As VLSI technology has developed, the interest in implementing an entire or significant part of a parallel computer system using wafer scale integration is growing. The major problem for the case is the possibility of drastically low yield and/or reliability of the system if there is no strategy for coping with such situations. Various strategies to restructure the faulty physical system into the fault-free target logical system are described in the literature [1]-[5]. In this paper, we propose an efficient approximate method which can reconstruct the 1 1/2 track-switch mesh arrays with faulty PEs using hardware as well as software. A logical circuit added to each PE and a network connecting the circuits are used to decide spare PEs which compensate for faulty PEs. The hardware compexity of each circuit is much less than that of a PE where the size of each additional circuit is independent of array sizes and constant. By using the exclusive hardware scheme, a built-in self-reconfigurable system without using a host computer is realizable and the time for reconfiguring arrays becomes very short. The simulation result of the performance of the method shows that the reconstructing efficiency of our algorithm is a little less than those of the exaustive and Shigei's ones [6] and [7], but much better than that of the neural one [3]. We also compare the time complexities of reconstructions by hardware as well as software, and the hardware complexity in terms of the number of gates in the logical circuit added to each PE among the other methods.

In this paper, we will present a study of the time-dependence effect in alumite perpendicular media at different thicknesses. Important parameters of the time-dependence effect such as magnetic viscosity and activation volume are investigated. Viscosity as a function of applied field (viscosity curve) exhibits a short plateau at a low field and then decreases monotonously with increasing field. After correcting for the demagnetizing field, the shape of the intrinsic viscosity curves changes to the well-known shape of the viscosity curve of in-plane media, i. e. , they have a peak near Hc. The intrinsic viscosity curves obtained from the experiments were fitted to an analytical model by Chantrell et al., from which, we found that the effective switching volumes obtained by fitting are much smaller than the column volumes, indicating that the reversal mechanism is incoherent.

Non-stationary glint noise is often observed in a radar tracking system. The distribution of glint noise is non-Gaussian and heavy-tailed. Conventional recursive identification algorithms use the stochastic approximation (SA) method. However, the SA method converges slowly and is invalid for non-stationary noise. This paper proposes an adaptive algorithm, which uses the stochastic gradient descent (SGD) method, to overcome these problems. The SGD method retains the simple structure of the SA method and is suitable for real-world implementation. Convergence behavior of the SGD method is analyzed and closed-form expressions for sufficient step size bounds are derived. Since noise data are usually not available in practice, we then propose a noise extraction scheme. Combining the SGD method, we can perform on-line adaptive noise identification directly from radar measurements. Simulation results show that the performance of the SGD method is comparable to that of the maximum-likelihood (ML) method. Also, the noise extraction scheme is effective that the identification results from the radar measurements are close to those from pure glint noise data.

An accurate and efficient numerical solution is presented for a two-dimensional electromagnetic wave scattering from a multilayered resistive strip grating embedded in a dielectric slab. Both E- and H-waves are treated. The problem is formulated into a set of integral equations, which is solved by the moment method accompanied by a regularization procedure. The resultant set of linear algebraic equations has the form of the Fredholm second kind, and therefore yields stable and accurate numerical solutions. The power distribution is computed for several grating parameters. Attention is paid to seek a set of parameters that maximizes absorption in the strips. The low frequency approximate formulas are also derived. This analysis would be useful in designing electromagnetic wave absorbers.

In this paper, an effective method of system modeling and dynamic scheduling to improve operation and control for the Back-End process of semiconductor manufacturing is developed by using Colored Timed Petri-Nets (CTPNs). The simulator of a CTPNs model was utilized to generate a new heuristic scheduling method with genetic algorithm(GA) which enables us to obtain the optimal values of the weighted delay time and standard deviation of lead time.

In this paper we present a Quality of Service (QoS) management architecture for distributed multimedia applications in heterogeneous communication environments of wired and wireless networks. Gaps in network performance such as bandwidths and error rates between wired and wireless networks, as well as gaps in terminal performance in media handling between desktop computers and handheld computers, bring about heterogeneities. Furthermore, even performance gaps among various desktop computers cause heterogeneities. As a result of these heterogeneities in network and terminal performances and various user preferences, the QoS requirement from each receiver is different. Therefore, mechanisms that adjust and satisfy each QoS requirement are needed. We propose a proxy server called Communication Coordination Server (CCS), which intermediates a video server and a receiver and manages the QoS coordination. The CCS performs QoS admission, adjustment, and allocation mechanisms to satisfy the user's QoS requirement. Then transcoding is used to realize the allocated QoS, and it decodes the input video stream from the video server and encodes it within the CCS. A QoS mapping mechanism that translates application-level QoS into resource-level QoS is needed for the QoS admission. We also propose a new QoS mapping mechanism using spline functions that enables a continuous QoS translation. We have built a CCS prototype in our laboratory testbed, and have verified that the CCS can resolve the heterogeneities between the server and receiver by the QoS adjustment mechanism of the transcoding and the QoS admission.

