Efficient modeling of RF CMOS spiral inductors using generalized knowledge-based neural network

Efficient modeling of RF CMOS spiral inductors by virtue of a novel generalized knowledge-based neural network (GKBNN) is presented. Prior knowledge of on-chip inductors is used for constructing the GKBNN. This new modeling approach also exploits merits of the iterative multi stage algorithm. This GKBNN has much enhanced learning and generalization capabilities. Comparing with the conventional neural network or the knowledge-based neural network, this new GKBNN model can map the input–output relationships with fewer hidden neurons and has higher reliability for generalization. As a consequence, this GKBNN model can run as fast as an approximate equivalent circuit model yet generate results as accurate as detailed electromagnetic simulations. Experiments are included to demonstrate merits and efficiency of this new approach.     

An improved charge-averaging charge-pump scheme

An improved charge-averaging charge-pump scheme is described in the paper. The circuit mainly includes four pairs of switches and four sampling capacitors. Two operation modes (normal/charge-averaging) are achieved by switching on/off. SpectreVerilog simulation results show the proposed scheme can strongly reduce the energy of spurs, and could be applied to high performance frequency synthesizers.     

CMOS inverters based positive type second generation current conveyor

A new versatile class AB low-voltage second generation current conveyor based on CMOS inverters operating in transconductance mode is presented in this letter. Against traditional design based on CCII+, the circuit is able to operate at low supply voltages and offers numerous advantages like class AB operation, large voltage and current swing, synthesis from digital inverters. Simulation results from a typical 0.35 μm CMOS process had demonstrated the circuit capability to operate at high frequency over wide voltage and wide current swings. The proposed circuit operation has been acted from measurements with the HEF4069UBP from Philips semiconductors [1].     

Smart TDI readout circuit for long-wavelength IR detector

A smart time delay and integration (TDI) readout circuit is suggested which performs background suppression, cell-to-cell non-uniformity compensation, and dead pixel correction. Using the smart TDI readout circuit, the integration capacitor area occupying almost the whole area of a unit-cell can be reduced to one-fifth and transimpedance gain can increase by five times. From measurement results, it is found that the skimming current error for a few hundred nA background current is < 1.25 nA corresponding to LSB/2 of ADC and the non-uniformity introduced by cell-to-cell background current variation is reduced to 1.02 nA     

Simulation and design methodology for a 50-Gb/s multiplexer/demultiplexer package

A 50 Gb/s package for SiGe BiCMOS 4:1 multiplexer and 1:4 demultiplexer targeting SONET OC-768 serial communication systems is introduced in this work. The package was designed to facilitate bit-error-rate tests and constructed with high-speed coaxial connectors, transmission lines on ceramic substrate, ribbon bonds for chip-to-package interconnects, and a metal composite housing. Numerical simulations were conducted to guide the package design, and both small signal measurements and operational tests were performed thereafter to verify the design and modeling concepts. To keep the model structure under the existing computing capability, the simulation was segmented into three sections - coaxial connector to transmission line, transmission line alone, and transmission line to ribbon bond, and then the results were assembled to predict the performance of the entire package. The package was operated up to 50 Gb/s with low degradation to input digital waveforms and free of error.     

Characterization of picosecond electric-pulse propagation on CPW components by transient near-field mapping

For the first time, the picosecond electric-pulse propagation characteristics of a few coplanar waveguide (CPW) components, including transmission lines, capacitive open and short terminations and a right-angle bend, have been successfully measured by mapping the time-dependent tangential near-electric-field distributions. The measurement was performed using a two-dimensional photoconductive (PC) near-field mapping system incorporating a recently developed novel PC electric-field probe, which could measure the separate picosecond orthogonal electric-field components with minimal loading effects. The resulting field-images mapped over the structural capacitive open termination and the short termination, showed previously unknown, and quite remarkable, transient picosecond electric-pulse reflection phenomena, which cannot be measured using conventional test instruments. In addition, the images showed that reflection-phenomenon itself required about 1 ps of propagation delay, and that the pulse-reflection from the capacitive open-termination required additional time delay matched to the length of the open-gap between the two slots of the CPW.     

柔性有源OLED显示器制造几点技术方面的考虑

AMOLED在柔性显示领域前途似锦。LG显示展示了一款全彩4in柔性AMOLED样机,衬底为80μm厚的不锈钢薄片,曲率达到5cm弯曲半径。本文就此项柔性显示技术所面临的挑战进行了探讨．包括柔性衬底衬底的传送、如何获得特性稳定可靠的TFT以使OLED的亮度和一致性适用于此项技术的商业化推广。     

Signaling and Control Procedures Using Generalized MPLS Protocol for IP over an Optical Network

This paper reviews the existing research activities on signaling and control procedures for IP over optical networks. We focus on the IP‐centric signaling and control architecture based on the generalized multi‐protocol label switching (GMPLS) protocol and analyze various scenarios and technical issues for deploying the IP over an optical network. We analyze the signaling and operations and administration and maintenance requirements for integrating an IP network and an optical network in order to cope with the high bandwidth and poor resource granularity of the optical network, including the optical cross‐connect system. On the basis of network architecture and a reference configuration model, we investigate the GMPLS‐based control architecture and interconnection model appropriate for controlling IP bandwidth and optical lambda resources. The signaling and control procedure based on GMPLS on optical user‐network interface and network‐network interface are comparatively investigated to provide the optical lightpath. We also study protection and restoration procedures to protect link failure when it applies to GMPLS signaling.     

An on-demand TV service architecture for networked home appliances

The recent advent of networked home appliances has enabled enhanced on-demand multimedia services to be feasible. Especially, personal video recorders have made it more convenient to manage broadcast TV programs at home by storing them on their own disks. However, standalone PVRs have limitations on availability of TV programs due to limited reception channel and storage capacity. Networked PVR services can get over these limitations by storing all broadcast TV programs in remote storage systems that servers provide. Thus, these networked PVR services can have several advantages over standalone PVRs in terms of cost and availability. In this article we therefore propose an on-demand TV service architecture for networked PVRs to reduce response times to interactive operations and save network bandwidth. Finally, we show from simulation results that our proposed architecture both decreases the number of programs whose initial time-shifting operations are delayed and increases the hit ratio in cache servers significantly.     

Blind identification of a class of nonlinear systems with cyclostationary input

This letter deals with blind identification of nonlinear discrete Hammerstein system under the input signal that is cyclostationary. The first-order moment of the specific input as well as the inverse nonlinear mapping of the Hammerstein model are combined to establish a relationship between the system output and the system parameters, which implies an approach to identifying the system blindly. Simulation results demonstrate the effectiveness of this approach to blind identification of a class of nonlinear systems.