The optical recording on an azopolymer surface by the optical fiber probe with a 100 nm diameter aperture was demonstrated. The 150 nm diameter pit was formed by the optical fiber probe coupled with a 50 ns pulse of 10 mW and 488 nm wavelength coherent light.

A radio network is a distributed system with no central shared resource, consisting of n stations each equipped with a radio transceiver. One of the most important parameters to evaluate protocols in the radio networks is the number of awake time slots in which each individual station sends/receives a data packet. We are interested in devising energy-efficient initialization protocols in the single-hop radio network (RN, for short) that assign unique IDs in the range [1,n] to the n stations using few awake time slots. It is known that the RN can be initialized in O(log log n) awake time slots, with high probability, if every station knows the number n of stations in the RN. Also, it has been shown that the RN can be initialized in O(log n) awake time slots even if no station knows n. However, it has been open whether the initialization can be performed in O(log log n) awake time slots when no station knows n. Our main contribution is to provide the breakthrough: we show that even if no station knows n, the RN can be initialized by our protocol that terminates, with high probability, in O(n) time slots with no station being awake for more than O(log log n) time slots. We then go on to design an initialization protocol for the k-channel RN that terminates, with high probability, in O(n/k + (log n)2) time slots with no station being awake for more than O(log log n) time slots.

Precise interconnect analysis is strongly required for giga-scale integration the operation frequency of which is excess 10 GHz. In this study, detailed and accurate analyses of a coaxial interconnect and an actual rectangular interconnect have been performed by the direct evaluation of Maxwell's equations and the finite element method, respectively. It has been revealed that there are two propagation modes for LSI interconnects: skin depth limited propagation mode and interconnect induced slow wave mode. In a miniaturized interconnect, the propagation mode is the interconnect induced slow wave mode; therefore, we cannot obtain the light-speed propagation due to such an interconnect-induced effect. In order to overcome this speed limitation or to improve signal integrity, it is essential to introduce a short interconnect for a miniaturized structure, and a much larger interconnect than the skin depth. We propose a gas-isolated interconnect as a candidate for an ultimately low-k structure in order to increase the signal-propagation speed. By the introduction of such structures, the performance of miniaturized devices in the deep submicron region will be effectively enhanced.

For ubiquitous computing and communication, a mobile node needs to change upon movements its network- and link-layer points of attachment to the Internet. Conventionally, the network-layer protocol is unaware of the link-layer changes, so IP datagrams to and from a moving node could be lost. Considerable system performance degradation might therefore result. This paper presents a scheme to integrate the handoff procedures of the two layers in the context of Mobile IP with route optimization. Experiment results show that our scheme is promising and can reduce packet loss due to host migrations substantially.

A novel delta-sigma modulator that utilizes a resonant-tunneling diode (RTD) quantizer is proposed and its operation is investigated by HSPICE simulations. In order to eliminate the signal-to-noise-and-distortion ratio (SINAD) degradation caused from the poor isolation of a single-stage quantizer (1SQ), a three-stage quantizer (3SQ), which consists of three cascoded RTD quantizers, is introduced. At a sample rate of 10 Gsps (samples per a second) and a signal bandwidth of 40 MHz (oversampling ratio of 128), the modulator demonstrates a SINAD of 56 dB, which corresponds to the effective number of bits of 9.3.

We consider the polymatroid packing and covering problems. The polynomial time algorithm with the best approximation bound known for either problem is the greedy algorithm, yielding guaranteed approximation factors of 1/k for polymatroid packing and H(k) for polymatroid covering, where k is the largest rank of an element in a polymatroid, and H(k)=Σi=1k 1/i is the kth Harmonic number. The main contribution of this note is to improve these bounds by slightly extending the greedy heuristics. Specifically, it will be shown how to obtain approximation factors of 2/(k+1) for packing and H(k)-1/6 for covering, generalizing some existing results on k-set packing, matroid matching, and k-set cover problems.

We propose two novel blind LMS algorithms, called exponential step size LMS algorithms (ES-LMS), for adaptive array antennas with fast convergence speeds. Both of the proposed algorithms are much better at tracking signal sources than the conventional LMS algorithms. In addition, they require neither spatial knowledge nor reference signals since they use the finite symbol property of digital signal. Computer simulations verify performances of the two proposed algorithms.

