This paper introduces a new class of machines called multihead marker finite automata, and investigates how the number of markers affects its accepting power. Let HM{0}(i, j)(NHM{0}(i, j))denote the class of languages over a one-letter alphabet accepted by two-way deterministic (nondeterminstic) i-head finite automata with j markers. We show that HM{0} (i, j) HM{0}(i, j1) and NHM{0}(i, j) NHM{0}(i, j+1) for each i2, j0.

Recently, many researches on associative memories have been made a lot of neural network models have been proposed. Bidirectional Associative Memory (BAM) is one of them. The BAM uses Hebbian learning. However, unless the traning vectors are orthogonal, Hebbian learning does not guarantee the recall of all training pairs. Namely, the BAM which is trained by Hebbian learning suffers from low memory capacity. To improve the storage capacity of the BAM, Pseudo-Relaxation Learning Algorithm for BAM (PRLAB) has been proposed. However, PRLAB needs long learning epochs because of random initial weights. In this paper, we propose Quick Learning for BAM which greatly reduces learning epochs and guarantees the recall of all training pairs. In the proposed algorithm, the BAM is trained by Hebbian learning in the first stage and then trained by PRLAB. Owing to the use of Hebbian learning in the first stage, the weights are much closer to the solution space than the initial weights chosen randomly. As a result, the proposed algorithm can reduce the learning epocks. The features of the proposed algorithm are: 1) It requires much less learning epochs. 2) It guarantees the recall of all training pairs. 3) It is robust for noisy inputs. 4) The memory capacity is much larger than conventional BAM. In addition, we made clear several important chracteristics of the conventional and the proposed algorithms such as noise reduction characteristics, storage capacity and the finding of an index which relates to the noise reduction.

This report describes 4-2 compressors composed of Complementary Pass-Transistor Logic (CPL). We will show that circuit designs of the 4-2 compressors can be optimized for high speed and small size using only exclusive-OR's and multiplexers. According to a circuit simulation with 0.8µm CMOS device parameters, the maximum propagation delay and the average power consumption per unit adder are 1.32 ns and 11.6 pJ, respectively.

In this paper we analize the performance of Trellis Coded Modulation (TCM) schemes with coherent detection operating in a frequency flat, mobile Rayleigh fading environment, and with different knowledge levels on both the amplitude and phase fading processes (the latter is not assumed as usual to be ideally tracked), or Channel State Information (CSI). For example, whereas ideal CSI means that both the amplitude and phase fading characteristics are perfectly known by the receiver, other situations that are treated consider perfect knowledge of the amplitude (or phase) with complete disregard of the phase (or amplitude), as well as non concern on any of them. Since these are extreme cases, intermediate situations can be also defined to get extended bounds based on Chernoff which allow the phase errors, in either form of constant phase shifts or randomly distributed phase jitter, to be included in the upper bounds attainable by transfer function methods, and are applicable to multiphase/level signaling schemes. We found that when both fading characteristics are considered, the availability of CSI enhances significatively the performance. Furthermore, for non constant envelope schemes with non ideal CSI and for constant envelope schemes with phase errors, an asymmetry property of the pairwise error probability is identified. Theoretical and simulation results are shown in support of the analysis.

This paper describes a 32-bit fully asynchronous microprocessor, with 4-stage pipeline based on a RISC-like architecture. Issues relevant to the processor such as design of self-timed datapath, asynchronous controller and interconnection circuits are discussed. Simulation results are included using parameters extracted from layout, which showed about the 300 MIPS processing speed and used 71,000 transistors with 0.5 µm CMOS technology.

We propose a large capacity broadband packet switch architecture using multiple optical star couplers and tunable devices whose tuning range is restricted. The proposed switch has the conventional three-stage switch structure. With the use of the generalized knockout principle and tunable lasers arranged in an appropriate manner, the switch becomes an output queueing system that yields the best possible delay/throughput performance. This switch requires minimal hardware at the cost of the increased number of wavelengths.

A new regularization cost function for generalization in real-valued function learning is proposed. This cost function is derived from the maximum likelihood method using a modified sample distribution, and consists of a sum of square errors and a stabilizer which is a function of integrated square derivatives. Each of the regularization parameters which gives the minimum estimation error can be obtained uniquely and non-empirically. The parameters are not constants and change in value during learning. Numerical simulation shows that this cost function predicts the true error accurately and is effective in neural network learning.

