Prospects of photonic switching technologies are described from the viewpoint of an experienced switching system engineer. Photonic switching technologies are classified into five categories, and each of them has the explanation about basic device operation principles with the experimental switching system technologies. The author's view of present technologies is explained with comparison table about these five typical system technologies. Finally, the expectation to the photonic switching system is described that it will be introduced at the beginning of 21st century.

This paper proposes a multilingual set of speech samples serving as the data base for standardizing an artificial voice which will be used to evaluate the talker dependency of digital coding algorithms. To investigate the impartiality of this data base, the fundamental statistical speech characteristics of these samples-long-term average spectrum, instantaneous amplitude distribution, segmental power, fundamental frequency, and voiced/ unvoiced ratio-are analyzed. Detailed dispersions in the samples as well as average values are derived. Taking the dispersions into account, the results agree with previous research. Speech quality dependency on talker and language is investigated when this set of speech samples is coded using three typical digital coding algorithms: PCM, ADPCM, and APC-AB. This set of speech samples reduces the evaluation bias caused by small speech-sample sizes. These multilingual speech samples are stored on CD-ROM and are publicly available.

Recent advances in VLSI memories have enabled integration of 10 to 30 million devices on prototype chips for 16 Mbit DRAMs, 16 Mbit EPROMs, and 4 Mbit SRAMs. An experimental 64 Mbit DRAM recently reported clearly shows that an integration density of more than 100 million devices on a chip will be feasible in the near future. These advances have been made not only by progress in fine processing technology, but also by the development of three-dimensional memory cells such as trench capacitor cells and stacked capacitor cells for DRAMs and polysilicon PMOS load cells for SRAMs. Various circuit innovations to increase the signal-to-noise ratio and circuit speed have been, and will continue to be, essential. Future circuits will be required to operate at very low voltages, and the prototype 64 Mbit DRAM has shown that an operating voltage as low as 1.5 volts is feasible. Improvements in packaging technology for reducing package volume and footprint area, as well as for production of multipackage modules, are also becoming more and more important.

A new concept of fuzzy tensions and fuzzy potentials is proposed as a mathematical tool for qualitative analysis of network problems. In this paper we first describe modeling of curriculum networks and fail-safe networks in which the concept of fuzzy tensions appears naturally as a well-balanced curriculum plan or reliability assignment. Algebraic properties of fuzzy tensions and potentials are thoroughly investigated and consistency of specifications is shown to be expressible by a simple condition on circuits of the underlying network.

Two new technologies, coherent optical fiber communications and optical fiber amplifiers, are now combined together to bring forth a new era in optical fiber communications research. The purpose of this paper is to describe two topics in this innovative research area: various diversity techniques in coherent optical fiber communications, and combination of the coherent optical fiber communications with erbium-doped fiber amplifiers (EDFAs).

Recently, a great deal of progress has been made in the area of optical amplifiers and many experiments using them have been conducted aiming at practical applications to various transmission systems. This paper reviews recent developments in optical amplifiers, focusing on Er-doped fiber amplifiers (EDFAs), and their prospective applications. Included are introductory amplification theory to clarify the features of optical amplifiers, comparison of optical amplifiers focusing on EDFAs and semiconductor laser amplifiers (SLAs), and recent typical system applications using EDFAs, including IM/DD transmission, analog transmission, coherent transmission, FDM transmission and other applications.

This paper presents the synchronization mechanisms of the highly parallel dataflow machine EM-4 with some results of measurement. First, various synchronization mechanisms of parallel computers are surveyed and compared, including dataflow synchronization. Then, the fundamental synchronization mechanisms of the EM-4 are shown, examining the reason why they are adopted. There are three types of synchronizations: (1) strongly connected instruction sequencing, (2) instruction level direct matching, and (3) function level synchronization. These mechanisms are preliminary evaluated on the EM-4 prototype, and the results are reported and analyzed. Next, synchronization mechanisms for resource managements are described.

Focusing on gate array and standard cell design LSIs, technology trends in ASIC are discussed. MOS transistors with LDD or modified LDD structures will be effective down to around 0.4 µm. Upon further miniaturization, simple single-drain structure MOS transistors will become prominent. SOI-structure MOS transistors may be even more effective with a smaller short channel effect. The supply voltage should be lowered to 3.3 V for 0.5 µm CMOS LSIs to decrease power dissipation owing to the increase of gate count and operation speed. SOG gate arrays will increase their share of the ASIC market. BiNMOS circuits will be useful under 5 V VDD condition. However, below 3.0 V, CMOS circuits will be preferable. In the near future, tpd of 20-30 ps/gate and 40-50 ps/gate should be attained for GaAs FET and Si-bipolar LSIs, respectively.

