The cooperative and competitive network suitable for circuit realization is presented, based on the network proposed by Amari and Arbib. To ensure WTA process, the output function of the original network is replaced with the piecewise linear function and supplying the inputs as pulse waveforms is obtained. In the SPICE simulations, it is confirmed that the network constructed by operational amplifiers attains WTA process, even if the scale of the network becomes large.

A GaInNAs alloy on GaAs substrate has been very promising for long-wavelength vertical-cavity surface-emitting lasers (VCSELs) as an active layer. In spite of many groups reported the excellent temperature characteristics of the threshold current of the GaInNAs/GaAs edge-emitting lasers, discussions of the temperature dependence of the lasing characteristics except threshold current is few. In this paper, temperature characteristics of GaInNAs lasers grown by chemical beam epitaxy (CBE) emitting at λ=1.27 µm and λ=1.30 µm were investigated in detail. The characteristic temperature (T0) ranging from 10 to 80 varies from 60 to 130 K and decreased with decreasing cavity length for shorter cavity (< 400 µm) devices. On the other hand, longer cavity (< 400 µm) devices show that the cavity length does not affect so much to T0. The internal losses did not increase with increasing temperature. On the other hand, internal quantum efficiencies decreased with increasing temperature. It is considered that non-radiative recombination center with large temperature dependence may influence the decrease of the internal quantum efficiency due to the insufficient crystal quality of GaInNAs layer. The transparency current densities were unchanged for all temperature range, however, the gain constants decreased with increasing temperature. Thus, the decrease of the gain constant is considered to be due to decreasing of gain. Unchanged both transparency current density and internal loss may also express that these temperature characteristics were not induced by carrier overflow but be done by decreasing of the gain. From the results, it is considered that the temperature dependence of the gain originated from the Fermi-Dirac distribution of carriers was dominant for the temperature characteristics of GaInNAs/GaAs lasers. Due to the temperature dependence on the gain, the T0 decreases with increasing mirror loss.

In this paper, we describe a frame synchronization method for bi-orthogonal modulation systems. In bi-orthogonal modulation systems, several bi-orthogonal sequences are used for data transmission. Frame synchronization in bi-orthogonal modulation systems is difficult because transmitted sequences can change every frame. In the proposed method, each bi-orthogonal sequence consists of two different inner sequences. Each bi-orthogonal sequence has the same arrangement of two different inner sequences. A receiver can track the frame timing by observing the arrangement of inner sequences. In this paper, we analyze the bit error rate performance that takes into account the tracking performance of a system we developed based on our method. The spectral efficiency of the proposed system in code division multiple access (CDMA) systems is also investigated. As a result, we found that the proposed system is effective in synchronous CDMA systems.

We report here on the electrical and structural characteristics of InGaP/GaInAsN DHBTs with up to a 50 mV reduction in turn-on voltage relative to standard InGaP/GaAs HBTs. High p-type doping levels ( 3 1019 cm-3) and dc current gain (βmax up to 100) are achieved in GaInAsN base layer structures ranging in base sheet resistance between 250 and 750 Ω/. The separate effects of a base-emitter conduction band spike and base layer energy-gap on turn-on voltage are ascertained by comparing the collector current characteristics of several different GaAs-based bipolar transistors. Photoluminescence measurements are made on the InGaP/GaInAsN DHBTs to confirm the base layer energy gap, and double crystal x-ray diffraction spectrums are used to assess strain levels in the GaInAsN base layer.

A novel InGaAs/InAlAs insulated gate (IG) pseudomorphic high electron mobility transistor (PHEMT) having a silicon interface control layer (Si ICL) is successfully fabricated and characterized. Systematic efforts to characterize and optimize the insulated gate structure and the PHEMT fabrication process were made by using in-situ X-ray photoelectron spectroscopy (XPS) and capacitance-voltage (C-V) techniques. This led to successful fabrication of a novel IG-PHEMT showing excellent stable DC characteristics with a good pinch off and a high transconductance (177 mS/mm), very small gate leakage currents, very high gate breakdown voltages (about 40 V) and respectable RF characteristics fT = 9 GHz and fmax=38 GHz.

