Two kinds of cyclic switched-capacitor A/D converters using only one Op-Amp are presented. The effects of the nonidealities of the circuit such as a capacitor-ratio mismatch, an Op-Amp's offset voltage, and a parasitic capacitance, on a conversion accuracy are discussed.

This paper shows a novel automatic classification method for the electroencephalogram (EEG) based on syntactical pattern recognition. The syntactical method is effective to represent the complicated structure of the features of the EEG which contains transient-waves as well as the background-wave. For the extraction of transient-waves an adaptive autoregressive-moving average (ARMA) model fitting is utilized where the input to the model is replaced by the modified input if the prediction error grows more than a given threshold. By the adaptive ARMA model transient-waves and the spectrum of the background-wave are obtained from the prediction error and ARMA parameters, respectively. Since transient-waves may contain noisy patterns or variances, a relaxation scheme is applied. As the second stage all of the features of the EEG including the spectrum are described syntactically according to the generative grammar. Then the syntactical description Ti inherent to the diagnosis is obtained. In order to reduce the ambiguity and to suppress the complexity of syntactical descriptions, numerical values representing the details of EEG are separated from the syntactical description, and are added as the attributes (this method is generally called the attributed grammar). For the input EEG having syntactical description Ti, the final diagnostic decision is made by using the statistical Bayes estimation about the attributes within the group for Ti. As the result of automatic EEG classification for 200 EEG samples correct recognition of about 80 percent is observed.

On the basis of the controllability gramians, the observability gramians and the second order modes, this paper shows that optimal realizations (filter structures having minimum roundoff noises) of quarter-plane-causal, recursive and separable in denominator 2-D digital filters (CRSD filters for short) are scaled and rotated balanced realizations. Two applications of this relation are given. The first one gives a simple proof of the absence of overflow oscillations in optimal realizations. The second one, which is the main result of this paper, gives a direct design method of CRSD filters in the spatial domain. This method simplifies traditional two-step design (approximation and synthesis) into a one-step design with much less computational complexity. Resulting filters of this direct design method can approximate given 2-D impulse responses closely. In addition, they are always guaranteed to be stable, nearly optimal with respect to roundoff noise and free of overflow oscillations. The efficiency of the direct design method is shown by numerical examples.

This letter presents the theoretical analysis of error properties of coded modulation using several quantum state lights. The direct modulation of general code into light pulse has advantage in the case of the photon state light, while PPM is superior in the case of coherent state light.

The Karhunen-Loève (K-L) expansion of the continuous type is the expansion of a stochastic process with respect to the eigenfunctions of a correlation integral operator. In most practical applications, however, only a finite number of sample values are available. This paper proposes conditions and methods by which the exact eigenvalues and eigenfunctions of the integral operator can be obtained from those of a related matrix.

Mass-transport and conventional buying techniques were combined to fabricate narrow active stripe lasers. Lowest threshold current of 9.6 mA was obtained under CW operation for the active layer width of 1µm in 1.5 µm wavelength region. The maximum light output power and differential quantum efficiency were 10.6 mW and 17 percent/facet, respectively. With this structure, complete single transverse mode operation was achieved both for Fabry-Perot (FP) type and bundle-integrated-guide distributed Bragg reflecter (BIB-DBR) type lasers.

Jumbo is the three layer switch-box router based on Rivest and Fiduccia's greedy channel router. Jumbo uses three layers of wires to generate the solution with two choices: a. Two wiring layers for vertical routing and one for horizontal routingVHV b. Two wiring layers for horizontal routing and one for vertical routingHVH. The router scans first from left to right until no more split nets remain or reach the right hand side edge. If it hasn't reached the edge, then it scans in the opposite direction and stops at the column where it had stopped earlier where no more split nets had remained. Jumbo is capable of routing switchboxes and channels containing obstacles such as preexisting wiring by hand or by other routers. It jogs nets around multi-layer obstacles such as contacts, and routes over single-layer obstacles. If there are no obstacles, results for VHV channel routing are comparable to other good channel routers. Three layer HVH channel routing requires fewer tracks and even upto about half the number of tracks required by the two layer routing. Results from several benchmark examples are evaluated. Jumbo router found a 19 track VHV solution for Deutsch's difficult example. The computational complexity of the greedy algorithm is linearly proportional to the number of columns in a channel for the average case. Jumbo thus combines the effectiveness of traditional channel routers with the flexibility of net-at-a-time routers.

