The visual inspection of printed circuit boards (PCBs) at the final production stage is necessary for quality assurance and the requirements for an automated inspection system are very high. However, consistent inspection of patterns on these PCBs is very difficult due to pattern complexity. Most of the previously developed techniques are not sensitive enough to detect defects in complex patterns. To solve this problem, we propose a new optical system that discriminates pattern types existing on a PCB, such as copper, solder resist and silk-screen printing. We have also developed a hybrid defect detection technique to inspect discriminated patterns. This technique is based on shape measurement and features extraction methods. We used the proposed techniques in an actual automated inspection system, realizing real time transactions with a combination of hardware equipped with image processing LSIs and PC software. Evaluation with this inspection system ensures a 100% defect detection rate and a fairly low false alarm rate (0.06%). The present paper describes the inspection algorithm and briefly explains the automated inspection system.

In this paper, we propose the high-level service architecture supporting multimedia multicast. The proposed architecture specifies network-oriented and lightweight communication management, which includes group management, multiparty call control and multicast connection/multiple connections control. Many of the existing approaches handling multimedia multicast applications are dependent on specific transport technologies, such as the Internet and ATM, and lack effective communication management. On the other hand, our approach defines flexible and extensible communication management that can be applied to a variety of multimedia multicast applications, independently of transport technologies. Our architecture supports the separation of control from a terminal, which enables remote control and control mobility, so that a user can use a multicast service in a more various way. The architecture is overlaid above legacy transport networks so that the existing network protocols are used for connection control. This minimizes the modification to a legacy transport network and enhances the practicality of the architecture. In addition, terminal manager and virtual device concepts are introduced that hide the details of physical devices from an application designer. The architecture consists of several service components that effectively interact with each other on a distributed platform. To verify and evaluate our architecture, we have prototyped the high-level service architecture on a CORBA platform and analyzed the architecture using a simulation.

This paper proposes a new simple method for network measurement. It extracts 6-bit control flags of TCP (Transmission Control Protocol) packets. The idea is based on the unique feature of flag ratios which is discovered by our exhaustive search for the new indexes of network traffic. By the use of flag ratios, one can tell if the network is really congested. It is much simpler than the conventional network monitoring by a network analyzer. The well-known monitoring method is based on the utilization parameter of a communication circuit which ranges from 0% to 100%. One cannot tell the line is congested even if the factor is 100%. 100% means full utilization and does not give any further information. To calculate the real performance of the network, one should estimate the throughput or effective speed of each user. The estimation needs much calculation. Our new method tries to correlate ratios of TCP control flags and network congestion. The result shows the usefulness of this new method. This paper analyzes the reason why the flag ratios show the unique feature.

A management of WWW server is still relying on the expertise and heuristic of administrators, because the comprehensive understandings of server behavior are missing. The administrators should maintain the WWW server with good states that they should investigate the WWW server in real time. Therefore, it is exactly desirable to provide a measurement application that enables the WWW server administrators to monitor WWW servers in the actual operational environment. We developed a measurement application called ENMA (Enhanced Network Measurement Agent) which is specially designed for WWW server state analysis. Furthermore, we applied this application to the large scale WWW server operation to show its implementation and advantages. In this paper, we analyze the WWW server states based on precise monitoring of performance indices of WWW system to help the server management.

A rapid growth of the Internet and proliferation of new multimedia applications lead to demands of high speed and broadband network technologies. Routers are also necessary to follow up the growth of link bandwidths. From this reason, there have been many researches on high speed routers having switching capabilities. To have an expected effect, however, a control parameters set based on traffic characteristics are necessary. In this paper, we analyze the network traffic using the network traffic monitor and investigate the Internet traffic characteristics through a statistical analysis. We next show the application of our analytical results to parameter settings of high speed switching routers. Simulation results show that our approach makes highly utilized VC space and high performance in packet processing delay. We also show the effect of flow aggregation on MPLS. From our results, the flow aggregation has a great impact on the performance of MPLS.

A software requirements specification (SRS) is a document at the first phase of software development. Since it is difficult to make an accurate SRS at the beginning of software development, we propose a supporting method to detect and interpret the inconsistency of SRS. First, we classify and define the inconsistency of SRS. Next, we describe how to detect and interpret the inconsistency of SRS. We use the Requirements Frame Model to detect the inconsistency of SRS. We apply the Dempster and Shafer's theory to interpret the inconsistency of SRS. We illustrate our method with an example.

