A function can be represented in many ways. A representation O of a function f is called 'obscure' if O is different from the representation D used as the definition of f and if it is (or, seems to be) computationally infeasible to get D from O. Such an obscure representation is useful for cryptographic techniques so that it is important to estimate its descriptive and executive complexity. We present a complexity-estimation method for certain functions used to constructing asymmetric cryptosystems. Let m be a positive integer and let K, Km, and L denote the field {0, 1}, the set of all m-tuples over K, and an extention field or order m over K, respectively. The objective function is a composit g:Km Km of three functions s, e, and t, where s:Km L and t:L Km are affine and e:L L is defined by a univariate polynomial e over L. The obscure representation of g is an m-tuple g of m-variate polynomials over K. The complexity respect to g is well measured by its degree. So we give a theorem for estimating the degree of g in terms of a characteristic quantity of the polynomial e.

This paper presents a rigorous Wiener-Hopf solution to the electromagnetic wave exitation by a waveguide mode, concerning a two-dimensional tunnel which might be the most simplified model of a rectangular tunnel. Surface impedance boundaries are assumed on the walls of tunnels surrounded by lossy dielectric materials, such as concrete, rock and others. A microwave simulation is also performed to determine whether the assumption used in the theory is good or not. From comparison of experimental results with theoretical ones, it is shown that the surface impedance model is an excellent approximation especially for tunnels of which dimensions are not so large compared with the wave length in the free space.

Three-wave interactions in a linear, active, and lossless system are investigated. The stability properties of the coupled-waves describing the interaction are discussed at length by taking into consideration possible combinations of parities and directions of group velocities of the component waves. Several schematic dispersion diagrams are presented for the classification of existing instabilities in this system. The power gains of traveling-wave amplification in the three-wave interaction are also analyzed. It is shown that when one passive-wave couples to other two active-waves propagating in the same direction, a power gain considerably larger than that of corresponding two-wave interaction can be obtained.

Intensity noise reduction effects by superimposed pulse modulation are theoretically analyzed for LD inherent noise, feedback light noise, and modal noise. It is shown that noise spectrum spreading effects due to LD mode partition noise are very effective to reduce mode hopping noise and modal noise, that a short pulse width completely suppress beat noise and mode hopping noise based on submode structures constructed by feedback light, and that modal noise is effectively reduced by LD longitudinal mode linewidth broadening effect as long as a fiber bandwidth is sufficiently narrow compared with the broadened LD linewidth. It is also shown that LD quantum intensity noise increases for LD multimode operation, and that feedback light noise based on injection locking effect can not be suppressed without system parameter adjustment.

A 1.544 MHz timing-extraction ceramic filter is examined. This filter has been successfully introduced into a PCM carrier system. Second-order Gaussian approximated effective-parameter theory with gentle frequency-phase inclination characteristics is applied in the circuit design. A ladder transformer filter construction, employing two resonator units and a coupling capacitor, is adopted. A resonator unit, comprising a trapped-energy piezoelectric ceramic strip resonator operating in the width-shear vibration mode and a ceramic capacitor connected in series with the resonator, fulfils the function for easy frequency adjustment and temperature compensation of the resonant frequency. Filter input and output terminating resistances are 30 Ω and 1500 Ω, respectively. The 3 dB fractional bandwidth is 1.07% and the insertion loss at center frequency f0 is -4.8 dB. The frequency deviation of f0 is within 200 ppm over a -20 to 60 temperature range and the aging rate of f0 is about 0.01%/decade. The frequency deviation over the operating temperature range is less than one-third and the aging rate is about one-half of the conventional LC filter used for this application. The filter size is reduced to one-fifth that of the LC filter.

The thickness dependence of structure, magnetic properties and read-write characteristics has been studied for sputtered iron oxide thin films prepared by RF reactive sputtering. For sputter-deposited Fe3O4 films, [111] preferential orientation perpendicular to film surface is improved and crystal size increases with an increase in thickness (δ). Magnetic properties for Fe3O4 films are affected by thickness. In particular, saturation magnetization rapidly decreases at thicknesses less than about 0.07 µm. From this perspective, it is concluded that a thickness larger than about 0.07 µm is available for practical use. For oxidation-annealed γ- Fe2O3 media, signal-output and media transition length are proportional to δ0.7 and δ0.5, respectively. A γ-(Co0.025Cu0.03Fe0.945)2O3 medium with a 0.071 µm thickness shows 0.51 mV of Epp, 2000 frpm of D-6 dB, -35 dB of O/W and 38 dB of SNR.

This paper describes on optical transmission system for a Time Compression Multiplexing (TCM) High Definition TV (HDTV) signal with a bandwidth of 20 MHz. Negative feedback is being employed to improve some of the shortcomings of the laser diode. In this paper the stability of the negative feedback circuit and its effect on reducing harmonic distortion and noise in the laser diode are examined. A laser diode module used in this experiment consists of a TJS-LD operating at a wavelength of 0.89 µm and a photo diode. In the final system, a method that superimposes high frequency pulses is employed for reducing modal noise. The signal-to-noise ratio (SNR) of the optical sender is achieved at 49 dB and harmonic distortion at about 44 dB, where the optical output level is 6.0 dBm and modulation depth is 0.5.

The quadratic Liénard equation, that is, the Liénard differential equation in which the coefficients of and x are polynomials of degree two in x is studied with respect to its periodic solutions of small amplitude. The equation can represent a hard or a soft oscillation depending on parameter values. The existence of periodic solutions is proved and a simple formula for amplitude is given.