We propose a fast computation method of stochastic relaxation for the continuous-valued Markov random field (MRF) whose energy function is represented in the quadratic form. In the case of regularization in visual information processing, the probability density function of a state transition can be transformed to a Gaussian function, therefore, the probablistic state transition is realized with Gaussian random numbers whose mean value and variance are calculated based on the condition of the input data and the neighborhood. Early visual information processing can be represented with a coupled MRF model which consists of continuity and discontinuity processes. Each of the continuity or discontinuity processes represents a visual property, which is like an intensity pattern, or a discontinuity of the continuity process. Since most of the energy function for early visual information processing can be represented by the quadratic form in the continuity process, the probability density of local computation variables in the continuity process is equivalent to the Gaussian function. If we use this characteristic, it is not necessary for the discrimination function computation to calculate the summation of the probabilities corresponding to all possible states, therefore, the computation load for the state transition is drastically decreased. Furthermore, if the continuous-valued discontinuity process is introduced, the MRF model can directly represent the strength of discontinuity. Moreover, the discrimination function of this energy function in the discontinuity process, which is linear, can also be calculated without probability summation. In this paper, a fast method for calculating the state transition probability for the continuous-valued MRF on the visual informtion processing is theoretically explained. Next, initial condition dependency, computation time and dependency on the statistical estimation of the condition are investigated in comparison with conventional methods using the examples of the data restoration for a corrupted square wave and a corrupted one-dimensional slice of a natural image.

A network system is proposed for segmenting and extracting multiple moving objects in 2D images. The system uses an interconnected neural network in which grouping factors, such as edge proximity, smoothness of edge orientatio, and smoothness of velocity perpendicular to an edge, are embedded. The system groups edges so that the network energy may be minimized, i.e. edges may be organized into perceptually plausible configuration. Experimantal results are provided to indicate the performance and noise robustness of the system in extracting objects in synthetic images.

The number of nodes in a hidden layer of a feed-forward layered network reflects an optimality condition of the network in coding a function. It also affects the computation time and the ability of the network to generalize. When an arbitrary number of hidden nodes is used in designing the network, redundancy of hidden nodes often can be seen. In this paper, a method of reducing hidden nodes is proposed on the condition that a reduced network maintains the performances of the original network within an accepted level of tolerance. This method can be applied to estimate the performances of a network with fewer hidden nodes. The estimated performances indicate the lower bounds of the actual performances of the network. Experiments were performed using the Fisher's IRIS data, a set of SONAR data, and the XOR data for classification. The results suggest that sufficient number of hidden nodes, fewer than the original number, can be estimated by the proposed method.

In this paper, a middle-mapping learning algorithm for cellular associative memories is presented. This algorithm makes full use of the properties of the cellular neural network so that the associative memory has some advantages compared with the memory designed by the ourter product method. It can guarantee each prototype is stored at an equilibrium point. In the practical implementation, it is easy to build up the circuit because the weight matrix presenting the connection between cells is not symmetric. The synchronous updating rule makes its associative speed very fast compared to the Hopfield associative memory.

Graph parameters such as connectivity and diameter have been studied extensively due to their intrinsic importance in graph theory, combinatorics and their relations to (and applications in) fault tolerance and transmission delay in communications networks. The advent of VLSI technology and fiber optics material science has enabled us to design massively parallel processing computer systems and fast and complicated communications networks. All these systems increase their reliability by studying (among other) the existence of two (or more) disjoint paths connecting any two nodes. This paper addresses these issues by studying the width and length of containers in graphs and networks. In particular, the notions of w-distance and w-diameter on a graph are defined and studied which generalize both concepts of connectivity and diameter. Thses notions are also considered in finite groups. Other closely related parameters will be explored in the contexts of fault tolerance and routing. Known results are surveyed and open problems are offered for further investigation.

A method to improve the efficiency of dynamic fault imaging (DFI) by fully utilizing the CAD data in the CAD-linked electron beam test system is proposed. In the method, in order to shorten the long acquisition time of the stroboscopic voltage contrast images over the whole area of the chip during the entire test cycle, only the area and phase (time) required for fault tracing are selected by utilizing the CAD data. Furthermore, image processing techniques are combined with the method to improve the efficiency of the DFI. In particular, the signal averaging technique is used in order to improve the signal-to-noise ratio in the stroboscopic images where all voltage information data on the equipotential electrode recognized by the CAD layout data are averaged. This enables us to reduce the acquisition time of images. Moreover, the experimental system is set up so that the image processing can be performed in parallel with the acquisition of the stroboscopic images. The proposed method is applied to part of a 2k-transistor block of a nonpassivated CMOS LSI where a marginal fault is detected. The result shows that the method is an efficient approach to the fully automatic fault diagnosis in the CAD-linked electron beam test system. The proposed method could improve the efficiency of the conventional DFI by a factor of more than 1000.

