In this paper, we propose a new denoising algorithm based on the dyadic wavelet transform (DWT) for ECG signals corrupted with different types of synthesized noise. Using the property that DWT is overcomplete, we define some convex sets in the set of wavelet coefficients and give an iterative method of the projection on the convex sets. The results show that the noises are not only removed from ECG signals, but also the ECG signals are reconstructed, which is used in detecting QRS complex. The performance of the proposed algorithm is demonstrated by some experiments in comparison with the conventional methods.

Topological Walk is an algorithm that can sweep an arrangement of n lines in O(n2) time and O(n) space. This paper revisits Topological Walk to give its new interpretation in contrast with Topological Sweep. We also survey applications of Topological Walk to make the distinction clearer.

We have developed a low bit-rate video coding using a video digital signal processor (DSP) called VDSP1χ, which performs real-time encoding and decoding for discrete cosine transform-(DCT-) based algorithms, such as ITU-T H. 261, H. 263 and wavelet-based subband encoding algorithms. This LSI features a processing unit which implements wavelet filters at high speeds, a compact DCT circuit, and a fast, flexible DRAM interface for low-cost systems. This system is capable of processing quarter common intermediate format (QCIF)(176144 pixels) size pictures at a rate greater than 15 frames/s.

We introduce the concepts of conservative cell loss ratio (CLR) estimation and worst case cell arrival patterns, and apply them to cell arrival patterns that conform to the generic cell rate algorithm (GCRA). We define new sets of cell arrival patterns which contain the worst case cell arrival patterns for conforming cell arrival patterns. Based on these sets, we propose an upper bound formula using the burst tolerance as well as peak cell rate and sustainable cell rate, and develope a connection admission control method that guarantees cell loss ratio performance satisfying its objective.

This paper describes two attacks against blind decryption (decode) based on the commutative random-self reducibility and RSA systems utilizing the transformability of digital signatures proposed in [2]. The transformable digital signature was introduced in [2],[8] for defeating an oracle attack, where the decrypter could be abused as an oracle to release useful information for an attacker acting as a requester of blind decryption. It was believed in [2],[8] that the correctness of a query to an oracle was ensured by the transformable signature derived from an original signature issued by the decrypter in advance, and a malicious query to an oracle could be detected before the blind decryption by the decrypter or would lead to release no useful information to an attacker. The first attack can decrypt all encrypted data with one access to an oracle. The second one generates a valid signature for an arbitrary message selected by an attacker abusing the validation check procedure.

This paper deals with overload control during congestion in a shared buffer ATM switch via selective cell discard and buffer management. Specifically, we consider the question of efficiency in buffer control schemes in order to reduce the number of cells that have to be discarded during congestion, in the meantime provide "fair" access to the shared buffer by all users. To prevent performance degradation of the shared buffer switch under imbalance traffic conditions, a "gated" buffer control scheme is proposed. The concept of the "gated" control policy is that we add a control gate in front of the corresponding logical queue of each overloaded output port. Some incoming cells destined for the overloaded ports can be blocked before entering the shared buffer. It will make rooms in the shared buffer for those incoming cells destined for the non-overloaded ports. This gated buffer control scheme can be modeled as a variation of SMXQ (sharing with a maximum queue length) scheme with a set of dynamically adjusted queue length thresholds. The simulation study of the gated buffer control is applied to a shared buffer ATM switch under various cell discard mechanisms. In most cases the proposed gated buffer control scheme can not only reduce the overall cell loss but also satisfy the "fair" access requirement under network congestion conditions, if we adjust the dynamical queue length thresholds properly.

In many applications, information retrieval is a very important research field. In several key strategies, the binary trie is famous as a fast access method able to retrieve keys in order. Especially, a Patricia trie gives the shallowest trie by eliminating all nodes which have only one arc, and it requires the smallest storage among the other trie structures. If trie structures are implemented, however, the greater the number of the registered keys, the larger storage is required. In order to solve this problem, Jonge et al. proposed a method to change the normal binary trie into a compact bit stream. This paper proposes the improved trie representation for the Patricia trie, as well as the methods for searching and inserting the key on it. The theoretical and experimental results, using 50,000 Japanese nouns and 50,000 English words, show that this method generates 25-39 percent shorter bit streams than the traditional method. This method, thus, enables us to provide more compact storage and faster access than the traditional method.

