We investigated a tunable delay-line using an optical single-sideband modulator and an optical fiber loop. The single-sideband modulator consists of four optical modulators and an RF electric signal source. The fiber loop has a fiber Bragg grating and a couple of optical circulators. The number of times light circulates in the loop depends on the frequency of the rf-signal fed to the modulators. By using numerical simulations, we discussed the deformation of the waveform in the delay-line due to the fiber Bragg gratings, the modulators and the optical amplifiers put in the loop.

Probabilistic inference by means of a massive probabilistic model usually has exponential-order computational complexity. For such massive probabilistic model, loopy belief propagation was proposed as a scheme to obtain the approximate inference. It is known that the generalized loopy belief propagation is constructed by using a cluster variation method. However, it is difficult to calculate the correlation in every pair of nodes which are not connected directly to each other by means of the generalized loopy belief propagation. In the present paper, we propose a general scheme for calculating an approximate correlation in every pair of nodes in a probabilistic model for probabilistic inference. The general scheme is formulated by combining a cluster variation method with a linear response theory.

In this paper, we propose a new configuration to implement transversal filters with negligible temperature sensitivity and low cost. These microwave filters are based on uniform fibre Bragg gratings as slicing elements of a broadband optical source. By using a tapered fibre Bragg grating as a delay line, we show that the temperature effects are the same over each component of the RF-filter. Therefore, it is possible a total cancellation of the thermal effects. The performance of these filters is compared to previous techniques, such as a laser array approach.

In the image recognition problem, it is very important how we represent the image. Considering this, we propose a new representational method of images based on the stability in scale-space. In our method, the image is segmented and represented as a hierarchical region graph in scale-space. The object is represented as feature graph, which is subgraph of region graph. In detail, the region graph is defined on the image with the relation of each segment hierarchically. And the feature graph is determined based on the "life-time" of the graph of the object in scale-space. This "life-time" means how long feature graph lives when the scale parameter is increased. We apply our method to the face detection problem, which is foundmental and difficult problem in face recognition. We determine the feature graph of the frontal human face statistical point of view. We also build the face detection system using this feature graph to show how our method works efficiently.

An adaptive blind signal separation filter is proposed using a risk-sensitive criterion framework. This criterion adopts an exponential type function. Hence, the proposed criterion varies the consideration weight of an adaptation quantity depending on errors in the estimates: the adaptation is accelerated when the estimation error is large, and unnecessary acceleration of the adaptation does not occur close to convergence. In addition, since the algorithm derivation process relates to an H filtering, the derived algorithm has robustness to perturbations or estimation errors. Hence, this method converges faster than conventional least squares methods. Such effectiveness of the new algorithm is demonstrated by simulation.

We investigate the capacity limitations of a WDM ring fiber-radio backbone incorporating wavelength interleaving where each base station drives a sectorized antenna interface. We also investigate the issues related to the merging of such networks with standard WDM infrastructures. The investigations show that re-allocating the interleaved WDM channels to fit within a 100 GHz block enables the millimeter-wave (mm-wave) fiber-radio system with sectorized antenna interfaces to integrate easily with WDM systems. The performance of a variety of channel allocations for the merged fiber-radio network is examined and simulation studies of the transmission of multiple channels are carried out. The overall network capacity of the merged mm-wave fiber-radio network is improved with the proposed channel allocation schemes.

A low power MPEG-4 video codec LSI with the capability for core profile decoding is presented. A 16-b DSP with a vector pipeline architecture and a 32-b arithmetic unit, eight dedicated hardware engines to accelerate MPEG-4 SP@L1 codec, CP@L1 decoding and post video processing, 20-Mb embedded DRAM, and three peripheral blocks are integrated together on a single chip. MPEG-4 SP@L1 codec, CP@L1 decoding and post video processing are realized with a hybrid architecture consisting of a programmable DSP and dedicated hardware engines at low operating frequency. In order to reduce the power consumption, clock gating technique is fully adopted in each hardware block and embedded DRAM is employed. The chip is implemented using 0.18-µm quad-metal CMOS technology, and its die area is 8.8 mm 8.6 mm. The power consumption is 90 mW at a SP@L1 codec and 110 mW at a CP@L1 decoding.

