We have implemented a distributed sensor system based on an array of fiber Bragg gratings (FBGs), which can measure up to 1000 points with a single piece of fiber. The system consists of FBGs having the same resonant wavelengths and small reflectivities (0.1 dB), and a wavelength tunable optical time-domain reflectometer (OTDR). To interrogate the distributed grating sensors and to address the event locations simultaneously, we have utilized the tunable OTDR. A thermoelectric temperature controller was used to tune the emission wavelength of the OTDR. The operating temperature of the laser diode was changed. By tuning the pulse wavelength of the OTDR, we could identify the FBGs whose resonant wavelengths were under change within the operating wavelength range of the DFB LD. A novel sensor cable with dry core structure and tensile cable was fabricated to realize significant construction savings at an industrial field and in-door and out-door applications. For experiments, a sensor cable having 52 gratings with 10 m separations was fabricated. To prevent confusion with unexpected signals from the front-panel connector zone of the OTDR, a 1 km buffer cable was installed in front of the OTDR. The proposed system could distinguish and locate the gratings that were under temperature variation from 20 to 70.

We implemented all-fiber delay line using linearly chirped fiber Bragg gratings (CFBG), which can be applicable for reflectometry or optical coherence tomography (OCT). Compared with the previously reported delay lines, the proposed fiber-based optical delay line has in principle novel advantages such as automatic dispersion cancellations without additional treatment and a gain in optical delay that is dependent on parameters of used CFBGs. Dispersion compensation in optical delay line (ODL), which is the indispensable problem in bulk optics based ODL, is demonstrated in fiber by using two identical but reversely ordered CFBGs. Amplified variable optical delay of around 2.5 mm can be obtained by applying small physical stretching of one of CFBGs in the proposed scheme. The operational principles of the all-fiber variable optical delay line, which are based on the distributed reflection characteristic of a CFBG employed, are described. Especially properties such as in-line automatic dispersion cancellation and amplified optical delay under strain are dealt. To demonstrate the properties of the proposed scheme, which is theoretical consequences under assumptions, an all-fiber optical delay line have been implemented using fiber optic components such as fiber couplers and fiber circulators. With the implanted ODL, the group delay and amplified optical delay length was measured with/without strain. The wavelength independent group delay measured within reflection bandwidth of the CFBG has proved the property of automatic dispersion cancellations in the proposed fiber delay line. Optical delay length of 2.5 mm was obtained when we apply small physical stretching to the CFBG by 100 µm and this is expressed by the amplification factor of 25. Amplification factor 25, which is less than theoretical value of 34 due to slipping of fiber in the fiber holder, shows that the proposed scheme can provide large optical delay with applying small physical stretching to the CFBG. We measure slide glass thickness to check the performance of the fiber delay line and the good agreement in measured and physical thickness of slide glass (1 mm thick) validates the potential of proposed delay line in the applications of optical reflectometry and OCT. We also discuss the problem and the solution to improve the performance.

This paper investigates the application of the computational model of Cellular Automata (CA) for pattern classification of real valued data. A special class of CA referred to as Fuzzy CA (FCA) is employed to design the pattern classifier. It is a natural extension of conventional CA, which operates on binary string employing boolean logic as next state function of a cell. By contrast, FCA employs fuzzy logic suitable for modeling real valued functions. A matrix algebraic formulation has been proposed for analysis and synthesis of FCA. An efficient formulation of Genetic Algorithm (GA) is reported for evolution of desired FCA to be employed as a classifier of datasets having attributes expressed as real numbers. Extensive experimental results confirm the scalability of the proposed FCA based classifier to handle large volume of datasets irrespective of the number of classes, tuples, and attributes. Excellent classification accuracy has established the FCA based pattern classifier as an efficient and cost-effective solutions for the classification problem.

This paper describes a formal verification for a shared-memory-style communication system. We first describe two versions (i.e. abstract and concrete) of the communication system based on an I/O-automaton, which is a formal system for distributed algorithms. Then, we prove the concrete version can perform all the external operations of the abstract version. This result, together with a former result, leads to the equivalence of the two versions. The proof is done by Larch theorem prover, and is the ever largest case study using I/O-automata.