This paper describes the implementation of a proposed image filter into a Discrete-Time Cellular Neural Network (DT-CNN). The three stages that compose the filter are described, showing that the resultant filter is capable of (1) erasing or detecting several concentric shapes simultaneously, (2) thresholding and (3) thinning of gray-scale images. Because the DT-CNN has to fill certain conditions for this filter to be implemented, it becomes a modified version of a DT-CNN. Those conditions are described and also experimental results are clearly shown.

Detection of nonlinearly distorted signals is an essential problem in telecommunications. Recently, neural network combined conventional equalizer has been used to improve the performance especially in compensating for nonlinear distortions. In this paper, the self-organizing map (SOM) combined with the conventional symbol-by-symbol detector is used as an adaptive detector after the output of the decision feedback equalizer (DFE), which updates the decision levels to follow up the nonlinear distortions. In the proposed scheme, we use the box distance to define the neighborhood of the winning neuron of the SOM algorithm. The error performance has been investigated in both 16 QAM and 64 QAM systems with nonlinear distortions. Simulation results have shown that the system performance is remarkably improved by using SOM detector compared with the conventional DFE scheme.

This study presents two new bit-parallel cellular multipliers based on an irreducible all one polynomial (AOP) over the finite field GF(2m). Using the property of the AOP, this work also presents an efficient algorithm of inner-product multiplication for computing AB2 multiplications is proposed, with a structure that can simplify the time and space complexity for hardware implementations. The first structure employs the new inner-product multiplication algorithm to construct the bit-parallel cellular architecture. The designed multiplier only requires the computational delays of (m+1)(TAND+TXOR). The second proposed structure is a modification of the first structure, and it requires (m+2) TXOR delays. Moreover, the proposed multipliers can perform A2iB2j computations by shuffling the coefficients to make i and j integers. For the computing multiplication in GF(2m), the novel multipliers turn out to be efficient as they simplify architecture and accelerate computation. The two novel architectures are highly regular, simpler, and have shorter computation delays than the conventional cellular multipliers.

Our recent progress in improving the performance of the GaInNAs laser is fully reviewed here. We improved the crystal quality of GaInNAs by optimizing the conditions for its grown by gas-source molecular beam epitaxy (MBE) using N radicals as a N source. We found that the temperature window for obtaining GaInNAs with high crystal quality, good surface morphology, and good photoluminescence (PL) characteristics is smaller than that for obtaining this kind of GaInAs. Like dopant atoms such as Si or Be in GaAs, the N radicals produced by an RF discharge have a high sticking coefficient. Their use is therefore effective when we want to increase and control the N content of GaInNAs. We found that the AsH3-flow-rate mainly affected crystal quality of GaInNAs rather than incorporation of nitrogen atoms. We also investigated the effects of thermal annealing on the optical properties of as-grown GaInNAs layers and found that it greatly increased the PL intensity and produced the large shift in the PL wavelength. The absorption spectra of the GaInNAs bulk layer revealed that the large shift in the PL wavelength is probably caused by a bandgap shift in the GaInNAs well layer, and cathodeluminescence measurements revealed that the increased PL intensity is due to the improved emission being more uniform spatially: uniformity from the entire region; in comparison, nonuniform dot-like regions exist in an as-grown GaInNAs layer. Optimizing the growth conditions and using thermal annealing effect, we made a 1.3-µm GaInNAs/GaAs single-quantum-well laser that has a high characteristic temperature (215 K) under pulsed operation. To our knowledge, this is the highest characteristic temperature reported for a 1.3-µm band-edge emitter suitable for used in optical-fiber communication systems. The use of GaInNAs as an active layer is, therefore, very promising for the fabrication of long-wavelength laser diodes with excellent high-temperature performance.