In our previous research, we proposed a nonlinear digital filter to Estimate the Smoothed and Differential values of the sensor inputs by using Sliding mode system (ESDS). This estimator is able to eliminate impulsive noise efficiently from time series data. We applied this filter to processing outputs of robot sensors, and it became possible to perform robust environment recognition. ESDS is designed using a theory of variable structure system (VSS) with sliding mode. In short, ESDS is a nonlinear filter. Therefore, it is very difficult to clarify the behavior of the system analytically. Consequentially, we deal with the step function with impulsive noise as an example, and we attempt to eliminate this impulsive noise by keeping the sudden shift of signals. In this case, there is a trade-off between the noise elimination ability and the tracking performance for an input signal. Although ESDS is a nonlinear filter, it has the same trade-off as linear filters such as a low-pass filter. In order to satisfy these two conditions simultaneously, we use two filters whose parameters are independent of each other. Furthermore, in order to repress the adverse affect of impulsive noise in the steady-state, we introduced the boundary layer. Generally, a boundary layer is used so as to inhibit the harmful effect of chattering. Chattering is caused in the sliding mode system when the state of the system vibrates on the switching line of a sliding mode system. By introducing the boundary layer to ESDS, we can repress the adverse effect of impulsive noise in the steady-state. According to these considerations, we clarify the relationship between these characteristics of ESDS and the arbitrary parameters.

In the paper, we propose a new method for chromatic dispersion measurement of WDM components in both transmission and reflection, employing photonic microwave technology. The dispersion can be determined by measuring the frequency spectrum range change of the microwave notch filter. The method features the advantages of low-cost and simplicity. Experimental results demonstrate that our setup is capable of measuring relative group delay with better than 1 ps time resolution and the measurement results show a good agreement with that measured by the conventional phase-shift technique.

In this paper, we propose a simple method to find the optimal rational function, with a fixed denominator, which minimizes an integral of polynomially weighted squared error to given analytic function. Firstly, we present a generalization of the Walsh's theorem. By using the knowledge on the zeros of the fixed denominator, this theorem characterizes the optimal rational function with a system of linear equations on the coefficients of its numerator polynomial. Moreover when the analytic function is specially given as a polynomial, we show that the optimal numerator can be derived without using any numerical integration or any root finding technique. Numerical examples demonstrate the practical applicability of the proposed method.

A decentralized estimation system usually contains a number of remotely located local sensors that can pre-process observed signal and convey the processed data to a fusion center that makes a final estimation. The local sensors are linked to the data fusion center by transmission channels. When the observation (or estimate of parameter) is quantized at the peripheral sensors and an assumption of conditionally independent sensor data is made, due to potential communication constraints on the channels, the problem of quantization design and bandwidth allocation among the channels linking local sensors to the fusion center is studied in this letter.

For common-mode noise current reduction, a CMC (Common Mode Choke) is widely used in signal transmission line circuits consisting of a ground and two conductors (a balanced transmission line). However, a common-mode noise current reduction characteristic is not clearly analyzed yet in the case that a CMC is inserted in a balanced transmission line. In this paper we propose the calculation method of CMC insertion losses and derive an equation to analyze the common-mode current characteristics of a balanced transmission line with a CMC inserted. The analyzed frequency range is from 100 kHz to 100 MHz. We conclude that in the frequency range up to 30 MHz: (1) the proposed insertion loss calculation method is useful for analyzing CMC insertion losses in differential-mode and in common-mode; (2) the derived circuit equation can be applied for analyzing the common-mode current characteristics of a balanced transmission line locally unbalanced with conditions of a CMC inserted; (3) the proposed calculation method may give the expected results that a CMC should be placed in a signal source side of an unbalanced point of a pair-cable for reduction of common-mode currents; and (4) if it is placed in a terminal (or load) side of an unbalanced point, there is no effect, or rather common-mode currents are increased by the insertion of a CMC.

We present a new method called the slope nucleation method (SNM) for the growth of high-quality organic 4-dimethylamino-N-methyl-4-stilbazolium tosylate (DAST) crystals. The SNM features the ability to control the nucleation position and the growth orientation of DAST crystals in spontaneous nucleated growth. X-ray diffraction (XRD) rocking curve measurements indicate that the SNM is effective for obtaining high-quality DAST crystals as compared to conventional spontaneous nucleation methods. We evaluated the electro-optic (EO) properties of DAST crystals by an external EO probing technique because DAST crystals are expected to be used in transverse-field probing. DAST crystals exhibits nearly five-times EO sensitivity enhancement as compared to inorganic KTiOPO4 (KTP) crystals at 90 kHz. The larger EO signal power obtained from the DAST crystal was almost constant at low frequencies (30 Hz to 90 kHz), whereas the KTP crystal could not respond below 180 Hz. We also observed excellent signals at all measured points due to the improved crystallinity of the crystal grown by the SNM.

