We investigated perceptual color assimilation of a strip (test field) displayed on a CRT close to a green or red strip (inducing field) with a dark background. The maximal distance to induce assimilation was about 7 for a red inducing field, and 24 for a blue one. The intensity of assimilation was almost inversely proportional to the width of test field.

In this paper, an analysis of a slotted ring type network, the Pierce type ring network, is carried out and characteristics of the ring are derived. The ring type network is one of the fundamental computer network topologies. It is easy to implement and has many merits such as low cost and expansion flexibility. However, possible hogging, or domination of slots by one station, is one of its drawbacks. At first, when we regard vacant slots as usable, we show how system performance is fairly dependent on traffic patterns. We also show quantitatively that when slots are fixed to each station, domination of the ring can be completely prevented, although the performance of the system will be reduced. We propose a method to control the transmission of packets in which we have positively adopted the insertion of shift registers into the transmission line to prevent the deterioration of the performance. With the insertion of shift registers, we show the feasibility of improving the performance of the fixed slot allocation method at the same time as preventing domination of the ring. We will discuss the merits of the shift register insertion method, which is introduced to improve the fixed slot and semi-fixed slot allocation methods. In the slotted ring type system, four types of slot allocation method including a register insertion method are considered for the improvement of the system performance. The merits and demerits of the methods are discussed. We show how system performance depends on the traffic patterns considered. We analyse the characteristics of the ring system under various conditions.

We believe that virtual world communication will subsume BBS and visual communication. Accordingly, we proposed the networked virtual world "Interspace" for visual communication. If we are to achieve education and training in this world, techniques to receive and transmit information without any special training are necessary. The traditional "easy" ways of transmitting information are talking and drawing. In Interspace, users can already talk each other. In this paper we focus on drawing. In daily life we communicate through drawings in various situations. At this time it is important to recognize who is drawing and where the participants are watching. It is difficult to realize these functions using conventional media, but it is possible to realize them in virtual space. In virtual space, the system can clearly represent who is drawing and where participants are watching; expressing topics in virtual space frees us from many physical restrictions. In this paper we discuss the process of drawing when many participants share topics in virtual space; the necessary conditions for our system are considered. We design a system that offers functions to make drawing sheets, to display the view fields of participants, and to share visual fields. Furthermore, we propose the mode of InterSheet called "InterMirror" which shows mirror images of partners and their drawings. We make a prototype and evaluate it. The results indicate the synergistic effect of drawing with voice and the usefulness of drawing for communication in virtual space.

Discrete higher-order modes of stripline, both proper and leaky, as well as its continuous spectrum are conceptualized and quantified by a full-wave integral operator formulation through complex analysis in the axial Fourier-transform plane. Poles associated with the parallel-plate background environment lead to branch points in the axial transform plane. A criterion for choice of associated branch cuts to restrict the migration of poles in the transverse transform plane is identified. The higher-order discrete modes are both bound (proper), with a low-frequency cutoff, and leaky (improper). The higher-order proper mode has a propagation constant very nearly equal to that for the TM parallel-plate mode. Distributions of the continuous-spectrum currents appear to consist of a smooth transition from those of the highest propagating proper discrete mode, as might be expected physically. The continuous spectrum is dominated by the propagating portions associated with any TM background modes (poles) which are above cutoff, and in fact by spectral points in a region very near to the branch point.

Research in radar polarimetry is hampered by shortcomings of the conventional formulation of polarimetric backscatter concepts. In particular the correct form of the Sinclair backscatter matrix under changes of polarization bases is derived from the antenna voltage (energy transfer) equation yielding the erroneous impression that radar polarimetry is a mongrel between scattering behavior and network performance. The present contribution restores logical consistency in a natural way by introducing the concept of an antilinear backscatter operator. This approach decouples scattering process and network performance, illuminates matrix analytical properties of the radar backscatter matrix and highlights characteristic states of polarization.

The growth characteristics of a two-dimensional Cherenkov laser composed of a planar relativistic electron beam and a parallel plate waveguide one plate of which is loaded with a nonlinear dielectric sheet are analyzed. The permittivity of the nonlinear dielectric sheet becomes inhomogeneous due to the Kerr effect as the electromagnetic wave grows along the waveguide. For the analysis of the electromagnetic fields in the nonlinear dielectric sheet, it is replaced by a number of thin dielectric layers each of which is assumed to be homogeneous. From numerical analysis, it is found that just a few homogeneous layers for the nonlinear dielectric sheet are enough to get the same results as obtained previously by means of the finite element method. This is because the variation of the permittivity across the nonlinear dielectric sheet is as small as within 10% of the linear permittivity of the nonlinear material. Thus the multilayer approximation method is found to be more simple and more efficient for the analysis of the Cherenkov laser loaded with a Kerr-like medium than the finite element method.