A kind of online edge-coloring problems on bipartite graphs is considered. The input is a graph (typically with no edges) and a sequence of operations (edge addition and edge deletion) under the restriction that at any time the graph is bipartite and degree-bounded by k, where k is a prescribed integer. At the time of edge addition, the added edge can be irrevocably assigned a color or be left uncolored. No other coloring or color alteration is allowed. The problem is to assign colors as many times as possible using k colors. Two algorithms are presented: one with competitiveness coefficient 1/4 against oblivious adversaries, and one with competitiveness coefficient between 1/4 and 1/2 with the cost of requiring more random bits than the former algorithm, also against oblivious adversaries.

If there exist any two vertices in G whose distance becomes longer when a vertex u is removed, then u is defined as a hinge vertex. Finding the set of hinge vertices in a graph is useful for identifying critical nodes in an actual network. A number of studies concerning hinge vertices have been made in recent years. In a number of graph problems, it is known that more efficient sequential or parallel algorithms can be developed by restricting classes of graphs. In this paper, we shall propose a parallel algorithm which runs in O(log n) time with O(n) processors on CREW PRAM for finding all hinge vertices of a trapezoid graph.

We consider equalizer initialization problems when the transmitted symbol rate is higher than the available channel bandwidth. In this case, the coefficients of an adaptive equalizer in the receiver can be updated only once per a predefined symbol period, requiring unacceptably long training time. The training time can be reduced significantly if the equalizer begins the training process from a properly initialized condition. In this letter, a fast initialization method is analytically designed for a minimum mean squared error (MMSE) type equalizer. Finally, the initialization performance is verified by computer simulation.

M-convex functions have various desirable properties as convexity in discrete optimization. We can find a global minimum of an M-convex function by a greedy algorithm, i.e., so-called descent algorithms work for the minimization. In this paper, we apply a scaling technique to a greedy algorithm and propose an efficient algorithm for the minimization of an M-convex function. Computational results are also reported.

Most conventional studies on self-stabilization have been indifferent to the vulnerability under convergence. This paper investigates how mutual exclusion property can be achieved in self-stabilizing rings even for illegitimate configurations. We present a new method which uses a state with a large state space to detect faults. If some faults are detected, every process is reset and not given a privilege. Even if the reset values are different between processes, our protocol mimics the behavior of Dijkstra's unidirectional K-state protocol. Then we have a fast and safe mutual exclusion protocol. Simulation study also examines its performance.

The main objective of vehicle suspensions is to improve ride comfort and road holding ability. Though passive suspensions consist of spring and damper, active suspensions adopt an actuator in addition to passive suspensions. In this paper, a quarter car model with an asymmetric hydraulic actuator is used. Moreover, the damping coefficient of the damper, which is changed according to the actuator velocity, is considered. The LPV (Linear Parameter Varying) model is obtained by applying feedback linearization technique. Next, a gain-scheduled controller, based on LQ regulator with different weighting factor, is designed according to the actuator velocity and the stability of the proposed controller is also proved. The effectiveness of the proposed controller is shown by numerical simulations.

This paper discusses the application of a fuzzy-ARTMAP neural network to digital communications channel equalization. This approach provides new solutions for solving the problems, such as complexity and long training, which found when implementing the previously developed neural-basis equalizers. The proposed fuzzy-ARTMAP equalizer is fast and easy to train and includes capabilities not found in other neural network approaches; a small number of parameters, no requirements for the choice of initial weights, automatic increase of hidden units, no risk of getting trapped in local minima, and the capability of adding new data without retraining previously trained data. In simulation studies, binary signals were generated at random in a linear channel with Gaussian noise. The performance of the proposed equalizer is compared with other neural net basis equalizers, specifically MLP and RBF equalizers.