A theoretical analysis of the lasing gain and the threshold current in GaAs-AlGaAs Graded-Index Separate Confinement Hetero (GRIN-SCH) structure lasers is given. Energy levels and the state densities for electron distribution in both the well layer and the optical guiding regions are formulated. Confinement ratios of the injected electrons and holes and the optical field in the well layer are examined. The injection current, the electron lifetime due to the radiative recombination, the lasing gain and the threshold current are examined by taking into account the intraband relaxation effect of the electron wave. Optimum conditions to obtain low threshold current density are numerically discussed.

In order to reduce the threshold current of GaInAsP/InP surface emitting (SE) laser, we introduce a flat surface circular buried heterostructure (FCBH), which has merits of both planar buried heterostructure (PBH) and buried heterostructure with a SiO2 mask (BH), into GaInAsP/InP SE laser. Its minimum threshold current was 20 mA at 77 K, which provide better result than previous PBH-SE lasers or BH-SE lasers grown with a SiO2 mask.

Design of a waveguide-type optical isolator using the Faraday and Cotton-Mouton effects for nonreciprocal and reciprocal mode conversions, respectively, is presented. A noteworthy conclusion is obtained that, comparing with the previous design using a phase-matched waveguide, the device length is reduced by allowing phase-mismatch between TE and TM modes in the waveguide.

Quasi-periodic or harmonic oscillation transforms into fundamental oscillation as external force increases in forced van der Pol oscillator. Numerical analyses reveal that quasi-periodic or harmonic oscillation can exist metastably in a finite time interval even in a region where fundamental oscillation is stable, which has not been predicted theoretically.

A new parallel sorting algorithm is proposed, assuming a linearly connected multiprocessor system with a limited number of processors. The algorithm based on merging and exchanging data between adjacent processors. The upper bound is analytically investigated, and the throughput is proportional to the number of processors. The time complexity is O(N) for N data with O(log N) processors. Average number of iterations for a multiprocessor with two processing elements is also analytically investigated. Execution time for real situation is numerically simulated on a conventional computer.

Relations among three types of tree polynomials are clarified. A method for calculating those coefficients is given. Then introduced is an index of graph related to the coefficients. This index expresses to what an extent the given graph destroys spanning trees of the complete graph.

This paper considers a fleet system which consists of n equipments. The system is operated intermittently repeating a system operation period and a system standby period alternatively. In each system operation period, any c of n equipments are required to operate. Each equipment undergoes overhaul (O/H) after prespecified periods (or numbers) of operation. We obtain the availability of the system assuming the following disciplines for the O/H interval and the failure probability. The O/H interval is scheduled by calendar time (Case 1) or actual operating time (Case 2). The failure probability of an equipment is constant in each operation (Model 1) or depends on the number of operations after O/H (Model 2). This analysis aims at the determination of the rational O/H interval and the number of the equipments in the system. Also it can be applied to the problem of spares provisioning for a system with O/H.

