Experiments were carried out at several voltages to clarify the influence of the voltage on various characteristics, i.e. arc duration, contact resistance, arc energy, and the change in electrode mass. The voltage was varied from DC100 V to 160 V, the load current was fixed at 5 A constant, and the electromagnetic contactor was operated continuously up to 100,000 times. The experiments were carried out under the three operation modes which are classified by the arc discharge. As a result, the relation between the operation mode and contact resistance was clarified. When only a make arc was generated, the contact resistance was smallest. In addition, the contact resistance was not affected by the source voltage.

A facial components based facial expression recognition algorithm with sparse representation classifier is proposed. Sparse representation classifier is based on sparse representation and computed by L1-norm minimization problem on facial components. The features of “important” training samples are selected to represent test sample. Furthermore, fuzzy integral is utilized to fuse individual classifiers for facial components. Experiments for frontal views and partially occluded facial images show that this method is efficient and robust to partial occlusion on facial images.

This paper proposes the test case prioritization in regression testing. The large size of a test suite to be executed in regression testing often causes large amount of testing cost. It is important to reduce the size of test cases according to prioritized test sequence. In this paper, we apply the Markov chain Monte Carlo random testing (MCMC-RT) scheme, which is a promising approach to effectively generate test cases in the framework of random testing. To apply MCMC-RT to the test case prioritization, we consider the coverage-based distance and develop the algorithm of the MCMC-RT test case prioritization using the coverage-based distance. Furthermore, the MCMC-RT test case prioritization technique is consistently comparable to coverage-based adaptive random testing (ART) prioritization techniques and involves much less time cost.

In this paper, we propose a hardware solution for overcoming the problem of high computational demands in a nearest neighbor (NN) based multi-prototype learning system. The multiple prototypes are obtained by a high-speed K-means clustering algorithm utilizing a concept of software-hardware cooperation that takes advantage of the flexibility of the software and the efficiency of the hardware. The one nearest neighbor (1-NN) classifier is used to recognize an object by searching for the nearest Euclidean distance among the prototypes. The major deficiency in conventional implementations for both K-means and 1-NN is the high computational demand of the nearest neighbor searching. This deficiency is resolved by an FPGA-implemented coprocessor that is a VLSI circuit for searching the nearest Euclidean distance. The coprocessor requires 12.9% logic elements and 58% block memory bits of an Altera Stratix III E110 FPGA device. The hardware communicates with the software by a PCI Express (4) local-bus-compatible interface. We benchmark our learning system against the popular case of handwritten digit recognition in which abundant previous works for comparison are available. In the case of the MNIST database, we could attain the most efficient accuracy rate of 97.91% with 930 prototypes, the learning speed of 1.310-4 s/sample and the classification speed of 3.9410-8 s/character.

A low complexity log-likelihood ratio (LLR) calculation for high-order amplitude phase shift keying (APSK) signals is proposed. Using proper constellation partitioning together with a look-up table, the number of terms for the comparison of Euclidean distances can be significantly reduced. Compared with the log-sum LLR approximation, the proposed method reduces the computational complexity by more than 65% and 75% for 16-APSK and 32-APSK signals, respectively, with very small bit error rate performance degradation.

In this paper, an improved B-picture coding algorithm based on the symmetric bi-directional motion estimation (ME) is proposed. In addition to the block match error between blocks in the forward and backward reference frames, the proposed method exploits the previously-reconstructed template regions in the current and reference frames for bi-directional ME. The side match error between the predicted target block and its template is also employed in order to alleviate block discontinuities. To efficiently perform ME, an initial motion vector (MV) is adaptively derived by exploiting temporal correlations. Experimental results show that the number of generated bits is reduced by up to 9.31% when the proposed algorithm is employed as a new macroblock (MB) coding mode for the H.264/AVC standard.