Investigations were carried out on metalorganic-chemical-vapor-deposition (MOCVD)-grown strained AlGaN/ GaN/sapphire structures using X-ray diffratometry. While AlGaN with lower AlN molar fraction (< 0.1) is under the in-plane compressive stress, it is under the in-plane tensile stress with high AlN molar fraction (> 0.1). Though tensile stress caused the cracks in AlGaN layer with high AlN molar fraction, we found that the cracks dramatically reduced when the GaN layer quality was not good. Using this technique, we fabricated a GaInN multi-quantum-well (MQW) surface emitting diodes were fabricated on 15 pairs of AlGaN/GaN distributed Bragg reflector (DBR) structures. The reflectivity of 15 pairs of AlGaN/GaN DBR structure has been shown as 75% at 435 nm. Considerably higher output power (1.5 times) has been observed for DBR based GaInN MQW LED when compared with non-DBR based MQW structures.

An analog vector quantization processor has been designed based on the neuron-MOS (νMOS) technology. In order to achieve a high integrating density, template information is merged into the matching cell (the absolute value circuitry) using the νMOS ROM technology. A new-architecture νMOS winner-take-all (WTA) circuit is employed for fully-parallel search for the minimum-distance vector. The WTA performs multi-resolution winner search with an automatic feedback gain control. A test chip having 256 16-element fixed template vectors has been built in a 1.5-µm double-polysilicon CMOS technology with the chip size of 7.2 mm 7.2 mm, and the basic operation of the circuits has been demonstrated.

We propose a new associative memory named Multi-Winner Associative Memory (MWAM) and study its bidirectional association properties in this paper. The proposed MWAM has two processes for pattern pairs storage: storage process and recall process. For the storage process, the proposed MWAM can represent a half of pattern pair in the distributed representation layer and can store the correspondence of pattern and its representation using the upward weights. In addition, the MWAM can store the correspondence of the distributed representation and the other half of pattern pair in the downward weights. For the recall process, the MWAM can recall information bidirectionally: a half of the stored pattern pair can be recalled by receiving the other half in the input-output layer for any stored pattern pairs.

The objective of thinning is to reduce the amount of information in image patterns to the minimum needed for recognition. Thinned image helps the extraction of important features such as end points, junction points, and connections from image patterns. The ultimate goal of parallel algorithms is to minimize the execution time while producing high quality thinned image. Though much research has been performed for parallel thinning algorithms, there has been no systematical approach for comparing the execution speed of parallel thinning algorithms. Several rough comparisons have been done in terms of iteration numbers. But, such comparisons may lead to wrong guides since the time required for iterations varies from one algorithm to the other algorithm. This paper proposes a formal method to analyze the performance of parallel thinning algorithms based on PRAM (Parallel Random Access Machine) model. Besides, the quality of skeletons, robustness to boundary noise sensitivity, and execution speed are considered. Six parallel algorithms, which shows relatively high performance, are selected, and analyzed based on the proposed analysis method. Experiments show that the proposed analysis method is sufficiently accurate to evaluate the performance of parallel thinning algorithms.

We have proposed a non-invasive method for diagnosis of the early stage of atherosclerosis, namely, the detection of small vibrations on the aortic wall near the heart by using ultrasound diagnostic equipment. It is, however, necessary to confirm the effectiveness of such measurement of the pulse wave velocity for quantitative evaluation of the local characteristics of atherosclerosis. It is well known that Young's modulus of a tube wall, estimated from measured pulse wave velocity, depends on inner pressure because of the non-linear relationship between the inner pressure and the change of volume in the tube. The inner pressure, however, changes during the period of one heartbeat. In this experimental study, we found for the first time that Young's modulus of the tube wall, estimated from the measured pulse wave velocity, depends not only on the diastolic pressure but also on the pulse pressure and the pressure gradient of the systolic period.