This letter derives the theoretical lower bound on image correlation coefficient that judges the extent of image degradation. It is shown that the correlation coefficient depends on phase-error variance in antenna aperture domain. Thereby, one can predict image quality before image formation. The theoretical bound is verified by experimental data where the dominant scatterer algorithm (DSA) is used for phase synchronization.

Despite the fact that the necessity of FDT (formal description techniques) had been emphasized, many approaches to verifying feature interactions were insufficient. This paper proposes formal definitions for feature interactions which can occur when telecommunication services specified independently are operated in parallel. Seven types of feature interactions are defined formally.

In this paper we derive the expressions of the spectra of waveform relaxation operators for linear differential-algebraic equations which stem from circuit simulation. These expressions suggest ways to split the matrices of the circuit equations such that waveform relaxation will converge. Numerical experimental results are given.

A maximal-ratio-combining (MRC) digital beamformer has been studied to attain open-loop and automatic self-beam steering towards both desired and multipath signals at the same time and diversity combining of the signals, which are made possible by spatial digital signal processing. This paper describes the performance of this beamformer under the multipath signal arrivals with various path delays using numerical simulation, aimed at application to future mobile radios with high spatial utilization efficiency. The results indicate the robustness of the MRC beamformer in a multipath environment. It features multidirectional beam steering when there is small path delay in the multipath signal and gain suppression in multipath signals when their path delay is more than about one symbol. Moreover, improvement in suppression by employing low-sidelobe amplitude distribution is discussed as a means to reduce inter-symbol interference without null-beam steering.

In a distributed concurrent system such as a computer communication network, the system components communicate with each other via communication links in order to accomplish a desired distributed application. If the links are dynamically established among the components, the system configuration as well as its behavior becomes complex. In this paper, we give formal specification of such a dynamically reconfigurable system in which the components are modeled by communicating finite state machines executed concurrently with the communication links which are dynamically established and disconnected. We also present an algorithm to validate the safety and link-related properties in the specified behavior. Finally, we design and implement a simulator and a validator that enables execution and validation of the given specification, respectively.

The structure of bidirectional syndrome decoding for binary rate (n-1)/n convolutional codes is investigated. It is shown that for backward decoding based on the trellis of a syndrome former HT, the syndrome sequence must be generated in time-reversed order using an extra syndrome former H*T, where H* is a generator matrix of the reciprocal dual code of the original code. It is also shown that if the syndrome bits are generated once and only once using HT and H*T, then the corresponding two error sequences have the intersection of n error symbols, where is the memory length of HT.

A training data selection method is proposed for multilayer neural networks (MLNNs). This method selects a small number of the training data, which guarantee both generalization and fast training of the MLNNs applied to pattern classification. The generalization will be satisfied using the data locate close to the boundary of the pattern classes. However, if these data are only used in the training, convergence is slow. This phenomenon is analyzed in this paper. Therefore, in the proposed method, the MLNN is first trained using some number of the data, which are randomly selected (Step 1). The data, for which the output error is relatively large, are selected. Furthermore, they are paired with the nearest data belong to the different class. The newly selected data are further paired with the nearest data. Finally, pairs of the data, which locate close to the boundary, can be found. Using these pairs of the data, the MLNNs are further trained (Step 2). Since, there are some variations to combine Steps 1 and 2, the proposed method can be applied to both off-line and on-line training. The proposed method can reduce the number of the training data, at the same time, can hasten the training. Usefulness is confirmed through computer simulation.

