This paper focuses on the underdetermined blind source separation (BSS) of three speech signals mixed in a real environment from measurements provided by two sensors. To date, solutions to the underdetermined BSS problem have mainly been based on the assumption that the speech signals are sufficiently sparse. They involve designing binary masks that extract signals at time-frequency points where only one signal was assumed to exist. The major issue encountered in previous work relates to the occurrence of distortion, which affects a separated signal with loud musical noise. To overcome this problem, we propose combining sparseness with the use of an estimated mixing matrix. First, we use a geometrical approach to detect when only one source is active and to perform a preliminary separation with a time-frequency mask. This information is then used to estimate the mixing matrix, which allows us to improve our separation. Experimental results show that this combination of time-frequency mask and mixing matrix estimation provides separated signals of better quality (less distortion, less musical noise) than those extracted without using the estimated mixing matrix in reverberant conditions where the reverberant time (TR) was 130 ms and 200 ms. Furthermore, informal listening tests clearly show that musical noise is deeply lowered by the proposed method comparatively to the classical approaches.

A robust microphone array for speech enhancement and noise suppression is studied in this paper. To overcome target signal cancellation problem of conventional beamformer caused by array imperfections or reverberation effects of acoustic enclosure, the proposed microphone array adopts an arbitrary model of channel transfer function (TF) relating microphone and speech source. Since the estimation of channel TF itself is often intractable, herein, transfer function ratio (TFR) is estimated instead and used to form a suboptimal beamformer. A robust TFR estimation method is proposed based on signal subspace analysis technique against stationary or slowly varying noise. Experiments using simulated signal and actual signal recorded in a real room illustrate that the proposed method has high performance in adverse environment.

The authors realize a 50% length reduction of short-slot couplers in a post-wall dielectric substrate by two techniques. One is to introduce hollow rectangular holes near the side walls of the coupled region. The difference of phase constant between the TE10 and TE20 propagating modes increases and the required length to realize a desired dividing ratio is reduced. Another is to remove two reflection-suppressing posts in the coupled region. The length of the coupled region is determined to cancel the reflections at both ends of the coupled region. The total length of a 4-way Butler matrix can be reduced to 48% in comparison with the conventional one and the couplers still maintain good dividing characteristics; the dividing ratio of the hybrid is less than 0.1 dB and the isolations of the couplers are more than 20 dB.

This paper presents a study on the blind separation of a convoluted mixture of speech signals using Frequency Domain Independent Component Analysis (FDICA) algorithm based on the negentropy maximization of Time Frequency Series of Speech (TFSS). The comparative studies on the negentropy approximation of TFSS using generalized Higher Order Statistics (HOS) of different nonquadratic, nonlinear functions are presented. A new nonlinear function based on the statistical modeling of TFSS by exponential power functions has also been proposed. The estimation of standard error and bias, obtained using the sequential delete-one jackknifing method, in the approximation of negentropy of TFSS by different nonlinear functions along with their signal separation performance indicate the superlative power of the exponential-power-based nonlinear function. The proposed nonlinear function has been found to speed-up convergence with slight improvement in the separation quality under reverberant conditions.

In this paper, a low-voltage low-power sigma-delta modulator dedicated to implantable sensing devices is presented. This second-order single-loop sigma-delta modulator is implemented with half-delay integrators. These integrators are based on new fully-differential CMOS class AB switched-Operational Transconductance Amplifier (switched-OTA). An on-chip voltage doubler is introduced to locally boost a supply voltage at the input stage of a conventional OTA in order to allow rail-to-rail signal swing. Experimental results of the modulator fabricated in CMOS 0.18 µm technology confirm its expected features of a peak signal-to-noise ratio (SNR) of 72 dB, a signal-to-noise distortion ratio (SNDR) of 62 dB in a 5 kHz signal bandwidth, and a power consumption lower than 66 µW with a 900 mV voltage supply.

In adaptive motion estimation, spatial-temporal correlation based motion type inference has been recognized as an effective way to guide the motion estimation strategy adjustment according to video contents. However, the complexity and the reliability of those methods remain two crucial problems. In this paper, a motion vector field model is introduced as the basis for a new spatial-temporal correlation based motion type inference method. For each block, Full Search with Adaptive Search Window (ASW) and Three Step Search (TSS), as two search strategy candidates, can be employed alternatively. Simulation results show that the proposed method can constantly reduce the dynamic computational cost to as low as 3% to 4% of that of Full Search (FS), while remaining a closer approximation to FS in terms of visual quality than other fast algorithms for various video sequences. Due to its efficiency and reliability, this method is expected to be a favorable contribution to the mobile video communication where low power real-time video coding is necessary.