To bridge a wide gap between the end users and the requirements engineers, we propose a business flow diagram for acquiring users' requirements of the object oriented software development in the business application domain. Each field of this diagram shows either a role or a responsibility of a particular person or an organization. This paper proposes a development method that the engineers acquire the requirements by using our diagrams. We have implemented a supporting tool based on this study for collaborating the requirements engineers with their users. At first, the end users of an information system to be developed draw diagrams representing the flows of information and physical objects in their work from their own points of view. Sometimes the engineers write them with the users. If all users submit their diagrams, then our tool collects them and constructs a total diagram. The requirements engineers analyze the total diagram for improving the business flow. After the engineers complete this diagram, our tool can automatically transform it into an initial version of the class diagram. We show the effectiveness of our approach with some experiments. Comparing the related works, we discuss some issues of the practical aspects of this proposal.

This paper describes a fiber-optic level sensor designed to measure the level of liquid propane-butane in a relatively short range (60 cm) in the top part of storage tanks at oil refineries with the purpose of monitoring the level of this product in the filled or slightly underfilled or overfilled tanks during various measuring operations. A discrete multi-element device employing novel refractometric transducers was selected because it yields both a large measurement range and high resolution. Several innovations offer a competitive advantage to industrial users: 1) Special micro-optical refractometric transducer; 2) Efficient and economical sensor multiplexing scheme; 3) Fast level-tracking operational algorithm. The vertical resolution of the sensor -1 cm, the maximum excess pressure in the tank -40 atm (4 MPa). The sensor has the spark-proof and explosion-proof design and optical fiber interface for the transmission of the output data. The sensor successfully measured liquid propane-butane level in storage tanks during numerous cycles of measuring operations.

This paper describes a load carrying test for a concrete pipe designed to study the effectiveness of distributed strain measurement using an optical fiber sensor. We performed a load carrying test on a concrete pipe and attempted to detect the distributed strain inside it using an optical fiber sensor mounted inside the pipe. We confirmed that it was possible to detect the strain in a concrete structure by using an optical fiber sensor after a crack had occurred on the concrete surface. This paper shows that measurement using the optical fiber sensor was effective despite great changes in the strain conditions of the measured object over a short distance.

A correlation-based technique for measuring Brillouin gain spectrum distribution along an optical fiber is proposed, which employs frequency-modulated pump and probe lightwaves. The spatial-resolution of about 40 cm is demonstrated, which cannot be realized by the conventional pulse-based technique.

The design method for sensitivity and frequency response of an electric field sensor using a Mach-Zehnder interferometer (an optical E-field sensor) has been developed in order to measure electromagnetic environments and the performance of measuring facilities. The designs of the optical modulator, sensor elements, and sensitivity were analyzed theoretically by using an accurate equivalent circuit of the sensor. Then an actual sensor was fabricated, and its characteristics of the sensor were evaluated experimentally. The results show that the designed sensitivity and frequency response were optimal. The optical output deviation when the temperature increased from 0 to 40 was reduced to within 2 dB. The minimum detectable electric field strength was 17 dBµV/m (8 µV/m), and the dynamic range was more than 100 dB. The frequency response of the sensitivity was almost flat between 200 Hz and 900 MHz.

This paper proposes a new self-healing scheme that differentiates the bandwidth requirement for each network service on ATM networks. First, we show the necessity of our proposed scheme. In the future network, we must satisfy two demands, rapid restoration from failure and differentiated bandwidth requirements. The conventional restoration scheme, called the self-healing scheme, realizes rapid restoration, but does not support bandwidth differentiation; the new self-healing scheme proposed herein does. We also show that the proposed scheme reduces the spare resources required for backup. The scheme can be realized as a simple extension of the conventional self-healing scheme. Finally, simulations show that the proposed scheme requires fewer spare resources while offering comparable restoration time to the conventional approach against any demand pattern.