It is known that the clock-period can be shorter than the maximum of signal-delays between registers if the clock arrival time to each register is properly scheduled. The algorithm to design an optimal clock-schedule was given. In this paper, we propose a clock-tree routing algorithm that realizes a given clock-schedule using the Elmore-delay model. Following the deferred-merge-embedding (DME) framework, the algorithm generates a topology of the clock-tree and simultaneously determines the locations and sizes of intermediate buffers. The experimental results showed that this method constructs a clock-tree with moderate wire length for a random layout of scheduled registers. Notably, the required wire length for a gentle layout of scheduled registers was shown to be almost equal to that of zero-skew clock-trees.

A procedure for radar target discrimination is presented in this paper. The scheme includes an enhancement of late-time noisy scattering data based on a proposed signal processing algorithm and a decision procedure using previously known resonance annihilation filters. The signal processing stage is specifically adapted to scattering signals and makes use of the results of the singularity expansion method. It is based on a signal reconstruction using the SVD of a data matrix with a suitable choice of the number of singular vectors employed. To justify the inclusion of this stage, this procedure is shown to maintain the signal characteristics necessary to identify the scattered response. Simulation results clearly reveal a significant improvement due to the inclusion of the proposed stage. This improvement becomes especially important when the noise level is high or the targets to be discriminated (five regular polygonal loops) have a similar geometry.

An approximate formula is proposed for the equivalent susceptance of a circular iris in a parallel plate waveguide when the TEM mode cylindrical wave is incident from the center of the iris. Schwinger's variational method for a linear iris is generalized to the cylindrical case, and an approximate closed form formula is obtained which recovers the result of the linear iris when the radius of the circular iris is sufficiently larger than the wavelength. For verification of the formula, an exact integral equation is formulated and solved numerically by Galerkin's method. A comparison between them shows good agreement.

This letter is concerned with a new algorithm which can be used to design a time-domain equalizer (TEQ) for xDSL systems employing the discrete multitone (DMT) modulation. The proposed algorithm, derived by neglecting the terms which do not affect the performance of a DMT system in ARMA modeling, is shown to have a good performance compared with the previous TEQ algorithms even with a significantly lower computational complexity. In addition, the proposed algorithm does not require the channel impulse response or training sequence, since all processing is made only with the received data.

The Matrix-Pencil (MP) method is applied to the estimation of the undesired radiation from the microstrip line discontinuities. The Q factors are obtained from the complex resonant frequencies estimated from FDTD transient field by using MP. The number of the damped oscillations is estimated by using MDL which is widely used as an information theoretic criterion for the model order estimation.

In this study, wave propagation phenomena of phase states are observed at van der Pol oscillators coupled by inductors as a ladder. For the case of 17 oscillators, interesting wave propagation phenomena of phase states are found. By using the relationship between phase states and oscillation frequencies, the mechanisms of the propagation and the reflection of wave are explained. Circuit experimental results agree well with computer calculated results qualitatively.

This paper presents an algorithm for the scan-chain optimization problem in multiple-scan design methodology. The proposed algorithm, which consists of four phases, first determines pairs of scan-in and scan-out pins (Phase 1), and then assigns flip-flops to scan-paths by using a graph theoretical method (Phase 2). Next the algorithm decides connection-order of flip-flops in each scan-path by using TSP (Traveling Salesman Problem) heuristics (Phase 3), and finally exchanges flip-flops among scan-paths in order to reduce total scan-path length (Phase 4). Experiments using actual design data show that, for ten scan-paths, our algorithm achieved a 90% reduction in scan-test time at the expense of a 7% total scan-path length increase as compared with the length of a single optimized scan-path. Also, our algorithm produced less total scan-path length than other three possible algorithms in a reasonable computing time.