An approach based on signal subspace analysis is proposed to blind estimation of the PN (Pseudo Noise) sequence from lower SNR (Signal to Noise Ratios) DS/SS (Direct Sequence Spread Spectrum) signals. The received signal is divided into vectors according to a temporal window, from which an autocorrelation matrix is computed and accumulated. The PN sequence can be reconstructed from principal eigenvectors of the matrix.

In this paper, we propose a statistical model for detecting article errors, which Japanese learners of English often make in English writing. It is based on the three head words--the verb head, the preposition, and the noun head. To overcome the data sparseness problem, we apply the backed-off estimate to it. Experiments show that its performance (F-measure=0.70) is better than that of other methods. Apart from the performance, it has two advantages: (i) Rules for detecting article errors are automatically generated as conditional probabilities once a corpus is given; (ii) Its recall and precision rates are adjustable.

In this letter, we propose a closed-form joint space-time channel and direction of arrival (DOA) blind estimation algorithm for space-time coded MC-CDMA systems equipped with a uniform linear array (ULA) at the base station in frequency-selective fading environments. We prove that the signal subspaces defined by the receive data covariance matrix can be determinately separated into an equivalent set of signal subspaces spanned by the space-time channel vector of an individual user. From these signal subspaces, the space-time channels of multiple users are estimated using the subspace method.

In this paper, we propose an OFDM scheme with pre-IDFT/DFT and the frequency domain equalization on frequency-selective Rayleigh fading channels. In this scheme, a two-dimensional block interleaving is used to randomize the correlated noise caused by the frequency domain linear equalizer. Then, the pre-DFT averages the interleaved noise enhancement and improves the error performance of the proposed scheme. Computer simulations confirm the bit error probability of the proposed scheme for multilevel modulations.

Based on the Si CMOS process, a low operating voltage and low power light emitting device is presented. It has a power transfer efficiency of 1 to 2 orders higher than previous reports and can be used as a high efficiency photodiode. Configurations using the same structure as both the light emitter and the optical receiver, and employing a simple modulation instrument is then proposed for applications in the chip-to-chip optical alignment and the signal transmission. Only single power supply is required in the emitter-receiver circuits and is compatible with other integrated circuits made by the CMOS process.

This paper presents the design of a novel method for improvement of the operation of distance relays during capacitive voltage transformer transients using artificial neural network. The proposed module uses voltage and current signals to learn the hidden relationship existing in the input patterns. Simulation studies are preformed and the influence of changing system parameters, such as fault resistance and source impedance is studied. Details of the design procedure and the results of performance studies with the proposed relay are given in the paper. Performance studies results show that the proposed algorithm decreases the effects of CVT transients and is fast and accurate.

When a dependency parser analyzes long sentences with fewer subjects than predicates, it is difficult for it to recognize which predicate governs which subject. To handle such syntactic ambiguity between subjects and predicates, we define an "a subject clause (s-clause)" as a group of words containing several predicates and their common subject. This paper proposes a two-phase method for S-clause segmentation. The first phase reduces the number of candidates of S-clause boundaries, and the second performs S-clause segmentation using decision trees. In experimental evaluation, the S-clause information turned out to be effective for determining the governor of a subject and that of a predicate in dependency parsing. Further syntactic analysis using S-clauses achieved an improvement in precision of 5 percent.

This paper presents an analysis of electrically large antennas using the adaptive integral method (AIM). The arbitrarily shaped perfectly conducting surfaces are modeled using triangular patches and the associated electric field integral equation (EFIE) is solved for computing the radiation patterns of these antennas. The method of moments (MoM) is used to discretize the integral equations and the resultant matrix system will be solved by an iterative solver. The AIM is employed in the iterative solver to speed up the matrix-vector multiplication and to reduce the memory requirement. As specific applications, radiation patterns of parabolic reflectors and X-band horns are computed using the proposed method.

This paper describes an ADPLL (All Digital Phase-Locked Loops) with a small DCO (Digitally Controlled Oscillator), low jitter, fine resolution and wide lock range suitable for mobile appplications. The novel DCO circuit is controlled by digital control codes with thermometer type instead of previous binary weighted type. Therefore, the DCO has small area and it has significantly small jitter when the control input is updated. The hierarchical DCO type with two loops makes it possible to have fine resolution and wide lock range. Functional verification and noise analysis of the ADPLL is performed by MATLAB simulink to improve design TAT (Turn-Around Time). And The ADPLL chip is in fabrication using a SEC 0.18 µm CMOS technology. The ADPLL has lock range between 520 MHz and 1.5 GHz and has peak-to-peak jitter 70 ps at 670 MHz.