Time series prediction is very important technology in a wide variety of fields. The actual time series contains both linear and nonlinear properties. The amplitude of the time series to be predicted is usually continuous value. For these reasons, we combine nonlinear and linear predictors in a cascade form. The nonlinear prediction problem is reduced to a pattern classification. A set of the past samples x(n-1),. . . ,x(n-N) is transformed into the output, which is the prediction of the next coming sample x(n). So, we employ a multi-layer neural network with a sigmoidal hidden layer and a single linear output neuron for the nonlinear prediction. It is called a Nonlinear Sub-Predictor (NSP). The NSP is trained by the supervised learning algorithm using the sample x(n) as a target. However, it is rather difficult to generate the continuous amplitude and to predict linear property. So, we employ a linear predictor after the NSP. An FIR filter is used for this purpose, which is called a Linear Sub-Predictor (LSP). The LSP is trained by the supervised learning algorithm using also x(n) as a target. In order to estimate the minimum size of the proposed predictor, we analyze the nonlinearity of the time series of interest. The prediction is equal to mapping a set of past samples to the next coming sample. The multi-layer neural network is good for this kind of pattern mapping. Still, difficult mappings may exist when several sets of very similar patterns are mapped onto very different samples. The degree of difficulty of the mapping is closely related to the nonlinearity. The necessary number of the past samples used for prediction is determined by this nonlinearity. The difficult mapping requires a large number of the past samples. Computer simulations using the sunspot data and the artificially generated discrete amplitude data have demonstrated the efficiency of the proposed predictor and the nonlinearity analysis.

An acoustic diagnosis technique for the blower by wavelet transform and neural networks is described. It is important for this diagnosis to detect surging phenomena, which lead to the destruction of the blower. Dyadic wavelet transform is used as the pre-processing method. A multi-layered neural network is used as the discrimination method. Experiment is performed for a blower. The results show that the neural network with wavelet transform can detect surging sound well.

As MOSFET sizes and wire widths become very small in recent years, influence of resistive component of interconnects on the estimation of propagation delay and power dissipation can no longer be neglected. In this paper we present formulas of output waveform at driving point and short-circuit power dissipation for static CMOS logic gates driving a CRC π load. By representing the short-circuit current and the current flowing in the resistance of a CRC π load by piece-wise linear functions, a closed-form formula is derived. On the gate delay the error of our formula is less than 8% from SPICE in our experiments. These formulas will contribute to faster estimation of circuit speed and power dissipation of VLSI chips on timing level simulators.

The tape deformation due to such particles as wear debris and dust in the head/tape contact region is one of the main causes of the signal quality deterioration in magnetic tape devices. Thus it is significant to make clear the tape deformation due to a particle for realizing higher recording densities. This paper investigates the tape deformation profile generated by a particle through an interferometric experiment and a simulation using a point loaded tape model. A rather good agreement between them is obtained, thereby the simulation is verified appropriate to estimate the tape deformation due to a particle. This paper also describes the relationship between the spacing loss and the particle height, considering the tape deformation profile obtained from the simulation. In addition, the influence of the particle height on the width of the tape deformed area is estimated, which can make a basis of the design of error correction code.

As expert system technology gains wider acceptance in digital system design, the need to build and maintain a large scale knowledge base will assume greater importance. However, how to build a correct and efficient rule base is even a hard part in the knowledge-based system development. In this paper, we develop FARHDL (Frame-And-Rule-based Hardware Description Language) to form a knowledge base. The FARHDL is simple but powerful to specify the hardware requirements and can be directly simulated by PROLOG. Through the knowledge base transformed from FARHDL, a formal method can be developed to design, implement, and validate the digital hardware systems. Furthermore, behavioral properties, anomaly properties, structural properties, and timing properties are applied to analyze the requirements specification. The purposes of those properties are used to detect explicit/implicit incorrect specification clauses and to capture some desired requirements, such as completeness and consistency. Finally, the analysis results can be a useful tool for finding obscure problems in tricky digital system designs and can also aid in the development of formal specifications.

Geometric region method is one of the techniques to handle real-time systems which have potentially infinite state spaces. However, the original geometric region method gives incorrect results for the CTL model checking of time Petri nets. In this paper, we discuss the sufficient condition for the geometric region graphs to be correct with respect to the CTL model checking of time Petri nets, and then propose a technique to partition given geometric regions so that the graphs satisfy the sufficient condition. Finally, we implement the proposed algorithm, and compare it with the other methods by using small examples.