Resonant slots are widely used for conventional slotted waveguide array. Reflection from each slot causes a standing wave in the waveguide and beam tilting technique is essential to suppress the reflection at the antenna input port. But the slot reflection narrows the overall frequency bandwidth and the design taking it into account is complicated. This paper proposes a reflection cancelling slot pair as an array element, which consists of two slots spaced by 1/4λg. Round trip path-length difference between them is 1/2λg and reflection waves from a pair disappear and traveling-wave excitation in the waveguide is realized. The full wave analysis reveals that mutual coupling between paired slots is large and seriously reduces the radiation from a pair. Offset arrangement of slots in a pair is recommended to decrease the mutual coupling and to realize strong coupling. In practical array design, the mutual couplings from other pairs were simulated by imposing periodic boundary conditions above the aperture. To clarify the advantages of the slot pair over a conventional resonant slot, the predicted characteristics are compared. Reflection characteristics of the array using the slot pair is excellent and a boresite beam array can be realized. In addition, a slot pair can realize stronger coupling than the conventional resonant slot, while the bandwidth of the former in terms of the aperture field phase illumination is narrower than that of the latter. These suggests that the slot pair array is much more suitable for a small array than conventional one. Finally, the predicted characteristics are confirmed by experiments.

This article deals with the binary neural network with negative self-feedback connections as a method for solving combinational optimization problems. Although the binary neural network has a high convergence speed, it hardly searches out the optimum solution, because the neuron is selected randomly at each state update. In thie article, an improvement using the negative self-feedback is proposed. First it is shown that the negative self-feedback can make some local minimums be unstable. Second a selection rule is proposed and its property is analyzed in detail. In the binary neural network with negative self-feedback, this selection rule is effective to escape a local minimum. In order to comfirm the effectiveness of this selection rule, some computer simulations are carried out for the N-Queens problem. For N=256, the network is not caught in any local minimum and provides the optimum solution within 2654 steps (about 10 minutes).

A novel pulse neural network model for sound localization has been proposed. Our model is based on the physiological auditory nervous system. Human beings can perceive the sound direction using inter-aural time difference (ILD) and inter-aural level difference (ILD) of two sounds. The model extracts these features using only pulse train information. The model is divided roughly into three sections: preprocessing for input signals; transforming continuous signals to pulse trains; and extracting features. The last section consists of two parts: ITD extractor and ILD extractor. Both extractors are implemented using a pulse neuron model. They have the same network structure, differing only in terms of parameters and arrangements of the pulse neuron model. The pulse neuron model receives pulse trains and outputs a pulse train. Because the pulses have only simple informations, their data structures are very simple and clear. Thus, a strict design is not required for the implementation of the model. These advantages are profitable for realizing this model by hardware. A computer simulation has demonstrated that time and level differences between two signals have been successfully extracted by the model.

Results of a DC motor driving test with a power sent by a microwave and extracted with a rectenna array are reported. No significant difference has been observed in the output DC power from the rectenna array between a motor load and a resistive load. Mechanical output could be extracted from the received microwave power with an efficiency of 26%.

In this paper, we show that the neural network can approximate the chaotic behavior in nonlinear dynamical system by experimental study. Chaotic neural activities have been reported in many respects including neural network field. On the contrary, can the neural network learn the chaotic behavior? There have been explored the neural network architecture for predicting successive elements of a sequence. Also there have been several studies related to learning algorithms for general recurrent neural networks. But they often require complicated procedure in time calculation. We use simple standard backpropagation for a kind of simple recurrent neural network. Two types of chaotic system, differential equation and difference equation, are examined to compare characteristics. In the experiments, Lorenz equation is used as an example of differential equation. One-dimensional logistic equation and Henon equation are used as examples of difference equation. As a result, we show the approximation ability of chaotic dynamics in difference equation, which is logistic equation and Henon equation, by neural network. To indicate the chaotic state, we use Lyapunov exponent which represents chaotic activity.

The mechanism for data retention failure of EPROM has been investigated by the Optical Beam Induced Current(OBIC) technique. It was found that the data of failure cells were changed from '1' to '0' during read-mode by laser irradiation by OBIC. The data in good cells was not changed. This result suggests the effective barrier height between Si and SiO2 is being lowered. In addition, the cross section technique revealed that gate electrode and gate oxide were exposed due to lack of dielectric layers. This defect seemed to be the cause of the barrier height lowering. The OBIC technique not only gives the failure location but a detailed information of the failure mechanism. We found that OBIC technique is a very powerful tool for the analysis of EPROM failure mechanisms. The usefulness of the Emission Micro Scope (EMS) technique is also discussed.

The origin of and a method of enhancing the Optical Beam Induced Resistance Change (OBIRCH) signal for defect observation in VLSI metal interconnections is discussed based on a numerical analysis of three-dimensional thermal conduction and experimental results. The numerical analysis shows that the OBIRCH signal originates from a slight increase in the resistance of the metal line caused by laser beam heating and that its effect is influenced by the temperature of the metal layer. Both simulations and experimental results suggest that cooling the sample is preferable to detect the OBIRCH signal. The decrease in the total resistance of the metal line without any change in the amount of the resistance increase under laser illumination is found to be the main cause of the OBIRCH signal enhancement under low temperature measurement.