A new CMOS 1/x circuit is presented in this letter. The output amplitude of the proposed circuit can be adjusted by a bias current. The proposed circuit can be used to realize a current-to-voltage converter and a current-mode divider. The proposed circuits have been fabricated in a 0.5 µm CMOS process. Experimental results show that under the linear error less than 1%, the input range of the proposed 1/x circuit can be up to 1.5 V for the supply voltages of 1.5 V and the power dissipation is 0.24 mW. The experimental results are given to demonstrate the proposed circuits.

The conservative mode and the greedy mode scheduling algorithms for OBS switch with shared buffer are presented and discussed. Their performance is evaluated by computer simulations, as well as that of the greedy mode with void-filling algorithm. Simulation results show that the conservative mode and the greedy mode have different characteristics under different input load. The greedy mode and the conservative mode are more applicable in a real system than that with void-filling, owing to their lower computational complexity and FIFO characteristic. Finally, a composite algorithm integrated by the conservative mode and the greedy mode is proposed, which is adapted to the input load with the help of an input load monitor. The simulation results reveal that it has favorable performance under different load.

Wall shadow removal problems differ due to configuration of walls in different propagation scenarios. Applying Weiler-Atherton polygon clipping method helps calculate illuminated regions on the building walls, but unfortunately the technique has limitation when there are many walls and the walls configuration is complex. The modified Weiler-Atherton polygon clipping method proposed can solve the problem by regarding all vertices of the subject polygon or clipping polygon, that are also intersection points as simply intersection points. In the case that a certain vertex of the subject polygon is a vertex of the clipping polygon, this vertex of the subject polygon is still considered a vertex. It is found that wall shadow removal using the proposed modified Weiler-Atherton polygon clipping method is more efficient.

A "rooted" plane triangulation is a plane triangulation with one designated vertex r and one designated edge incident to r on the outer face. In this paper we give a simple algorithm to generate all connected rooted plane triangulations with at most m edges. The algorithm uses O(m) space and generates such triangulations in O(1) time per triangulation without duplications. The algorithm does not output entire triangulations but the difference from the previous triangulation. By modifying the algorithm we can generate all connected (non-rooted) plane triangulations with at most m edges in O(m3) time per triangulation.

This paper proposes a real-time scheduling mechanism for web server that finds a way to solve so-called priority inversion problem between tasks handling real-time web requests with higher priorities and tasks with lower priorities like HTML requests. The priority inversion problem of web server stems from operating two independent schedulers, web scheduler and kernel scheduler, without exchanging scheduling information or considering the requirement of each other. In the proposed mechanism, two schedulers are integrated in a kernel and the integrated scheduler schedules real-time web tasks and other application tasks together based on their priorities. Since the unified scheduler sees all tasks simultaneously in system, it can schedule the tasks with their absolute priorities and reduce the execution delay caused by the priority inversion. The proposed mechanism provides web server with a better chance to efficiently serve real-time web requests. The feasibility and performance of the proposed mechanism are empirically shown on an embedded system.

A clustering scheme for ad hoc networks is aimed at managing a number of mobile devices by utilizing hierarchical structure of the networks. In order to construct and maintain an effective hierarchical structure in ad hoc networks where mobile devices may move at high mobility, the following requirements must be satisfied. (1) The role of each mobile device for the hierarchical structure is adaptive to dynamic change of the topology of the ad hoc networks. The role of each mobile device should thus change autonomously based on local information in each mobile device. (2) The overhead for management of the hierarchical structure is small. The number of mobile devices in each cluster should thus be almost equivalent. This paper proposes an adaptive multihop clustering scheme for highly mobile ad hoc networks. The results obtained by extensive simulation experiments show that the proposed scheme does not depend on mobility and node degree of mobile devices in the network, which satisfy the above requirements.

In this paper, we present a new analog multiplier with wide input range which is achieved by utilizing the variable threshold voltage characteristics of FG-MOSFET's. The performance of the proposed multiplier is evaluated through HSPICE simulation with 2.0 µm CMOS process parameters. From HSPICE simulation, we can find that the proposed multiplier can be operated at the supply voltage of 3.0 V with 3.0 Vp-p input range. Namely, the input voltage range of the multiplier is equal to the supply voltage. The maximum power consumption of the multiplier is 8.8 mW. The THD is 1.36% under the condition that the amplitude of the input signal is 3.0 Vp-p and the frequency is 1 MHz. Under the same condition, the linearity error is less than 0.5%. The -3 dB bandwidth of the proposed multiplier is 23 MHz.