Parallel, self-organizing, hierarchical neural networks. II

For pt.I see IEEE Trans. Neural Networks, vol.1, p.167-78 (1990). Parallel, self-organizing, hierarchical neural networks (PSHNNs) involve a number of stages with error detection at the end of each stage, i.e., rejection of error-causing vectors, which are then fed into the next stage after a nonlinear transformation. The stages operate in parallel during testing. Statistical properties and the mechanisms of vector rejection of the PSHNN are discussed in comparison to the maximum likelihood method and the backpropagation network. The PSHNN is highly fault tolerant and robust against errors in the weight values due to the adjustment of the error detection bounds to compensate errors in the weight values. These properties are exploited to develop architectures for programmable implementations in which the programmable parts are reduced to on-off or bipolar switching operations for bulk computations and attenuators for pointwise operations     

Synthesis of ω-amino- and ω-carboxyalkylamides of N-acetylmuramoyl-L-alanyl-D-isoglutamine

The synthesis has been effected of derivatives of MDP having a spacer with an amino or carboxy group. The final stage of the synthesis was the condensation of Boc-L-Ala-D-iGln with 6-aminohexanol (followed by the two-stage replacement of the hydroxy by an azido group) or with benzyl 6-aminohexanoate.Simferopol' State University. Translated from Khimiya Prirodnykh Soedinenii, No. 1, pp. 122–125, January–February, 1993.     

Synthesis of furaxanenitrolic acids

Previously unreported furoxanenitrolic acids have been synthesized.Translated from Khimiya Geterotsiklicheskikh Soedinenii, No. 8, pp. 1117–1119, August, 1993.     

Photolithographic packaging with selectively occupied repeatedtransfer (PL-Pack with SORT) for scalable film optical linkmultichip-module (S-FOLM)

"Photolithographic packaging (PL-pack) with selectively occupied repeated transfer (SORT)" is proposed for optoelectronic microsystem integration. PL-pack with SORT integrates different types of thin-film device pieces into one substrate with desired configurations using an all-photolithographic process. A process design example is presented for a scalable film optical link multichip-module (S-FOLM). A preliminary estimation reveals that PL-Pack with SORT will achieve III-V epitaxial material saving of <1/100 and module cost reduction of <1/10, compared with flip-chip-bonding-based packaging. The result indicates that the process will save on cost and resources simultaneously. A critical issue is how to simplify the procedure for distributing thin-film device pieces onto a substrate. SORT is found to reduce the distribution step count typically by factor of <1/10-1/10000 compared with the conventional one-by-one method. PL-pack with SORT will be extended to the 3R process (reduce, reuse, recycle), which is generally applied to a variety of device/module fabrications     

Using dynamic data exchange to exchange information between Visual Basic and Matlab : Application to a diode array spectrophotometer

Photodiode array spectrometers are increasingly being coupled to flow analysis and separation techniques. Computer programs are needed to control the data acquisition from each photodiode and to handle the data. The Microsoft Visual Basic programming language is used along with a DLL library and dynamic data exchange (DDE) to allow communication with the MATLAB computational environment.     

Far-infrared pump-probe measurement of the lifetime of the 2p−1 shallow donor level in n-GaAs

The low temperature lifetime of electrons excited in the 2p^?1 donor level of n-GaAs has been studied in a far-infrared pump-probe experiment. The measurement has been carried out using a pulsed far-infrared molecular gas laser working at a wavelength of 292µm, with the sample in a magnetic field of 5.1 T, resonant with the 1s_o?2p^?1 transition. Two FIR pulses are sliced from one FIR-laser pulse by means of optical switching techniques using two Q-switched Nd:YAG lasers. The first pulse is used to saturate the transition, while the second pulse probes the return of the population in the excited state towards thermal equilibrium as a function of the time delay after the excitation pulse. The value of 350±50 ns found for the lifetime falls in line with CW saturation results on materials with other doping concentrations.     

High speed p-type SiGe modulation-doped field-effect transistors

We report on the fabrication and characterization of high-speed p-type modulation-doped field-effect transistors (MODFETs) with 0.7-μm and 1-μm gate-lengths having unity current-gain cut-off frequencies (f_T) of 9.5 GHz and 5.3 GHz, respectively. The devices were fabricated on a high hole mobility SiGe heterostructure grown by ultra-high-vacuum chemical vapor deposition (UHV-CVD). The dc maximum extrinsic transconductance (g_m) is 105 mS/mm (205 mS/mm) at room temperature (77 K) for the 0.7-μm gate length devices. The fabricated devices show good pinch-off characteristics and have a very low gate leakage current of a few μA/mm at room temperature and a few nA/mm at 77 K     

线状锗等离子体X射线线谱的时间分辨测量

用自行研制的X射线条纹晶体谱仪首次测量了线状锗等离子体的X射线时间分辨谱。给出了类Ne-锗L线共振线的时间演化过程,并用类Ne-锗L线共振线与其双电子俘获伴线的相对强度比粗估了锗等离子体的电子温度及其随时间的变化,实验给出了X光激光增益区介质的电子温度为400～760eV,同时给出了电子温度保持相对恒定的时间不小于90ps(电子温度变化小于2％)。     

On the optimum set of polarization measurements for the scattering of two indentical fermions to determine the helicity amplitudes

The polarization asymmetries related to usual polarizations combine at least 4 different helicity amplitudes. The introduction of ±45° inclined transverse polarizations allows us to obtain the complete set of new polarization asymmetries combining only up to 3 helicity amplitudes. There are no further polarization asymmetries depending on a smaller number of helicity amplitudes. These expressions are most economical to determine the helicity amplitudes from observed data. Some of them are suitable to study especially the spin-flip effects. We give a complete Table of all such polarization asymmetries.     

A generalized suboptimum unequally spaced channel allocationtechnique. I. In IM/DD WDM systems

Four-wave mixing (FWM) is the most serious fiber nonlinearity associated with low-input optical power levels in long-haul multichannel optical systems employing dispersion-shifted fiber. To reduce the crosstalk due to FWM, a generalized suboptimum unequally spaced channel allocation (S-USCA) technique is proposed and investigated. Even though the developed technique is useful in combating FWM crosstalk in wavelength division multiplexing (WDM) lightwave systems with up to 12 channels, its main virtue is in designing multichannel WDM lightwave systems with more than 12 channels. Comparisons of power penalty due to FWM between equal channel spacing (ECS) systems and the S-USCA systems are presented. It is shown that for an intensity modulation/direct detection (IM/DD) transmission system operating in an optical bandwidth of 16 nm with 0 dBm (1 mW) peak optical input power per channel, while a conventional ECS WDM system with 0.84-nm channel spacing cannot even achieve a bit-error rate (BER)=10^-9, the suboptimum technique developed in this paper, for the same minimum channel spacing, can achieve a BER=10^-9 with an FWM crosstalk power of less than 1 dB at the worst channel in a 20-channel WDM system