The performance of English automatic speech recognition systems decreases when recognizing spontaneous speech mainly due to multiple pronunciation variants in the utterances. Previous approaches address this problem by modeling the alteration of the pronunciation on a phoneme to phoneme level. However, the phonetic transformation effects induced by the pronunciation of the whole sentence have not yet been considered. In this article, the sequence-based pronunciation variation is modeled using a noisy channel approach where the spontaneous phoneme sequence is considered as a “noisy” string and the goal is to recover the “clean” string of the word sequence. Hereby, the whole word sequence and its effect on the alternation of the phonemes will be taken into consideration. Moreover, the system not only learns the phoneme transformation but also the mapping from the phoneme to the word directly. In this study, first the phonemes will be recognized with the present recognition system and afterwards the pronunciation variation model based on the noisy channel approach will map from the phoneme to the word level. Two well-known natural language processing approaches are adopted and derived from the noisy channel model theory: Joint-sequence models and statistical machine translation. Both of them are applied and various experiments are conducted using microphone and telephone of spontaneous speech.

A low power and high performance with third order delta-sigma modulator for audio applications, fabricated in a 0.18 µm CMOS process, is presented. The modulator utilizes a third order noise shaping with only one opamp by using an opamp sharing technique. The opamp sharing among three integrator stages is achieved through the optimal operation timing, which makes use of the load capacitance differences between the three integrator stages. The designed modulator achieves 101.1 dB signal-to-noise ratio (A-weighted) and 101.5 dB dynamic range (A-weighted) with 7.5 mW power consumption from a 3.3 V supply. The die area is 1.27 mm2. The fabricated delta-sigma modulator achieves the highest figure-of-merit among published high performance low power audio delta-sigma modulators.

Recently, server virtualization technology, which is one of the key technologies to support cloud computing, has been making progress and gaining in maturity, resulting in an increase in the provision of cloud-based services and the integration of servers in enterprise networks. However, the progress in network virtualization technology, which is needed for the efficient and effective construction and operation of clouds, is lagging behind. It is only recently that all the required technical areas have started to be covered. This paper identifies network-related issues in cloud environments, describes the needs for network virtualization, and presents the recent trends in, and application fields of, network virtualization technology.

We investigate a sparsely encoded Hopfield model with unit replacement by using a statistical mechanical method called self-consistent signal-to-noise analysis. We theoretically obtain a relation between the storage capacity and the number of replacement units for each sparseness a. Moreover, we compare the unit replacement model with the forgetting model in terms of the network storage capacity. The results show that the unit replacement model has a finite value of the optimal sparseness on an open interval 0 (1/2 coding) < a < 1 (the limit of sparseness) to maximize the storage capacity for a large number of replacement units, although the forgetting model does not.

In this paper, we propose a high accuracy motion estimation method based on the spatio-temporal gradient method using high frame-rate images. In the method, we adopt spatial gradients with low estimated errors by the previous motion vectors. In addition, we evaluate the proposed method and confirm the effectiveness. Finally, we apply the method to super-resolution as an application of the proposed method.

This paper presents a novel CMOS readout circuit for satellite infrared time delay and integration (TDI) arrays. An integrate-while-read method is adopted, and a dead-pixel-elimination circuit for solving a critical problem of the TDI scheme is integrated within a chip. In addition, an adaptive charge capacity control method is proposed to improve the signal-to-noise ratio (SNR) for low-temperature targets. The readout circuit was fabricated with a 0.35-µm CMOS process for a 5004 mid-wavelength infrared (MWIR) HgCdTe detector array. Using the circuit, a 90% background-limited infrared photodetection (BLIP) is satisfied over a wide input range (∼200–330 K), and the SNR is improved by 11 dB for the target temperature of 200 K.

Orthogonal space-time block code (OSTBC) can achieve full diversity with a simple MLD, but OSTBC only achieves 3/4 of the maximum rate if more than two transmit antennas are used. To solve this problem, a quasi-orthogonal STBC (QOSTBC) scheme has been proposed. Even though a QOSTBC scheme can achieve the full rate, there are interference terms resulting from neighboring signals during detection. The existing QOSTBC using the pairs of transmitted symbols can be detected with two parallel MLD. Therefore, MLD based QOSTBC has higher complexity than OSTBC. To reduce the detection complexity, in this paper, we propose the heterogeneous constellation based QOSTBC for improving the detection property of QRD-MLD with maintaining a simple decoding structure.

