在频域引入交织器,使交织后频率分集各分量经历独立的哀落,从而提高了抗哀落能力。最后给出了用系统仿真软件SPW仿真的结果。