A novel technique has been developed for in situ sensing of thin film growth. In this method, a fiber optic probe is placed at an appropriate position in a deposition chamber, and the thin film builds up on the end of the fiber. This film is either the same as on the wafer where deposition occurs, or it bears a fixed relationship to the film on the wafer. By an analysis of the intensity of the light reflected from the film and guided by the fiber, information on the film may be obtained. With interference causing maxima, minima and a point of inflection as the film grows, it is possible to obtain near real time information on the following quantities: the real and imaginary parts of the refractive index of the film, a Gaussian parameter characterizing surface roughness, and the film thickness itself. To demonstrate this technique, we have studied the deposition of silicon nitride films in a CVD reactor and how reactor temperature and reactant flow rates influence film growth. This technique may be applied to measure in situ reflectivity of multi layer films, so that reflectance as a function of temperature and time may be obtained. Because the measurement is simple and direct and the information is optical, we believe that this technique has the potential to supplant quartz oscillators in the measurement of thin film growth.

We compare between four Connection Admission Control schemes that use either the Gaussian or the Effective Bandwidth model with and without real-time traffic measurements. We demonstrate that under heavy multiplexing, the Gaussian is more efficient than the Effective Bandwidth approach in either case.

A portable cryo-system using a high-Tc SQUID for the measurement of the remanant magnetic field of a rock specimen was designed and fabricated. The sensing surface of the SQUID faces upward in our system, although the system for bio-magnetics faces down. The SQUID is cooled by liquid nitrogen via a sapphire heat transfer rod. The total heat transfer of the system was measured by means of a boiling-off method and was found to be 1.65 W. It was demonstrated that the system can be operated for more than 17 hours without any maintenance such as filling with liquid nitrogen. The system was applied to the measurement of the remanent magnetic field distributions of rock samples cored from deep underground. We have successfully measured the distributions.

We explain three random sampling techniques that are simple but widely applicable for various problems involving huge data sets. The first technique is an immediate application of large deviation bounds. The second and the third ones are sequential sampling or adaptive sampling techniques. We fix one simple problem and explain these techniques by demonstrating algorithms for this problem and discussing their correctness and efficiency.

It is the purpose of this paper to review the generation of quantized voltage steps in Josephson-junctions, and also the recent practical application of these precise measurements. A 10-V Josephson-junction-array-voltage standard system has been established with a Josephson-junction-array, a phase-locked millimeter wave, and a precise null-detection system. Based on these technologies, the AC Josephson effect has been applied to other precise measurements such as DC error voltage of a multi-integrating analog-to-digital converter and for a pulse-width-modulation type precise voltage calibrator.

This paper proposes an associative memory neural network whose limiting state is the nearest point in a polyhedron from a given input. Two implementations of the proposed associative memory network are presented based on Dykstra's algorithm and a fixed point theorem for nonexpansive mappings. By these implementations, the set of all correctable errors by the network is characterized as a dual cone of the polyhedron at each pattern to be memorized, which leads to a simple amplifying technique to improve the error correction capability. It is shown by numerical examples that the proposed associative memory realizes much better error correction performance than the conventional one based on POCS at the expense of the increase of necessary number of iterations in the recalling stage.

This paper proposes a phase assignment algorithm "Silent Wave Algorithm (SWA)" for real-time MPEG traffic in ATM networks. Our algorithm decides when a new MPEG source should begin to transmit based on its notification parameters and traffic measurement of ongoing connections. Simulation results show that it is hard to accommodate MPEG traffic effectively without any control of phase assignment. On the other hand, the SWA can provide better QOS and improve the network utilization.

A method to recognize planar objects undergoing affine transformation is proposed in this paper. The method is based upon wavelet multiscale features and Hopfield neural networks. The feature vector consists of the multiscale wavelet transformed extremal evolution. The evolution contains the information of the contour primitives in a multiscale manner, which can be used to discriminate dominant points, hence a good initial state of the Hopfield network can be obtained. Such good initiation enables the network to converge more efficiently. A wavelet normalization scheme was applied to make our method scale invariant and to reduce the distortion resulting from normalizing the object contours. The Hopfield neural network was employed as a global processing mechanism for feature matching and made our method suitable to recognize planar objects whose shape distortion arising from an affine transformation. The Hopfield network was improved to guarantee unique and more stable matching results. A new matching evaluation scheme, which is computationally efficient, was proposed to evaluate the goodness of matching. Two sets of images, noiseless and noisy industrial tools, undergoing affine transformation were used to test the performance of the proposed method. Experimental results showed that our method is not only effective and robust under affine transformation but also can limit the effect of noises.