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.

Principle of a single shot demultiplextion by means of time-to-space conversion was investigated using femtosecond nonlinear optical response of absorption bleaching of squarylium dye (SQ) J-aggregates. Spincoated films of squarylium dye J-aggregates on glass substrates exhibit efficient and ultrafast transmittance change, which recovers 73% of its initial level (0 fs) within 1 ps. A simple method for time-to-space conversion was applied for this film. We took our attention to one of the characteristics of femtosecond pulse, which is the spatial thinness in its propagation direction. Femtosecond pulses of a single pump pulse and train of four probe pulses were illuminated to the same area (diameter of 10 mm) of the surface of the SQ J-aggregates film. Direction of the probe beam was normal to the surface of the film and that of the pump beam was oblique angle in horizontal plane. Caused by spatial delay of a pump pulse due to the illumination in oblique angle to the film, four probe pulses with interval time of 1 ps (1 THz) meet separate places on the film. Because of the fast response of the SQ J-aggregates, the film picked out part of each probe pulse, which has narrower shapes in horizontal direction compared to the initial circular one by transmittance change of the film. The spatially separated four lines were observed by a CCD camera for an image of the transmitted probe pulse train. These results suggest that the response time of SQ J-aggregate film, which determines the horizontal width of each line, to be enough for demultiplexing of 1 THz optical signals.

In this paper, we propose the majority algorithm to choose the connection weights for the neural networks with quantized connection weights of 1 and 0. We also obtained the layered network to solve the parity problem with the input of arbitrary number N through an application of this algorithm. The network can be expected to have the same ability of generalization as the network trained with learning rules. This is because it is possible to decide the connection weights, regardless of the size of the training set. One can decide connection weights without learning according to our case study. Thus, we expect that the proposed algorithm may be applied for a real-time processing.

We present two estimation methods for camera rotation from two images obtained by the active camera before and after rotation. Based on the representation of the projected rotation group, quasi moment features are constructed. Camera rotation can be estimated by applying the singular value decomposition (SVD) or Newton's method to tensor quasi moment features. In both cases, we can estimate 3D rotation of the active camera from only two projected images. We also give some experiments for the estimation of the actual active camera rotation to show the effectiveness of these methods.

We describe the fabrication method and characteristics of hybrid external cavity lasers composed of a spot-size converter integrated LD (SS-LD) and a UV written Bragg grating in a planar lightwave circuit (PLC) on a Si substrate. The SS-LD is passively aligned on a Si platform formed in the PLC, and the UV grating is created in the PLC with ArF laser irradiation through a phase mask. This structure enables us to fabricate a stable single-mode laser with a precisely controllable oscillation wavelength. By using the above techniques, we obtained a threshold current of 7-8 mA and a side mode suppression of 37 dB for an external cavity laser operating at 1.3 µm. Moreover, we successfully demonstrated a four-channel external cavity laser with a wavelength interval of 2 nm 0.1 nm by integrating 4 SS-LDs on a PLC and controlling the Bragg wavelengths with ArF laser irradiation without a phase mask.

We studied theoretically and experimentally an InGaAs/InAlAs/InP polarization-insensitive multiple quantum well (MQW) electroabsorption (EA) modulator operating over a very wide wavelength range in 1.55 µm wavelength region. One of the simplest possible potential-tailored quantum well, "pre-biased" quantum well (PBQW) is used to achieve wide-wavelength polarization insensitivity. PBQW is basically a rectangular quantum well with a thin barrier inserted near one edge of well. This thin barrier effectively introduces "pre-bias" to a rectangular quantum well and the same amount of Stark shift is achieved for electron-heavy hole and electron-light hole transition energies. By incorporating tensile strain into PBQW, polarization-insensitive modulation is achieved over 60 nm wavelength range, from 1510 nm to 1570 nm. This MQW-EA modulator plays an important role in wavelength division multiplexing (WDM) transmission and switching systems.

10-Gbit/s monolithic receiver OEIC's for 1.55-µm optical transmission systems were fabricated using a stacked layer structure of p-i-n photodiodes and HEMT's grown on InP substrates by single-step MOVPE. A receiver OEIC with a large O/E conversion factor was obtained by adding a three-stage differential amplifier to a conventional feedback amplifier monolithically integrated with a surface-illuminated p-i-n photodiode. The circuit configuration gave a preamplifier a transimpedance of 60 dBΩ. The receiver OEIC achieved error-free operation at 10 Gbit/s without a postamplifier even with the optical input as low as -10.3 dBm because of its large O/E conversion factor of 890 V/W. A two-channel receiver OEIC array for use in a 10-Gbit/s parallel photoreceiver module based on a PLC platform was made by monolithically integrating multimode WGPD's with HEMT preamplifiers. The side-illuminated structure of the WGPD is suitable for integration with other waveguide-type optical devices. The receiver OEIC arrays were fabricated on a 2-inch wafer with achieving excellent uniformity and a yield over 90%: average transimpedance and average 3-dB-down bandwidth were 43.8 dBΩ and 8.0 GHz. The two channels in the receiver OEIC array also showed sensitivities of -16.1 dBm and -15.3 dBm at 10 Gbit/s. The two-channel photoreceiver module was constructed by assembling the OEIC array on a PLC platform. The frequency response of the module was almost the same as that of the OEIC chip and the crosstalk between channels in the module was better than -27 dB in the frequency range below 6 GHz. These results demonstrate the feasibility of using our receiver OEIC's in various types of optical receiver systems.