This paper proposed an automated segmentation algorithm for MR brain images through the complementary use of T1-weighted, T2-weighted, and PD images. The proposed segmentation algorithm is composed of 3 steps. The first step involves the extraction of cerebrum images by placing a cerebrum mask over the three input images. In the second step, outstanding clusters that represent the inner tissues of the cerebrum are chosen from among the 3-dimensional (3D) clusters. The 3D clusters are determined by intersecting densely distributed parts of a 2D histogram in 3D space formed using three optimal scale images. The optimal scale image results from applying scale-space filtering to each 2D histogram and a searching graph structure. As a result, the optimal scale image can accurately describe the shape of the densely distributed pixel parts in the 2D histogram. In the final step, the cerebrum images are segmented by the FCM (Fuzzy c-means) algorithm using the outstanding cluster center value as the initial center value. The ability of the proposed segmentation algorithm to calculate the cluster center value accurately then compensates for the current limitation of the FCM algorithm, which is unduly restricted by the initial center value used. In addition, the proposed algorithm, which includes a multi spectral analysis, can achieve better segmentation results than a single spectral analysis.

We investigate a method to suppress the polarization-dependent loss (PDL) of long-period fiber gratings (LPFGs). We study the origins of the PDL and propose an azimuthally isotropic UV exposure to suppress the UV-induced birefringence and to realize low-PDL LPFGs. By using this technique and a low birefringent fiber together, the PDL of LPFGs can be reduced to a sufficiently low level required in high performance communication systems. Moreover, the validity of our theoretical modeling is confirmed by the experimental results.

The group organization used for group learning in a knowledge intensive domain like software development affects educational achievement. This paper proposes a group organization system for software engineering education done through group learning. The organizational problem itself is defined and why a Genetic Algorithm (GA) is an appropriate means of solving this problem is explained. This system is a Web application developed with open source software and runs on an open source software platform. Based on the group organization data collected from actual classes, we generated various group organizations by using different strategy parameter values. We then gave a questionnaire to actual students asking them which solution produced the fairest group organization. The replies received revealed that the candidate solution that set greater weight on leadership capability and system analysis capability was the fairest.

This paper proposes a scheme for joint synchronization between stored media with interactive control and live media in multicast communications. We deal with visual search control, such as fast-forward and fast-reverse, as interactive control. The proposed scheme enables visual search by enhancing the virtual-time rendering (VTR) media synchronization algorithm, which the authors previously proposed, and adjusts the timing of changing the visual search mode among destinations by carrying out group synchronization control. We also demonstrate the effectiveness of the scheme by experiment.

In modern high-capacity wavelength division multiplexing (WDM) transmission systems, there is increasing demand for large transmission capacity. To achieve this purpose, an L-band (1565-1625 nm) erbium-doped fiber amplifier (EDFA) is very effective method because the conventional silica-based EDF can be used. In EDFAs that used in WDM transmission systems, the gain flatness of EDFA is very important. A passive gain equalizer flattens the gain profile of EDFA. But the gain flatness in L-band deteriorates due to dynamic gain-tilt (DGT) and temperature gain-tilt (TGT) when the operating condition of the EDFA changes, while the EDFAs should maintain the gain flatness even if the operating condition has changed. To solve this problem, we propose an active gain-slope compensation technique for the L-band EDFA using a thulium-doped fiber (TDF). The EDFA actively gain-slope compensated by the TDF compensator keeps the gain profile constant for the wide input power range of more than 8 dB, a wide temperature range of 65 without gain-tilt in a wavelength band between 1575 nm and 1610 nm. Furthermore, the EDFA keeps a low noise figure of less than 7.5 dB.

The redundancy allocation problem for a series-parallel system is a well known as one of NP-hard combinatorial problems and it generally belongs to the class of nonlinear integer programming (nIP) problem. Many researchers have developed the various methods which can be roughly categorized into exact solution methods, approximate methods, and heuristic methods. Though each method has both advantages and disadvantage, the heuristic methods have been received much attention since other methods involve more computation effort and usually require larger computer memory. Genetic algorithm (GA) as one of heuristic optimization techniques is a robust evolutionary optimization search technique with very few restrictions concerning with the various design problems. However, GAs cannot guarantee the optimality and sometimes can suffer from the premature convergence situation of its solution, because it has some unknown parameters and it neither uses a priori knowledge nor exploits the local search information. To improve these problems in GA, this paper proposes an effective hybrid genetic algorithm based on, 1) fuzzy logic controller (FLC) to automatically regulate GA parameters and 2) incorporation of the iterative hill climbing method to perform local exploitation around the near optimum solution for solving redundancy allocation problem. The effectiveness of this proposed method is demonstrated by comparison results with other conventional methods on two different types of redundancy allocation problems.