This paper presents a new framework for effectively creating VoiceFonts for speech synthesis. A VoiceFont in this paper represents a voice inventory aimed at generating personalized voices. Creating well-formed voice inventories is a time-consuming and laborious task. This has become a critical issue for speech synthesis systems that make an attempt to synthesize many high quality voice personalities. The framework we propose here aims to drastically reduce the burden with a twofold approach. First, in order to substantially enhance the accuracy and robustness of automatic speech segmentation, we introduce a multi-layered speech segmentation algorithm with a new measure of segmental reliability. Secondly, to minimize the amount of human intervention in the process of VoiceFont creation, we provide easy-to-use functions in a data viewer and compiler to facilitate checking and validation of the automatically extracted data. We conducted experiments to investigate the accuracy of the automatic speech segmentation, and its robustness to speaker and style variations. The results of the experiments on six speech corpora with a fairly large variation of speaking styles show that the speech segmentation algorithm is quite accurate and robust in extracting segments of both phonemes and accentual phrases. In addition, to subjectively evaluate VoiceFonts created by using the framework, we conducted a listening test for speaker recognizability. The results show that the voice personalities of synthesized speech generated by the VoiceFont-based speech synthesizer are fairly close to those of the donor speakers.

A new automatic-threshold-control (ATC) system for the amplitude-shift-keying (ASK) transceivers without using automatic gain control (AGC) was developed. When signal-to-noise ratio (SNR) in a transmission system fluctuates, the new system optimizes the threshold by statistic processing. In this paper, the schematic block diagram, the theory of statistical processing, and the performance estimated by numerical simulations are shown. The simulations show that under ideal conditions, the system can control the threshold voltage in a broad SNR region (wider than 20 dB). Delay of response and trapping in real situation can occur, but, the problems can be avoided by waveform shaping. It is thus concluded that the new ATC system can be applied to ASK transceivers.

Appropriate language modeling is one of the major issues for automatic transcription of spontaneous speech. We propose an adaptation method for statistical language models based on both topic and speaker characteristics. This approach is applied for automatic transcription of meetings and panel discussions, in which multiple participants speak on a given topic in their own speaking style. A baseline language model is a mixture of two models, which are trained with different corpora covering various topics and speakers, respectively. Then, probabilistic latent semantic analysis (PLSA) is performed on the same respective corpora and the initial ASR result to provide two sets of unigram probabilities conditioned on input speech, with regard to topics and speaker characteristics, respectively. Finally, the baseline model is adapted by scaling N-gram probabilities with these unigram probabilities. For speaker adaptation purpose, we make use of a portion of the Corpus of Spontaneous Japanese (CSJ) in which a large number of speakers gave talks for given topics. Experimental evaluation with real discussions showed that both topic and speaker adaptation reduced test-set perplexity, and in total, an average reduction rate of 8.5% was obtained. Furthermore, improvement on word accuracy was also achieved by the proposed adaptation method.

Automatic labeling of prosodic features is an important topic when constructing large speech databases for speech synthesis or analysis purposes. Perceptually-related F0 parameters are proposed with the aim of automatically classifying phrase final tones. Analyses are conducted to verify how consistently subjects are able to categorize phrase final tones, and how perceptual features are related with the categories. Three types of acoustic parameters are proposed and analyzed for representing the perceptual features related to the tone categories: one related to pitch movement within the phrase final, one related to pitch reset prior to the phrase final, and one related to the length of the phrase final. A classification tree is constructed to evaluate automatic classification of phrase final tones, resulting in 79.2% accuracy for the consistently categorized samples, using the best combination among the proposed acoustic parameters.