A theoretical study on high slope-efficiency phase-shifted DFB laser diodes is presented. We have proposed a new grating structure called asymmetrically-pitch-modulated (APM) grating, and calculated its slope- efficiency and single-mode-yield. In order to take into account the modulated grating period; we have developed an F-matrix which directly includes a chirped grating structure. APM phase-shifted DFB laser diodes consist of a uniform grating in one half section of the cavity and a chirped grating in the other half. This structure causes asymmetrical field distribution inside the cavity and the optical output power from one facet is larger than that from the other facet. According to the simulation results, when the normalized coupling coefficient κ L is 3.0, the front-to-rear output power ratio is 2.6, while the single-mode-yield remains at 100%, and simultaneously the slope-efficiency improvement becomes 65% better than that of ordinary quarter-wave phase-shifted DFB lasers of the same κ L value.

A 160 GHz colliding-pulse mode-locked laser diode (CPM-LD) was stabilized by injection of a stable master laser pulse train repeated at a 16th-subharmonic-frequency (9.873 GHz) of the CPM-LD's mode-locking frequency. Synchroscan steak camera measurements revealed a clear pulse train with 16-times repetition frequency of the master laser pulse train for the stabilized CPM-LD output, indicating that CPM-LD output was synchronized to the master laser and that the timing jitter was also reduced. The timing jitter of the stabilized CPM-LD was quantitatively evaluated by an all-optical down converting technique using the nonlinearity of optical fiber. This technique is simple and has a wider bandwidth in comparison to a conventional technique, making it possible to accurately measure the phase noise of ultrafast optical pulse train when its repetition frequency exceeds 100 GHz. The electrical power spectra measurements indicated that the CPM-LD's mode-locking frequency was exactly locked by the injection of the master laser pulse train and that the timing jitter decreased as the injection power increased. The timing jitter was reduced from 2.2 ps in free running operation to 0.26 ps at an injection power of 57 mW, comparable to that of the master laser (0.21 ps).

The propagation and the gate operation of femtosecond pulses in nonlinear optical waveguides utilizing the saturation of the intersubband absorption at 1.55 µm in nitride multiple quantum wells are simulated for the first time. The calculation was carried out by a one-dimensional finite-difference time-domain (FD-TD) method combined with three-level rate equations describing the intersubband carrier dynamics. The absorption recovers within 1 ps when the pulse width is less than 200 fs, which will allow 1-Tb/s operation. However, the pulse shape may be deformed with the propagation due to the coherent effect and the interference between the signal and the control pulses, and thus, optimization of the pulse widths and the incident timing is required. Since the transparent window (width of the control pulse) becomes shorter according to the propagation, the width of the control pulse should be set broader than that of the signal pulse. As an example, we assume the case where a 1.6-µm, 100-fs signal pulse is gated by a 300-fs control pulse at a wavelength of 1.5 µm in a 500-µm length waveguide. A 140-fs gated signal pulse with a smooth envelope is expected to appear after the band-pass filter. The extinction ratio is expected to be greater than 15 dB.

A newly developed hybrid-integrated Symmetric Mach-Zehnder (HI-SMZ) all-optical switch is reported. For integration, we chose the Symmetric Mach-Zehnder (SMZ) structure rather than the Polarization-Discriminating Symmetric Mach-Zehnder (PD-SMZ) structure which is similar to SMZ but more often used in experiments using discrete optical components. We discuss advantages and disadvantages of SMZ and PD-SMZ to show that SMZ is more suitable for integration. We also discuss about the use of SOAs as nonlinear elements for all-optical switches. We conclude that, although the ultrafast switching capability of SMZ is limited by the gain compression of SOAs, the very low switching energy is more important for practical devices. We then describe the HI-SMZ all-optical switch. This integration scheme has advantages which include low loss, low dispersion silica waveguides for high speed operation and ease in large scale integration of many SMZs with other optical, electrical, and opto-electrical devices. We show that a very high dynamic extinction ratio is possible with HI-SMZ. We also examine HI-SMZ with 1 ps pulses to show its ultrafast capability. Finally, we describe a 168 to 10.5 Gbps error-free demultiplexing experiment which is to our best knowledge the fastest experiment with an integrated device.

A new digital watermark approach based on fractal image coding is proposed in this letter. We present a way to use the fractal code as a means of embedding a watermark. The proposed approach has shown to be resistant to the JPEG lossy compression. Moreover, the digital watermark can be simply extracted from the watermarked image without resorting to the original image.