The number of solutions of a nonlinear equation x = sgn(Wx) is discussed. The equation is derived for the determination of equilibrium points of a kind of Hopfield neural networks. We impose some conditions on W. The conditions correspond to the case where a Hopfield neural network has n neurons arranged on a ring, each neuron has connections only from k preceding neurons and the magnitude of k connections decrease as the distance between two neurons increases. We show that the maximum number of solutions for the above case is extremely few and is independent of the number of neurons, n, if k is less than or equal to 4. We also show that the number of solutions generally increases exponentially with n by considering the case where k = n-1.

We present a framework for the unsupervised segmentation of time series. It applies to non-stationary signals originating from different dynamical systems which alternate in time, a phenomenon which appears in many natural systems. In our approach, predictors compete for data points of a given time series. We combine competition and evolutionary inertia to a learning rule. Under this learning rule the system evolves such that the predictors, which finally survive, unambiguously identify the underlying processes. The segmentation achieved by this method is very precise and transients are included, a fact, which makes our approach promising for future applications.

A method is proposed for locating large gas leaks in subscriber cable networks composed of different type cables and/or branch cables that uses pressures periodically measured at several points. Theoretical pressure curves corresponding to these measured pressure curves are calculated using a model representing the actual cable network. These curves are called the calculated pressure curves. The position of the leak is varied in the calculation model and where the measured and calculated pressure curves coincide the most, the leak position in the calculation model indicates the estimated leak position. The calculated pressure curves are based on the Laplace transform and an analogy between the pneumatic characteristics in cable networks and the electrical characteristics in electrical transmission networks. An experiment using a cable network about 910m long resulted in a location error of less than 30m.

Cellular Neural Networks (CNNs) are nonlinear dynamic array processors with mainly local interconnections. In most of the applications, the local interconnection pattern, called cloning template, is translation invariant. In this paper, an optimal ring-coding method for rotation invariant description of given set of objects, is introduced. The design methodology of the templates based on the ring-codes and the synthesis of CNN analogic algorithms to detect standing and moving objects in a rotationally invariant way, discussed in detail. It is shown that the algorithms can be implemented using the CNN Universal Machine, the recently invented analogic visual microprocessor. The estimated time performance and the parallel detecting capability is emphasized, the limitations are also thoroughly investigated.

The classification of remotely sensed multispectral data using classical statistical methods has been worked on for several decades. Recently there have been many new developments in neural network (NN) research, and many new applications have been studied. It is well known that NN approaches have the ability to classify without assuming a distribution. We have proposed an NN model to combine the spectral and spacial information of a LANDSAT TM image. In this paper, we apply the NN approach with a normalization method to classify multi-temporal LANDSAT TM images in order to investigate the robustness of our approach. From our experiments, we have confirmed that our approach is more effective for the classification of multi-temporal data than the original NN approach and maximum likelihood approach.

One of the major open issues in neural network research includes a Network Designing Problem (NDP): find a polynomial-time procedure that produces minimal structures (the minimum intermediate size, thresholds and synapse weights) of multilayer threshold feed-forward networks so that they can yield outputs consistent with given sample sets of input-output data. The NDP includes as a sub-problem a Network Training Problem (NTP) where the intermediate size is given. The NTP has been studied mainly by use of iterative algorithms of network training. This paper, making use of both rate distortion theory in information theory and linear algebra, solves the NDP mathematically rigorously. On the basis of this mathematical solution, it furthermore develops a mathematical solution Procedure to the NDP that computes the minimal structure straightforwardly from the sample set. The Procedure precisely attains the minimum intermediate size, although its computational time complexity can be of non-polynomial order at worst cases. The paper also refers to a polynomial-time shortcut to the Procedure for practical use that can reach an approximately minimum intermediate size with its error measurable. The shortcut, when the intermediate size is pre-specified, reduces to a promising alternative as well to current network training algorithms to the NTP.

We propose a new high-performance message router for k-ary n-cube multicomputer systems, called the Tokky router. The router utilizes a small number of queues at the outputs of its communication ports to allow fully adaptive routing, misrouting to prevent deadlocks and randomization to prevent livelock. Uncongeste network performance is improved by the inclusion of the packet expressway. Accurate models are developed to predict the switch and buffer performance of routers for varying radix and dimension and these models can be used in the design of routers for networks other than those investigated here. The simulated performance of the router exceeds that of published results for oblivious routers and is equal to or exceeds those reported for other adaptive routers. These performance predictions are especially encouraging when the simplicity of the control structures required to implement the router are taken into consideration.