This paper presents the selective orthogonal matrix least-squares (SOM-LS) method for representing a multiport network characterized by sampled data with the rational matrix, improving the previous works, and providing new criteria. Recently, it is needed in a circuit design to evaluate physical effects of interconnects and package, and the evaluation is done by numerical electromagnetic analysis or measurement by network analyzer. Here, the SOM-LS method with the criteria will play an important role for generating the macromodels of interconnects and package in circuit simulation level. The accuracy of the macromodels is predictable and controllable, that is, the SOM-LS method fits the rational matrix to the sampled data, selecting the dominant poles of the rational matrix. In examples, simple PCB models are analyzed, where the rational matrices are described by Verilog-A, and some simulations are carried out on a commercial circuit simulator.

Multichannel superconducting quantum interference device (SQUID) systems based on double relaxation oscillation SQUIDs (DROS) were developed for measuring magnetocardiography (MCG) and magnetoencephalography (MEG) signals. Since DROS provides large flux-to-voltage transfer coefficients, about 10 times larger than the DC SQUIDs, direct readout of the SQUID output was possible using compact room-temperature electronics. Using DROSs, we fabricated two types of multichannel systems; a 37-channel magnetometer system with circular sensor distribution for measuring radial components of MEG signals, and two planar gradiometer systems of 40-channel and 62-channel measuring tangential components of MCG or MEG signals. The magnetometer system has external feedback to eliminate magnetic coupling with adjacent channels, and reference vector magnetometers were installed to form software gradiometers. The field noise of the magnetometers is around 3 fT/ at 100 Hz inside a magnetically shielded room. The planar gradiometer systems have integrated first-order gradiometer in thin-film form with a baseline of 40 mm. The magnetic field gradient noise of the planar gradiometers is about 1 fT/cm/ at 100 Hz. The planar gradiometers were arranged to measure field components tangential to the body surface, providing efficient measurement of especially MCG signals with smaller sensor coverage than the conventional normal component measurements.

We consider the edge-linking approach for accurate locating of moving object boundaries in video segmentation. We review the existing methods and propose a scheme designed for efficiency and better accuracy. The scheme first obtains a very rough outline of an object by a suitable means, e.g., change detection. It then forms a relatively compact image region that properly contains the object, through a procedure termed "mask sketch." Finally, the outermost edges in the region are found and linked via a shortest-path algorithm. Experiments show that the scheme yields good performance.

Δ-Timed Atomic Broadcast is the broadcast ensuring that all correct processes deliver the same messages in the same order, and that delivery latency of any message broadcast by any correct process is some predetermined time Δ or less. In this paper, we propose a Δ-timed atomic broadcast algorithm in a synchronous system where communication delay is bounded by a known constant d and processes suffer both crash faults and timing faults. The proposed algorithm can tolerate fc crash faults and ft timing faults as long as at least ft + 1 processes are correct, and its maximum delivery latency Δ is (2f' + 7)d where f' is the actual number of (crash or timing) faulty processes. That is, the algorithm attains the early-delivery in the sense that its delivery latency depends on the actual number of faults rather than the maximum number of faults that the algorithm can tolerate. Moreover, the algorithm has a distinct advantage of guaranteeing that timing-faulty processes also deliver the same messages in the same order as the correct processes (Uniformity). We also investigate the maximum number of faulty processes that can be tolerated. We show that no Δ-timed atomic broadcast algorithm can tolerate ft timing faults, if at most ft processes are correct. The impossibility result implies that the proposed algorithm achieves the maximum fault-resilience with respect to the number of faulty processes.

This paper discusses real-time language recognition by 1-dimensional one-way cellular automata (OCAs) and two-way cellular automata (CAs), focusing on limitations of the parallel computation power. To clarify the limitations, we investigate real-time recognition of cyclic strings of the form uk with u {0,1}+ and k 2. We show a version of pumping lemma for recognizing cyclic strings by OCAs, which can be used for proving that several languages are not recognizable by OCAs in real time. The paper also discusses the real-time language recognition of CAs by prefix and postfix computation, in which every prefix or postfix of an input string is also accepted, if the prefix or postfix is in the language. It is shown that there are languages L Σ+ such that L is not recognizable by OCA in real-time and the reversal of L and the concatenation LΣ* are recognizable by CA in real-time.

