Z-scan measurements at 1600 nm on single-crystal PTS (p-toluene sulfonate) with single, 65 ps pulses gave a complex nonlinear refractive index coefficient of n2=2.2(±0.3)×10-12 cm2/W at 1 GW/cm2 and α2<0.5 cm/GW. This is the first highly nonlinear, organic material to satisfy the conditions imposed by the figures of merit 相似文献
An efficient hybrid modulation/demodulation scheme using a short duration pulse in the time-domain for ultra wideband-impulse radio (UWB-IR) systems is proposed. The proposed modulation scheme is pulse position modulation (PPM) of the UWB-IR standard modulation combined with differential encoding, and non-coherent energy detection (ED) adopting differential decoding is proposed for demodulation. Differential encoding makes a pulse that can transfer additive information bit into bits assigned in one symbol without increasing the symbol period. The BER performance is evaluated for 2-PPM, 4-PPM and the proposed HD-2PPM (which has the same symbol duration as BPPM and includes two information bits per symbol). The error performance indicates that the proposed scheme is an outstanding 0.5 dB over existing schemes of UWB-IR, and the data-rate performance shows that the proposed method has higher spectral efficiency than conventional methods that occupy the same duration as the proposed scheme.
A highly efficient single-input, dual-output AC–DC converter for wireless power transfer in implantable devices is implemented using the 0.18-µm CMOS process. The proposed AC–DC converter, consisting of three rectifiers with cross-coupled NMOS transistors and comparator-driven PMOS transistors, achieves up to 79.5% power conversion efficiency at 13.56 MHz operation frequency in order to provide dual outputs of 1.2 V and 2.2 V DC voltages along with 6.2 mA and 22.6 mA of current, respectively, to the implant device from a single RF input. The designed IC consumes a core die area of 0.18 mm2. 相似文献
We propose a new method for the optical implementation of the Hopfield neural network with a universal tool. The tool is a matrix grating constituted with a group of element gratings. The algorithms for designing a matrix grating are proposed, and a matrix grating is created to execute recognition experiments by use of the Hopfield neural network. The experimental results demonstrate that the proposed method performs well. The stability of the light efficiencies of different optical components used in optical networks is also considered. 相似文献
We searched numerically dielectric-loaded cylindrical waveguide configurations with an injected electron beam in which the growth rate of the cyclotron-Cherenkov instability surpassed that of the Cherenkov instability, and found such a configuration. This configuration consists of a metallic core and an outer metallic cylinder with a dielectric liner on the inner surface. In order to investigate experimentally radiation due to the cyclotron-Cherenkov instability, we designed and assembled an experimental device using the computational results. We studied beam propagation in the dielectric-loaded coaxial waveguide and microwave radiation due to the cyclotron-Cherenkov instability and the Cherenkov instability. 相似文献
A novel ultracompact 2/spl times/2 wavelength division multiplexer (WDM) for 1.55-/spl mu/m operation based on highly dispersive two-mode interference (TMI) was designed, theoretically modeled, and verified using a finite-difference-time-domain (FDTD) method. A two-moded waveguide assisted with a dispersive tooth-shaped grating provided a mode-dependent reflection band of central wavelength at 1.55 /spl mu/m. The wavelengths of 1538 and 1572 nm that were at the band edges and had the lowest reflection losses and relatively high dispersion were selected for wavelength multiplexing. The result showed that the wavelengths were separated by grating dispersion in a coupler length of 75 /spl mu/m which was much shorter than the required length of 1.1 mm in a regular TMI multiplexer of no grating. Insertion loss of about 1.7 dB and channel contrast of about 12 dB were achieved. 相似文献
A novel silicon carbide (SiC) normally off lateral channel vertical junction field-effect transistor (LC-VJFET), namely a source-inserted double-gate structure with a supplementary highly doped region (SHDR), was proposed for achieving extremely low power losses in high-power switching applications. The proposed architecture was based on the combination of an additional source electrode inserted between two adjacent surface gate electrodes and a unique SHDR in the vertical channel region. Two-dimensional numerical simulations for the static and resistive switching characteristics were performed to analyze and optimize the SiC LC-VJFET structures for this purpose. Based on the simulation results, the excellent performance of the proposed structure was compared with optimized conventional structures with regard to total power losses. Finally, the proposed structure showed about a 20% reduction in on-state loss (P/sub on/) compared to the conventional structures, due to the effective suppression of the JFET effect. Furthermore, the switching loss (P/sub sw/) of the proposed structure was found to be much lower than the results of the conventional structures, about a 75% /spl sim/ 95% reduction, by significantly reducing both input capacitance (C/sub iss/) and reverse transfer capacitance (C/sub rss/) of the device. 相似文献