Introduction of fiber-optic systems to the subscriber network is the key issue for the next generation network. Fiber-optic systems have, so far, been limited to business applications. However, the large scale introduction of fiber-optic systems to all customers, especially to the home and to small business customers is indispensable for the smooth evolution of B-ISDN. Accordingly, NTT plans to introduce a fiberoptic subscriber network widely to the home from around 1995 and to achieve an all fiber subscriber network by 2015. This paper presents the development and deployment concept for a fiber-optic subscriber network. It also proposes the fiber-optic network services that should be offered upon fiber-optic system deployment. Near term system development and deployment strategies are also discussed.

The present paper describes an efficient heuristic approach to state encoding of sequential circuits so that the total usage of silicon area of PLA-based two-level implementation for combinational circuits can be minimized. Apart from those conventional methods in which state encoding is determined one after another in a predetermined order of states, this approachi is distinctive in that top priority is given to how to seek cubes which constitute a cover to be allocated to each group of states (where a group of states denotes a set of states which have the same next state and the same output for an input) on the basis of a magnitude estimation of cubes so that the total number of cubes to be used can be minimized. Implementation results are also shown to reveal how this approach is of practical use.

The advances in silicon IC technology have provided an incredible performance increase in MPU developments. This paper describes that history of microprocessor developments, as well as future direction of MPU developments from the viewpoint of architectural design. Also, an empirical study of the development shows that changes of MPU generation occurs every four years with rapid performance increase between generations.

The spherical electromagnetic wave backscattering from a random rough terrain is investigated. The incoherent backscattered power is computed to examine how the antenna beamwidth and wave sphericity influence the magnitude of copolarized terrain radar backscattering.

In this letter, we present a channel access strategy with both downward and upward (thus called both-way) overflow mechanisms, which allocates channel bandwidth dynamically to multiple heterogeneous traffics with differing intensities, bit rates and holding times. Each input traffic has its own service channel and finite waiting room and can overflow to any other channels. The threshold parameter which limits overflow is also considered, and the performance of this channel access strategy is analyzed as an M/M/S/N system with state-dependent arrival rates.

A rapid, stable, and simple triangulation method is presented for approximating equi-valued surfaces by polyhedra. It is applicable to various scientific and engineering fields, because equi-valued surfaces can be closed or open, and because spatial functions can be given by a parameter set of defining functions or by the numerical values of functions at grid points. The method is tested numerically to compare its accuracy, computation time, and data volume with those of other methods.

A simple method for calculating the optimum transit angles of QWITT diodes is proposed. The carrier diffusion effect and the drift velocity transient effect in QWITT diodes are considered. Reasonably good agreement between our results and those by a more rigorous analysis is obtained.

A graph associated with a linear code, which originates from a δ-decodable code pair for the two-user binary adder channel, is investigated based on the structure of the linear code. Subgraphs of the graph that are induced by cosets of the linear code are introduced. It is found that these are vertextransitive and are disconnected for uniquely decodable (1-decodable) code pair. Moreover, a class of graphs associated with linear codes is proved to consist of clique components and their independence numbers are successfully formulated. Applications to channel coding for the two-user binary adder channel are also discussed.

Time transient scattering field dur to a line source located at the center of the dielectric cylinder has been calculated. In the analysis, the corresponding time harmonic result has been formulated first rigorously, then the high frequency asymptotic expansion result has been derived. Thus obtained result is found to coincide wiht the one constructed directly by ray approximation. Fourier invesion for an impulsive response has been done by two methods, namely the Singularity Expansion Method and wavefront expansion method. While the former method collects the contributions around the singularities in the complex frequency domain, the latter gives us a result which is a summation of each successive wavefront arrivals. A finite Hilbert transform technique has been introduced to recover the causal responses of odd-time caustic passing wavefronts. Also derived are results of numerical inversion by Fast Fourier Transform Technique for the frequency band limited incident pulses. A Gaussian pulse has been introduced to simulate an impulse response result, and a raised cosine pulse which de-emphasizes the low frequency defects of asymptotically constructed frequency spectrum confirms the usefulness of ray solution.

This paper studies the method of parallel processing for two dimensional recursive filters on a multiprocessor system. Conventional recursive filterings are sequential and iterative local, i.e. global processing. We decompose their global processings into space partition processings with a few global communications. We derive an efficient parallel algorithm for two dimensional recursive filterings using Roesser's model and investigate their speed-up, rate, efficiency and degradation.