A simulation model for natural convection was developed for determining the surface temperature distribution in base plates with rectangular vertical fins in communication equipment. An estimated velocity derived from the buoyancy and pressure drop equations in a duct was used for laminar forced convection cooling simulations in parallel plates. Temperature distributions in finned plates were calculated by numerical integration of the heat conduction equation. An experimental study was also performed, to check these simulation results, by changing the height of fins, the pitch of fins, and the heat generation conditions. Experimental results and analytical results were found to agree well. Also, this simulation method was extended to analyze natural convection cooling in vertical base plates with inclined parallel fins. We placed alternately on the plates the sections without fins and the sections with fins on the plates. Using the inclined fins, air flow rate between fins was large and fresh air flew into the fins from the side of the plates. The natural convective heat-transfer rate for inclined fins was found to be 14% higher than that for vertical fins.

The basic operation characteristics of an asymmetric turnstile which transfers each electron one by one in one direction is described. A novel single electron counter circuit consisting of the asymmetric turnstiles, a load capacitor and an inverter which counts the number of high inputs is proposed. Monte Carlo circuit simulations reveal that the gate clock time of the counter circuit should be long enough to achieve allowable minimum error rate. The counter circuit implementing asymmetric single electron turnstiles is demonstrated to be applicable to a noise reduction system, a Winner-Take-All circuit and an artificial neuron circuit.

This paper focuses on competitive learning algorithms for WTA (winner-take-all) networks which perform rotation invariant pattern classification. Although WTA networks may theoretically be possible to achieve rotation invariant pattern classification with infinite memory capacities, actual networks cannot memorize all input data. To effectively memorize input patterns or the vectors to be classified, we present two algorithms for learning vectors in classes (LVC1 and LVC2), where the cells in the network memorize not only weight vectors but also their firing numbers as statistical values of the vectors. The LVC1 algorithm uses simple and ordinary competitive learning functions, but it incorporates the firing number into a coefficient of the weight change equation. In addition to all the functions of the LVC1, the LVC2 algorithm has a function to utilize under-utilized weight vectors. From theoretical analysis, the LVC2 algorithm works to minimize the energy of all weight vectors to form an effective memory. From computer simulation with two-dimensional rotated patterns, the LVC2 is shown to be better than the LVC1 in learning and generalization abilities, and both are better than the conventional Kohonen self-organizing feature map (SOFM) and the learning vector quantization (LVQ1). Furthermore, the incorporation of the firing number into the weight change equation is shown to be efficient for both the LVC1 and the LVC2 to achieve higher learning and generalization abilities. The theoretical analysis given here is not only for rotation invariant pattern classification, but it is also applicable to other WTA networks for learning vector quantization.

An alternative approach to increasing the functional capability of an integrated circuit chip other than the conventional scaling approach is presented and discussed. We will show the functional enhancement at a very elementary device level is essential in implementing intelligent functions at a system level. The concept of a four-terminal device is reviewed as a guiding principle in considering the device functionality enhancement. As an example of a four-terminal device, the neuron MOS transistor is presented. Applications of neuron MOS transistors to several new architecture circuits are demonstrated and the possibility of implementing intelligent functions directly on integrated circuit hardware is discussed.

This letter proposes a map-matching method for automotive navigation systems. The proposed method utilizes the innovation of the Kalman filter algorithm and can achieve more accurate positioning than the correlation method which is generally used for the navigation systems. In this letter, the performance of the proposed algorithm is verified by some simulations.

We have studied liquid crystal (LC) alignment on UV-irradiated poly(vinyl cinnamate)(PVCi)films by using the texture observations and the anchoring energy measurements. Irradiation of the PVCi films with linearly-polarized UV light creates the optical anisotropy in the films, and the anisotropy can well align LCs perpendicular to the UV polarization. We discuss the LC alignment mechanisms and point out the important contribution from non-dimerized side chains of PVCi molecules. The anchoring energies on photo-processed PVCi films are shown to be smaller than those on conventional rubbed polyimide films. We propose a new method to generate pretilt angle on the photo-processed PVCi films, and successful results of pretilt angle generation are demonstrated. This method is applied to fabricate TN- and super-multidomain TN-LCDs, and the good electro-optical performance of the LCDs is confirmed.