Nonvolatile Memory Effect in Indium Gallium Arsenide-Based Metal–Oxide–Semiconductor Devices Using II–VI Tunnel Insulators

This paper reports the successful use of ZnSe/ZnS/ZnMgS/ZnS/ZnSe as a gate insulator stack for an InGaAs-based metal–oxide–semiconductor (MOS) device, and demonstrates the threshold voltage shift required in nonvolatile memory devices using a floating gate quantum dot layer. An InGaAs-based nonvolatile memory MOS device was fabricated using a high-κ II–VI tunnel insulator stack and self-assembled GeO _x -cladded Ge quantum dots as the charge storage units. A Si₃N₄ layer was used as the control gate insulator. Capacitance–voltage data showed that, after applying a positive voltage to the gate of a MOS device, charges were being stored in the quantum dots. This was shown by the shift in the flat-band/threshold voltage, simulating the write process of a nonvolatile memory device.     

A Distortion-Sensitive Seam Carving Algorithm for Content-Aware Image Resizing

As displays become less expensive and are incorporated into more and more devices, there has been an increased focus on image resizing techniques to fill an image to an arbitrary screen size. Traditional methods such as cropping or resampling can introduce undesirable losses in information or distortion in perception. Recently, content-aware image retargeting methods have been proposed (Avidan and Shamir, ACM Trans Graphics 26(3), 2007; Guo et al., IEEE Trans Multimedia 11(5):856–867, 2009; Shamir and Avidan, Commun ACM 52(1), 2009; Simakov et al. 2008; Wolf et al. 2007), which produce exceptional results. In particular, seam carving, proposed by Avidan and Shamir, has gained attention as an effective solution. However, there are many cases where it can fail. In this paper we propose a distortion-sensitive seam carving algorithm for content-aware image resizing that improves edge preservation and decreases aliasing artifacts. In the proposed approach, we use local gradient information along with a thresholding technique to guide the seam selection process and provide a mechanism to halt seam carving when further processing would introduce unacceptable visual distortion in the resized image. Furthermore, anti-aliasing filter is used to reduce the aliasing artifacts caused by seam removal. Experiments have demonstrated superior performance over the current seam carving methods.     

The effects of boron penetration on p+ polysilicon gatedPMOS devices

The penetration of boron into and through the gate oxides of PMOS devices which employ p⁺ doped polysilicon gates is studied. Boron penetration results in large positive shifts in V_FB , increased PMOS subthreshold slope and electron trapping rate, and decreased low-field mobility and interface trap density. Fluorine-related effects caused by BF₂ implantations into the polysilicon gate are shown to result in PMOS threshold voltage instabilities. Inclusion of a phosphorus co-implant or TiSi₂ salicide prior to gate implantation is shown to minimize this effect. The boron penetration phenomenon is modeled by a very shallow, fully-depleted p-type layer in the silicon substrate close to the SiO ₂/Si interface     

Operation of integrated InGaAsP-InP optical amplifier-monitoringdetector with feedback control circuit

Monolithic integration of a monitoring detector with an optical amplifier simplifies the use of an amplifier in lightwave systems. The structure and performance are described of a monolithically integrated semiconductor optical amplifier with low-loss Y-branching waveguides and a monitoring p-i-n detector. The photocurrent of the integrated detector can be used as a single control parameter for amplifier output leveling, gain optimization, and in situ monitoring of facet antireflective coatings     

Dispersion-induced ultrafast pulse reshaping in 1.55-/spl mu/m InGaAs-InGaAsP optical amplifiers

Using the Foreman effective mass Hamiltonian, the electronic structure of the valence band and the interband dipole matrix elements in In/sub x/Ga/sub 1-x/As-In/sub y/Ga/sub 1-y/As/sub z/P/sub 1-z/ quantum-well optical amplifiers are calculated, taking into account the valence band mixing and the biaxial strain. The optical field of the amplified pulse is calculated by solving the wave equation with the computed polarization as a source term. A novel wavelet transform is introduced in analyzing the pulse chirp imposed by the optical amplifier. In the linear propagation regime, the spectrum of the amplified pulse can be either red-shifted or blue-shifted with respect to its initial center frequency, depending on the local gain dispersion spanned by the pulse spectrum. The output pulse shape can be retarded or advanced, depending on the local gain and group velocity dispersion. Furthermore, an initially unchirped pulse centered in the tail of the gain spectrum is significantly reshaped after propagating 600 /spl mu/m, and its spectrum is broadened and distorted considerably. In the spectral region where both gain and group velocity change rapidly, the frequency chirp for a linearly chirped input pulse is significantly weakened after propagation.     

Semiconductor Yield Analysis and Multi-Chip Package(MCP) Dic Pairing Optimization using Machine Learning

<正>~~     

Decoupled multiuser code-timing estimation for code-divisionmultiple-access communication systems

We present herein a decoupled multiuser acquisition (DEMA) algorithm for code-timing estimation in asynchronous code-division multiple-access (CDMA) communication systems. The DEMA estimator is an asymptotic (for large data samples) maximum-likelihood method that models the channel parameters as deterministic unknowns. By evoking the mild assumption that the transmitted data bits for all users are independently and identically distributed, we show that the multiuser timing estimation problem that usually requires a search over a multidimensional parameter space decouples into a set of noniterative one-dimensional problems. Hence, the proposed algorithm is computationally efficient. DEMA has the desired property that, in the absence of noise, it obtains the exact parameter estimates even with a finite number of data samples which can be heavily correlated. Another important feature of DEMA is that it exploits the structure of the receiver vectors and, therefore, is near-far resistant. Numerical examples are included to demonstrate and compare the performances of DEMA and a few other standard code-timing estimators     

Scalable VLSI implementations for neural networks

This paper discusses research on scalable VLSI implementations of feed-forward and recurrent neural networks. These two families of networks are useful in a wide variety of important applications—classification tasks for feed-forward nets and optimization problems for recurrent nets—but their differences affect the way they should be built. We find that analog computation with digitally programmable weights works best for feed-forward networks, while stochastic processing takes advantage of the integrative nature of recurrent networks. We have shown early prototypes of these networks which compute at rates of 1–2 billion connections per second. These general-purpose neural building blocks can be coupled with an overall data transmission framework that is electronically reconfigured in a local manner to produce arbitrarily large, fault-tolerant networks.     

On the scalability of ad hoc networks

We investigate the inherent scalability problem of ad hoc networks originated from the nature of multihop networks. First, the expected packet traffic at the center of a network is analyzed. The result shows that the expected packet traffic at the center of a network is linearly related with the network size, that is, the expected packet traffic at the center of a network is O(k), where k is the radius of a network. From the result, the upper bound of the diameter of a network D=2k, that guarantees the network is scalable, is obtained. The upper bound is given by C/r-1, where C is the channel capacity available to each node and r is the packet arrival rate at each node.