The rapid progress in semiconductor technology and chip mounting technology has made it possible to produce portable computes with high performance, such as notebook computers and PDAs. Portable computes can be used anytime and anywhere. By using the features of mobility, some cooperative works anywhere with multiple portable computers have been realized. However, present networks such as a wireless LAN or a public wireless network are too limited to permit portable computers to cooperate with each other. As the result, a new networking technology for portable computers has been highly requested. In this paper, we propose a novel network technology suitable for face-to-face communication, which we call "Wireless DAN" (Desk Area Network). Here, face-to-face communication is a communication between people who can directly see faces and hear voices with each other. In Wireless DAN, computers independently recognize the network configuration around them. Although present computer networks assume permanence and all computers must be registered before the system can be used, Wireless DAN enables a computer to communicate with computers nearby at any place the user goes without any complicated procedures or dedicated network equipment such as a base station. We also present a new support system for face-to-face cooperative work, which makes it possible to join or leave a meeting while it is in progress. We have developed an experimental system of Wireless DAN and the support system, which is comprised of notebook personal computers with an existing peer-to-peer infrared LAN card. The support system has been implemented using the functions provided by Wireless DAN and thus the system has been simplified. With the experimental system, we confirmed that Wireless DAN operates effectively for face-to-face communication.

This paper proposes a new multistage switch architecture for large-scale multicast ATM switching. The proposed architecture uses routing schemes both at the connection level and the cell level. This results in the reduction of the memory capacity required in the trunk number translators of the copy network modules. If Connection Splitting algorithm is used, a nonblocking switch can be constructed under the same nonblocking condition as that of point-to-point Clos network. It is shown that the memory requirements in the new switching network are less than the previous architectures.

This paper proposes a masked data transferring method for the write-back controlled disk cache system employing a fixed-length recording disk drive, enabling data transfer of discontinuous sectors on the same track between the cache and the disk. This paper also evaluates the method. In write-back controlled disk cache sytems, random write requests cause dirty data (write-pending data on a cache) on discontinuous areas on the cache. It is likely that several sectors on the same track become dirty. These dirty sectors must be written onto the disk according to the cache management scheme. In conventional data transferring methods between a disk cache and a disk drive, plural sectors can be transferred in one single operation when the sectors are adjacent, but discrete sectors must be transferred by individual operations. In the methods, an address of the head sector and number of sectors to be transferred are given to the transfer unit. For example, when two sectors on the same track are located closely but not adjacently, and data transfer is requested for those two sectors, the transfer operation for the second sector must be prepared after the first transfer had completed and before the second sector arrives under the disk head. Although the time for the head to pass by the uninterested sector is often too short for the software overhead for the first transfer to be completed and the second transfer to be prepared, which leads to an unwanted extra rotation of the disk. With the masked transferring method proposed in this paper, the micro program creates a bit-map specifying the target sectors to be transferred and passes it to the data transfer unit, enabling to transfer the discontinuous sectors without latency. The method was evaluated using OLTP warkloads. Results show an improvement in random I/O throughput of between 8% and 27%. The masked transferring method is adopted in Hitachi's A-6521 disk subsytems, shipped since December 1993.

A new algorithm for multiple-precision modular multiplication is proposed. It is fast and uses a small amount of main memory, and hence, is useful for application of a public-key cryptosystem to small computers, such as card computers.

In this work, a statistical analysis is performed for a simple constrained high-order Yule-Walker (YW) tone frequency estimator obtained from the first equation of the constrained high-order YW equations. Explicit expressions for its estimation bias and variance are efficiently derived by virtue of a Taylor series expansion technique. Especially, being explicit in terms of frequency, data length and Signal-to-Noise Ratio (SNR) value, the resulting bias expression can not be obtained by using the asymptotic analyses used for the parameter estimation methods. The obtained expressions are compared with their counterparts of the Pisarenko tone frequency estimator. Simulations are performed to support the theoretical results.

Empirical studies confirm that the received radio signals in certain cellular systems are well modelled by Nakagami statistics. Therefore, performing relevant systems studies can be potentially useful to a system designer. A very useful statistical measure for characterizing the performance of a mobile radio system is the probability of outage, which describes the fraction of time that the signal-to-interference ratio (SIR) drops below some threshold. A more refined criterion for the outage is the failure to simultaneously obtain a sufficient SIR and a minimum power level for the desired signal. Thus, we derive new expressions for the probability of outage where a mobile unit receives a Nakagami desired signal and multiple, independent, cochannel Nakagami interferers. A salient feature of our results is that, unlike some previous studies, the outage expressions do not restrict the Nakagami fading parameter, m, to strictly integer values. Furthermore, since the received signals in mobile radio also experience log-normal shadowing, we analyze the case where the received signals are modelled by a composite of Nakagami and log-normal distributions. Outage probabilities are computed and graphically presented for several cases. The effect of specifying a minimum signal requirement for adequate reception is found to introduce a floor on the outage probability. It is also found that shadowing in macrocellular systems severely degrades the desired quality of service by increasing the reuse distance necessary for a given outage level.

This paper describes an MOS charge-mode version of a T-Model neural-based PCM encoder. The neural-based PCM encoding networks are designed, simulated and implemented using MOS charge-mode circuits. Simulation results are given for both the T-Model and the Hopfield model CMOS charge-mode PCM encoders, and demonstrate the T-Model neural-based one performs the PCM encoding perfectly, while the Hopfield one fails to.