We propose a novel Bragg grating filter synthesis method using a Fourier transform of the target scattering matrix. Multiple scattering processes are taken into account by iteration to improve the synthesis accuracy.

This paper presents a constant-scale-factor radix-2-4-8 CORDIC algorithm for fast vector rotation and sine/cosine computation. The CORDIC algorithm is a well-known hardware algorithm for computing various elementary functions. Due to its sequential nature of computation, however, significant reduction in processing latency is required for real-time signal processing applications. The proposed radix-2-4-8 CORDIC algorithm dynamically changes the radix of computation during operation, and makes possible the reduction in the number of iterations by 37% for 64-bit precision. This paper also describes the hardware implementation of radix-2-4-8 CORDIC unit that can be installed into practical digital signal processors.

This paper introduces the construction of a family of complex-valued scaling functions and wavelets with symmetry/antisymmetry, compact support and orthogonality from the Daubechies polynomial, and applies them to solve electromagnetic scattering problems. For simplicity, only two extreme cases in the family, maximum-localized complex-valued wavelets and minimum-localized complex-valued wavelets are investigated. Regularity of root location of the Daubechies polynomial in spectral factorization are also presented to construct these two extreme genus of complex-valued wavelets. When wavelets are used as basis functions to solve electromagnetic scattering problems by the method of moment (MoM), they often lead to sparse matrix equations. We will compare the sparsity of MoM matrices by the real-valued Daubechies wavelets, minimum-localized complex-valued Daubechies and maximum-localized complex-valued Daubechies wavelets. Our research summarized in this paper shows that the wavelets with smaller signal width will result in a more sparse MoM matrix, especially when the scatterer is with many corners.

Software radio base and personal station prototypes are proposed and implemented. The prototypes are composed of RF/IF, A/D and D/A, pre- and post-processors, CPU, and DSP parts. System software is partitioned into CPU program and DSP program to use processor resources effectively. They support various air interfaces, some of which are equivalent to the 384 kbit/s transmission rate PHS (personal handy phone system) and a 96 kbit/s transmission rate system. The base station can also be used as a communication bridge between two systems. In order to ease IF filter requirements, the zero-stuff method is employed. Basic transmission and receiving performances are evaluated in an experiment and their results agree well with those expected.

A generalized class of consecutive-k-out-of-n:G systems, referred to as Con/k*/n:G systems, is studied. A Con/k*/n:G system has n ordered components and is good if and only if ki good consecutive components that originate at component i are all good, where ki is a function of i. Theorem 1 gives an O(n) time equation to compute the reliability of a linear system and Theorem 2 gives an O(n2) time equation for a circular system. A distributed computing system with a linear (ring) topology is an example of such system. This application is very important, since for other classes of topologies, such as general graphs, planar graphs, series-parallel graphs, tree graphs, and star graphs, this problem has been proven to be NP-hard.

We present three-wave mixing devices useful for signal processing functions in WDM and TDM systems, including wavelength conversion, spectral inversion, and gated mixing. These mixers exhibit extremely wide bandwidth, low noise, high efficiency, and format transparency.

The multiple access causes an interference problem in the direct-sequence code-division multiple access systems. An efficient adaptive algorithm should be used to suppress this interference for the improvement of system performance. In this paper, the new blind adaptive method is suggested using the constant modulus algorithm for the purpose of interference suppression. Simulation results show that the converged value of signal to interference ratio for the proposed method is approximately 6 [dB] larger than that of a conventional Blind-MOE receiver in the additive white Gaussian noise channel and channel with inter-symbol interference while the signal to interference ratio improvement is almost 4 [dB] better in the Rayleigh fading channel. The suggested method is also robust to the new user interference resulting the nearly 3 [dB] improvement of the SIR value comparing with the conventional receiver. Based on these results, it is shown that the BER of the proposed receiver is lower than that of any other conventional receiver. Therefore, using the newly suggested method, the considerable performance improvement can be obtained for the DS-CDMA systems.

We propose a rough classification system using Hierarchical Learning Vector Quantization (HLVQ) for large scale classification problems which involve many categories. HLVQ of proposed system divides categories hierarchically in the feature space, makes a tree and multiplies the nodes down the hierarchy. The feature space is divided by a few codebook vectors in each layer. The adjacent feature spaces overlap at the borders. HLVQ classification is both speedy and accurate due to the hierarchical architecture and the overlapping technique. In a classification experiment using ETL9B, the largest database of handwritten characters in Japan, (it contains a total of 607,200 samples from 3036 categories) the speed and accuracy of classification by HLVQ was found to be higher than that by Self-Organizing feature Map (SOM) and Learning Vector Quantization methods. We demonstrate that the classification rate of the proposed system which uses multi-codebook vectors for each category under HLVQ can achieve higher speed and accuracy than that of systems which use average vectors.