Low temperature excess loss of fusion-spliced fibers packaged by a new packaging method is investigated. The package consists of a heat-shrinkable tube, an adhesive tube and a resistance heating rod. Based on the theoretical analysis using the theory of elasticity, the packaging structure using a rigid resistance rod, which has a high compression modulus, low linear expansion coefficient and large cross-sectional area, is recommended for reducing the loss increase at low temperatures. By using carbon fiber-reinforced carbon (CFRC) for such a resistance rod, the loss increase at -40 is found to be less than 0.03 dB per package.

An improved theory to obtain circularly polarized (CP) waves from a singly fed microstrip antenna is presented. This theory indicates that a singly fed circularly polarized (SFCP) microstrip antenna, in general, is able to produce CP waves at two different frequencies, which are just slightly separated. The two CP operating frequencies can be found through an iterative process. Furthermore, the optimum feed location loci can be determined numerically. Finally, the theoretical results for two kinds of SFCP antennas are compared with the experimental results. The agreement is quite good and the theory validity is confirmed.

Phenomenon of silica-fiber loss increase, which is induced by an electric current applied in water, is reported. It is found from experimental results that the loss increase is caused by hydrogen molecules, which are electrolytically produced and are diffused into the fiber interstitially. Furthermore, both infrared loss and Ramanscattering spectra are measured for a fiber dissolved by H2, HD, and D2 molecules to make clear assignment for loss peak. As a result, it is found that the absorption loss peaks are associated with a connection between molecular vibration and SiO4 tetrahedral vibrations in the silica glass.

A recognition method of handprinted Thai characters by using local features is described. In the method, Freeman chain code and directional differences of contour tracing are utilized for extracting concavities and convexities of characters. Several local features are used to calculate similarities between arc portions and similarity between characters. Similar arcs are detected from characters of different categories to make a dictionary of arcs. Then, a dictionary of characters containing lists of names of character arcs is made to obtain a compact dictionary of models. By applying the method to 69 categories (828 data) of Thai characters, a recognition rate of 99.3% for learning data, and a recognition rate of 88.9% for test data have been obtained.

This paper presents a surface emitting injection laser with improved short cavity structure operating at 1.30 µm (77 K) of wavelength. The selective etching technique has been applied to take out the InP substrate in order to reduce the absorption loss and resultant cavity length was 7 µm. The threshold was as low as 50 mA at 77 K. The device operated up to 140 K with one single longitudinal mode and small wavelength change (0.64 /degree). The near-field diameter was 9 µm and the far-field angle (FWHM) was 9 degrees. The polarization was linear.

In the near future, the packet switching system will need a large switching capacity. But, large processor can not be obtained in cheap way. So, it seems better to use smaller processors in a distributed fashion to obtain a large switching capacity. If several small switching devices are interconnected, it will be possible to increase the traffic handling capacity. Depending on the topology of the network, overall processing characteristics are different. Switching networks like torus etc. have the problem, because their nodes have the transit traffic. According to the increase of the switching network size, the transit traffic becomes large and the overall traffic efficiency of the network decreases significantly. The mesh network, in which all nodes are interconnected, does not have that problem, because there is no transit traffic in any node. This paper describes the composition of mesh type distributed packet switching systems. Flexible architectures are proposed, which are based on the building block structure. A node of these architectures is composed of several switching elements, and one or two intranode switches. Using these architectures, the addition of a new node to the system can be done without any modification of the fundamental units. The experimental system has been designed and built.

A new type of passive TE/TM mode splitter device using the ion-exchanged LiNbO3 waveguide is proposed and demonstrated. The characteristics of this device is estimated by the Beam Propagation Method based on a simple model. The extinction ratio is predicted to amount to 40 dB in the optimal case. This device is experimentally demonstrated in the X-junction configuration by using the Ag+-exchanged waveguide an Ti-indiffused one, and the extinction ratio is measured to be 10.5 dB, while the calculated value is 15 dB.

We measured response times of visual pattern recognition for various figures and characters on a TV screen when visual field of subjects were restricted to various visual angles, using a television eye camera and a newly developed equipment for restricting human visual field effectively, and estimated size of central vision according to the result of measurement.

Various methods have been proposed for the generation of constant envelope signals. It is the purpose of this paper to show systematic methods to generate a class of constant envelope signals with linear or nonlinear phase. Unlike other methods, our technique does not require nonlinear elements and the distortion is controlled by the parameters of the signal.

Lasing gain and threshold current density in GaAs/AlGaAs multi-quantum-well lasers are theoretically analyzed. Split energy levels in multi-quantum-well structure are calculated. The lasing gain and electron lifetime by the spontaneous emission effect are calculated taken into account both the split energy levels and broadening effect by the intraband relaxation of electron wave. It is found that the gain can not be much higher than that in conventional double heterostructure, because the gain is broadened by the intraband relaxation effect even in the quantum-well structure. Conditions of laser structure to get much lower threshold current density are discussed. The most important condition to reduce the threshold current density is to increase confinement factor of the optical field into the well layers.

A new GaAs monolithic negative feedback amplifier using a dual-gate MESFET for A.G.C. function has been developed. Such a low NF value as 1.8 to 2.5 dB with a gain of 10 to 12 dB over a frequency range of 50 to 2,000 MHz has been obtained. The gain reduction of about 30 dB was achieved over the same frequency range by means of the second-gate bias. A comparison of the performance between negative feedback amplifiers using a dual-gate FET and a single-gate FET is also discussed.

A new coding scheme of binary images is proposed. This coding scheme not only obtains high compression efficiencies, but also produces coded data which are suitable for computer processing.