An analytical method is presented for deriving the BER of optical OOK systems using soft decision Viterbi decoding. An APD and photon counter are employed as a detector, and the background optical noise and dark current noise of the APD are taken into account in the analysis. We employ two kinds of the soft decision boundary, which are the optimum boundary maximizing the cutoff rate R0 and the equal spacing boundary. For the symbol metric quantization methods both the optimum metric of the logarithm of the soft decision probability and the linear increment metric are utilized. Since the optical channel is not symmetric, we derive the tight upper-bound of the BER conditioned on the data sequence with the aid of the generating function of the employed code. Then averaging the conditional BER by the data sequence, we obtain the average BER. Numerical results of the BER are given for hard decision and 4-level soft decision comparing with the no coding case, and show the significance of the coding gain. Effects of the soft decision boundary and metric quantization method are also examined from the view points of both the BER and the cutoff rate R0. It is shown that, the linear increment metric system is easily realized, but the BER degrades rapidly with the scintillation. Therefore the boundary spacing should be changed according to the signal level in the linear increment metric case. Furthermore some computer simulations are made to confirm the validity of the analysis.

This paper gives a method for transforming attribute grammars into efficient action routines. An action routine description is a set of fragments of programs associated with production rules. Those fragments of programs are activated according to the ordering given by a bottom-up syntax analyzer. We present a transformation method which we call patch introduction. Patch introduction consists of techniques which we call extended postfix transformation and broadcast type inherited attribute elimination. Extended postfix transformation is an extension of simple postfix transformation. Broadcast type inherited attribute elimination is a technique of making the evaluation into a one-pass bottom-up evaluation. The scenario of patch introduction is as follows. First we reverse the flow of broadoast type inherited attributes, and transform those inherited attributes into synthesized attributes whose values are sets of addresses. Then we introduce generation of incomplete code where code is an extended postfix attribute, and introduce patch operations at the point where values of the broadcast type inherited attributes are determined. We illustrate the usefulness of our patch introduction method using a problem of translation of Boolean expressions into an intermediate code of short-circuit evaluation form.

This paper describes the design of a software development system for real-time controllers (or small embedded computers), which provides an advanced envimonment for developing and testing of software for a real-time controller. This system offers a concurrent language called *C and its development tool including a simulator for evaluating the execution time. *C is fundamentally based on the C language, and incorporates facilities for concurrency. The simulator intends to facilitate testing of a *C program to be executed on a real-time controller. The experimental system, which targets a single CPU real-time controller with a real-time monitor, is implemented on the UNIX operating system.

An integrated laser having increased rate of the spontaneous emission is proposed a apply this device as a light source giving weaken optical coherency. This device consists of two sections, one is a Light-Emitting-Diode (LED)-section and another is a Laser-section. The spontaneous emission is generated and amplified in the LED-section, then injected into the Laser-section. Power of the incoherent light in the Laser-section becomes more than ten times of that in the conventional single section laser with the help of the light injection. Wider half width of the output spectrum more than 1.5 nm is obtained at injection current level of I/Ith=1.5 in GaAs DH structure. Mechanism and fundamental properties of this device are analyzed with the help of both classical Maxwell's equation and quantum mechanics.

In this paper, a combinatorial problem oriented multicomputer system called DON (Double-Tree Structured Network Machine) is proposed. And a parallel branch-and-bound program scheme for the DON system is described. The DON system is composed of two binary-tree structured subsystems and a system controller. The DON system works as a post-end processor of a host computer system. The DON system is designed to achieve high parallelism and efficient pipeline ability. One of the most distinctive features of the DON system, compared to a conventional single-tree machine, is that the algorithms with pipeline features can be easily implemented and executed more efficiently. From the experimental results through simulation, it appears that the DON system can solve large scale combinatorial problems more efficiently than a conventional single-tree machine.

This paper introduces a calligraphic character writing system in which the generated characters are really written out on paper by a robot holding a writing brush. Being different from the conventional researches, the data base is made by means of a new method in which features of different style of calligraphic characters can be extracted with a uniform algorithm. Besides, we really constructed a robot for writing generated characters and the problems introduced by adopting a brush and their solutions are also discussed. In particular, because of the softness and friction against paper of the brush head, the postponement of the brush touch may be caused and we solved it by developing a special compensation method. In addition, an experimental result based on several style of characters is described. It turns out that, since the writing process is much similar to the one made by man, the aesthetic perception and momentum as well as some intrinsic properties of calligraphy can be better demonstrated.