For each positive integer r 1, a nondeterministic machine M is r path-bounded if for any input word x, there are r computation paths of M on x. This paper investigates the accepting powers of path-bounded one-way (simple) multihead nondeterministic finite automata. It is shown that for each k 2 and r 1, there is a language accepted by an (r + 1) path-bounded one-way nondeterministic k head finite automaton, but not accepted by any r path-bounded one-way nondeterministic k head finite automaton whether or not simple.

This paper studies a cellular system with mobile customers. The network system consists of cells, the tagged cell and the adjacent cells which surround the tagged one. Each cell has a finite number of channels that give calls to the mobile customers. The service (holding) time distribution of the calls is general. Customers in the adjacent cells inflow into the tagged cell according to a Poisson process. The sojourn time distribution of each customer in the tagged cell is general. Each customer without call in progress generates his call according to a Poisson process. It is proved that the steady state distribution in the tagged cell is the generalized Erlang loss formula which is the joint distribution of the number of customers with calls and the number of customers without calls. The distribution depends on the service time distribution and the sojourn time distribution only through their means.

This paper treats multi-way microstrip power dividers composed of multi-step, multi-furcation, and corners. Since the design procedure is founded on the planar circuit approach in combination with the segmentation method, optimization of the circuit configuration can be performed in a reasonable short computation time when applying the Powell's optimization algorithm. Actually, broadband 3- and 4-way power dividers with mitered bends are designed, and fractional bandwidths of about 90% and 100% are realized for the power-split imbalance less than 0.2 dB and the return loss better than -20 dB, respectively. The validity of the design results is confirmed by an EM-simulator (HFSS) and experiments.

In this letter we first introduce a new generalized cyclotomic sequence of order four with respect to pq, then we calculate the linear complexity and minimal polynomial of this sequence. Our results show that the new binary sequence is quite good from the linear complexity viewpoint.

Inoue et al. introduced an automaton on a two-dimensional tape, which decides acceptance or rejection of an input tape by scanning the tape from various sides by various automata which move one way, and investigated the accepting power of such an automaton. This paper continues the investigation of this type of automata, especially, -type automata (obtained by combining four three-way two-dimensional deterministic finite automata (tr2-dfa's) in "or" fashion) and -type automata (obtained by combining four tr2-dfa's in "and" fashion). We first investigate a relationship between the accepting powers of -type automata and -type automata, and show that they are incomparable. Then, we investigate a hierarchy of the accepting powers based on the number of tr2-dfa's combined. Finally, we briefly describe a relationship between the accepting powers of automata obtained by combining three-way two-dimensional deterministic and nondeterministic finite automata.

PUE films have been found to exhibit an electrostriction effect. We propose the applying them to a moving device such as an actuator similar to artificial muscles using the electrostriction effect. The actuators are of monomorph type fabricated by PUE film and metal electrodes evaporated at different thickness on the film surfaces. Because these actuators work at a high voltage of more than 1 KV, we controlled the molecular structure of the films by doping C60 or CNT derivatives into PUE so that the actuators could operate under a low voltage. The bends of C60 and CNT-doped actuators were larger than those of non-doped actuators and the working voltage was also low. The force of the actuators increased in proportion to the electric field, and strongly depended on the thickness of the PUE films. Furthermore, in order to clarify the relationship between the stretch of PUE film and the bending mechanism of actuators, we measured the space charge of PUE films using the pulsed electroacoustic method.

An important issue in applying machine learning algorithms to Natural Language Processing areas such as Named Entity Recognition tasks is to overcome the lack of tagged corpora. Several bootstrapping methods such as co-training have been proposed as a solution. In this paper, we present a different approach using the Web resources. A Named Entity (NE) tagged corpus is generated from the Web using about 3,000 names as seeds. The generated corpus may have a lower quality than the manually tagged corpus but its size can be increased sufficiently. Several features are developed and the decision list is learned using the generated corpus. Our method is verified by comparing it to both the decision list learned on the manual corpus and the DL-CoTrain method. We also present a two-level classification by cascading highly precise lexical patterns and the decision list to improve the performance.