This paper presents a no-reference (NR) based video-quality estimation method for compressed videos which apply inter-frame prediction. The proposed method does not need bitstream information. Only pixel information of decoded videos is used for the video-quality estimation. An activity value which indicates a variance of luminance values is calculated for every given-size pixel block. The activity difference between an intra-coded frame and its adjacent frame is calculated and is employed for the video-quality estimation. In addition, a blockiness level and a blur level are also estimated at every frame by analyzing pixel information only. The estimated blockiness level and blur level are also taken into account to improve quality-estimation accuracy in the proposed method. Experimental results show that the proposed method achieves accurate video-quality estimation without the original video which does not include any artifacts by the video compression. The correlation coefficient between subjective video quality and estimated quality is 0.925. The proposed method is suitable for automatic video-quality checks when service providers cannot access the original videos.

Secure computation of the set intersection functionality allows n parties to find the intersection between their datasets without revealing anything else about them. An efficient protocol for such a task could have multiple potential applications in commerce, health care, and security. However, all currently known secure set intersection protocols for n > 2 parties have computational costs that are quadratic in the (maximum) number of entries in the dataset contributed by each party, making secure computation of the set intersection only practical for small datasets. In this paper, we describe the first multi-party protocol for securely computing the set intersection functionality with both the communication and the computation costs that are quasi-linear in the size of the datasets. For a fixed security parameter, our protocols require O(n2k) bits of communication and Õ(n2k) group multiplications per player in the malicious adversary setting, where k is the size of each dataset. Our protocol follows the basic idea of the protocol proposed by Kissner and Song, but we gain efficiency by using different representations of the polynomials associated with users' datasets and careful employment of algorithms that interpolate or evaluate polynomials on multiple points more efficiently. Moreover, the proposed protocol is robust. This means that the protocol outputs the desired result even if some corrupted players leave during the execution of the protocol.

This paper clarifies the bifurcation structure of the chaotic attractor in an interrupted circuit with switching delay from theoretical and experimental view points. First, we introduce the circuit model and its dynamics. Next, we define the return map in order to investigate the bifurcation structure of the chaotic attractor. Finally, we discuss the dynamical effect of switching delay in the existence region of the chaotic attractor compared with that of a circuit with ideal switching.

This paper proposes a novel satellite channel allocation algorithm for a demand assigned multiple access (DAMA) controller. In satellite communication systems, the channels' total bandwidth and total power are limited by the satellite's transponder bandwidth and transmission power (satellite resources). Our algorithm is based on multi-carrier transmission and adaptive modulation methods. It optimizes channel elements such as the number of sub-carriers, modulation level, and forward error correction (FEC) coding rate. As a result, the satellite's transponder bandwidth and transmission power can be simultaneously used to the maximum and the overall system capacity, i.e., total transmission bit rate, will increase. Simulation results show that our algorithm increases the overall system capacity by 1.3 times compared with the conventional fixed modulation algorithm.

A generalized version of sequential logic circuit based neuron models is presented, where the dynamics of the model is modeled by an asynchronous cellular automaton. Thanks to the generalizations in this paper, the model can exhibit various neuron-like waveforms of the membrane potential in response to excitatory and inhibitory stimulus. Also, the model can reproduce four groups of biological and model neurons, which are classified based on existence of bistability and subthreshold oscillations, as well as their underlying bifurcations mechanisms.

This paper proposes a new generative model which can deal with rotational data variations by extending Separable Lattice 2-D HMMs (SL2D-HMMs). In image recognition, geometrical variations such as size, location and rotation degrade the performance. Therefore, the appropriate normalization processes for such variations are required. SL2D-HMMs can perform an elastic matching in both horizontal and vertical directions; this makes it possible to model invariance to size and location. To deal with rotational variations, we introduce additional HMM states which represent the shifts of the state alignments among the observation lines in a particular direction. Face recognition experiments show that the proposed method improves the performance significantly for rotational variation data.

With the aim of achieving heterogeneous integration of compound semiconductors with silicon technology, the fabrication of an InP/InGaAs transferred-substrate HBT (TS-HBT) on a Si substrate is reported. A current gain of 70 and a maximum current density of 12.3 mA/µm2 were confirmed in a TS-HBT with a 340-nm-wide emitter. From microwave characteristics of the TS-HBT obtained after de-embedding, a cutoff frequency (fT) of 510 GHz and a 26% reduction of the base-collector capacitance were estimated. However, the observed fT was too high for an HBT with a 150-nm-thick collector. This discrepancy can be explained by the error in de-embedding, because an open pad is observed to have large capacitance and strong frequency dependence due to the conductivity of the Si substrate.