In this paper, we present a traffic scheduling algorithm, called the Delay-Bound Monotonic with Average Rate Reservation (DM/ARR), which generates minimum output burstiness streams. We assume that connection i is policed by the leaky bucket algorithm with parameters (σi,ρi) where σi is the bucket size (or burstiness) and ρi is the leaky rate. Compared with the totally isolated scheme where connection i is allocated a bandwidth ri=max{σi/di,ρi} (di is the delay bound requirement of connection i), the DM/ARR algorithm has a better performance in the sense that it has a larger admission region. We prove that, among all possible scheduling algorithms that satisfy the delay bound requirements of established connections, DM/ARR results in the minimum output burstiness. This is important because a smaller burstiness implies a smoother traffic and thus the receiver (or next switch node in a multihop network) can handle it more easily. Numerical results show that the admission region of the DM/ARR algorithm is close to that of the earliest deadline first algorithm. A packetized version is studied for ATM networks.

This paper deals with two-processor scheduling for a class of program nets, that are acyclic and SWITCH-less, and of which each node has unity node firing time. Firstly, we introduce a hybrid priority list L* that generates optimal schedules for the nets whose AND-nodes possess at most single input edge. Then we extend L* to suit for general program nets to give a new priority list L**. Finally, we use genetic algorithm to do the performance evaluation for the schedules generated by L** and show these schedules are quite close to optimal ones.

A new feature parameter based on a discrete wavelet transform is proposed for word boundary detection of isolated utterances. The sum of standard deviation of wavelet coefficients in the third coarse and weighted first detailed scale is defined as a new feature parameter for endpoint detection. Experimental results demonstrate the superiority of the proposed feature to the conventional ones in capturing word boundaries even in noisy speech.

Newer off-chip DRAM families, including Synchronous DRAMs (SDRAMs) and RAMBUS DRAMs (RDRAMs), are becoming standard choices for the design of high-performance systems. Although previous work in High-Level Synthesis (HLS) has addressed exploiting features of page-mode DRAMs, techniques do not exist for exploiting the two key features of these newer DRAM families that boost memory performance and help overcome bandwidth limitations: (1) burst mode access, and (2) interleaved access through multiple banks. We address pre-synthesis optimizations on the input behavior that extract and exploit the burst mode and multiple bank interleaved access modes of these newer DRAM families, so that these features can be exploited fully during the HLS trajectory. Our experiments, run on a suite of memory-intensive benchmarks using a contemporary SDRAM library, demonstrate significant performance improvements of up to 62.5% over the naive approach, and improvements of up to 16.7% over the previous approach that considered only page-mode or extended-data-out (EDO) DRAMS.

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.

Human visual system can perceive 3-D structure of an object by binocular disparity, gradient of illumination (shading), occlusion, textures, perspective and so forth. Among them, binocular disparity seems to be the essentially important cues for the 3-D space perception and it is used widely for displaying 3-D visual circumstances such as in VR (virtual reality) system or 3-D TV. Visual illusions seem to be one of the phenomena which are purely reflecting the mechanism of human visual system. In the recent several years, the authors found several new types of 3-D visual illusions with binocular viewing. Entire 3-D illusory object including volume perception, transparency, dynamic illusions can be perceived only from the visual stimuli of disparity given by some inducing objects arranged with suitable relations. In this report, the authors introduced these newly found visual illusions and made some considerations on the human visual mechanism of 3-D perception and on their exploitation for new effective techniques in 3-D display. They introduced especially on the visual effect in two kinds of arrangement with occlusion and sustaining relationship between the illusory object and inducing objects. In the former case, the inducing objects which provide the stimuli were named as occlusion cues and classified into two types: contour occlusion cues and bulky occlusion cues. In the later case, those inducing objects were named as sustaining cues and a 3-D fully transparent illusory object was perceived. The perception was just like imagined from the scenes of the actions and positions of the pantomimists; then this phenomena was named as "Mime (Pantomime) Effect. " According to the positions of sustaining cues, they played different actions in this perception, and they are classified into three types: front sustaining cues, side sustaining cues and back sustaining cues. In addition, dynamic fusion and separation of volumetrical illusory objects were perceived when the visual stimuli were moving continuously between two structurally different conditions. Then the hysteresis was recognized in geometrical position between the fusion and separation. The authors believe that the occlusion cues and sustaining cues introduced in this paper could be effective clues for exploiting the new techniques for 3-D display.

This paper addresses the problem of learning Bayesian belief networks (BBN) based on the minimum description length (MDL) principle. First, we give a formula of description length based on which the MDL-based procedure learns a BBN. Secondly, we point out that the difference between the MDL-based and Cooper and Herskovits procedures is essentially in the priors rather than in the approaches (MDL and Bayesian), and recommend a class of priors from which the formula is obtained. Finally, we show as a merit of using the formula that a modified version of the Chow and Liu algorithm is obtained. The modified algorithm finds a set of trees rather than a spanning tree based on the MDL principle.