This paper presents a technique for designing a dielectric Electronically Steerable Parasitic Array Radiator (Espar) antenna to achieve miniaturization of the conventional Espar antenna. The antenna's size is reduced by immersing the central active element in a dielectric cylinder, mounting the surrounding planar parasitic elements at the circumference of the cylinder, and decreasing the radius of the ground skirt to that of the parasitic elements. An example of a polycarbonate (εr = 2.9 + j0.006) Espar antenna operating at 2.484 GHz is optimised by using a genetic algorithm in conjunction with an FEM-based cost function. The designed antenna generates a half-power beam width of 78and a main lobe that elevates at an angle of only 5from the horizontal plane. The designed antenna is also fabricated and measured. Good agreement between the measurement and simulation results is obtained. We reduce the size of the designed Espar antenna to 1/8 the size of its conventional counterpart while achieving a 12improvement in half-power beam width.

It is not always possible to find a global parse for an input sentence owing to problems such as errors of a sentence, incompleteness of lexicon and grammar. Partial parsing is an alternative approach to respond to these problems. Partial parsing techniques try to recover syntactic information efficiently and reliably by sacrificing completeness and depth of analysis. One of the difficulties in partial parsing is how the grammar might be automatically extracted. In this paper we present a method of automatically extracting partial parsing rules from a tree-annotated corpus using the decision tree method. Our goal is deterministic global parsing using partial parsing rules, in other words, to extract partial parsing rules with higher accuracy and broader expansion. First, we define a rule template that enables to learn a subtree for a given substring, so that the resultant rules can be more specific and stricter to apply. Second, rule candidates extracted from a training corpus are enriched with contextual and lexical information using the decision tree method and verified through cross-validation. Last, we underspecify non-deterministic rules by merging substructures with ambiguity in those rules. The learned grammar is similar to phrase structure grammar with contextual and lexical information, but allows building structures of depth one or more. Thanks to automatic learning, the partial parsing rules can be consistent and domain-independent. Partial parsing with this grammar processes an input sentence deterministically using longest-match heuristics, and recursively applies rules to an input sentence. The experiments showed that the partial parser using automatically extracted rules is not only accurate and efficient but also achieves reasonable coverage for Korean.

According to the proliferation of ubiquitous computing, various products which contain large-size embedded software have been developed. One of most typical features of embedded software is concurrency of software and hardware factors. That is, software has connected deeply into hardware devices. The existence of various hardware make quality assurance of embedded software more difficult. In order to assure quality of embedded software more effectively, this paper discusses features of embedded software and an effective method for quality assurance for embedded software. In this paper, we first analyze a failure distribution of embedded software and discuss the effects of hardware devices on quality of embedded software. Currently, in order to reduce hardware related faults, huge effort for testing with large number of test items is required. Thus, one of the most important issues for quality assurance of embedded software is how to reduce the cost and effort of software testing. Next, focusing on hardware constraints as well as software specifications in embedded software, we propose an evaluation metrics for determinating important functions for quality of embedded software. Furthermore, by referring to the metrics, undesirable behaviors of important functions are identified as root nodes of fault tree analysis. From the result of case study applying the proposed method to actual project data, we confirmed that test items considering the property of embedded software are constructed. We also confirmed that the constructed test items are appropriate to detect hardware related faults in embedded systems.

In a direct conversion receiver for mobile communication, it is important to reduce power dissipation. Because a low pass filter in a direct conversion receiver must suppress adjacent channel signals, a high order and high power dissipation is often required in the low pass filter. We propose a new phase compensation technique suitable for a low power transconductor used in a GmC filter as a low pass filter. The new phase compensation technique reduces 10% of power dissipation.

An even-harmonic mixer using a bipolar differential pair (bipolar harmonic mixer;BHMIX) is theoretically analyzed from the direct conversion point of view; i.e, conversion gain, third-order input intercept point (IIP3), self-mixing induced dc offset level, and second-order input intercept point (IIP2). Also, noise are analyzed based on nonlinear large-signal model, and numerical results are given. Noises are treated as cyclostationary noises, thus all the folding effects are taken into account. Factors determining IIP3, IIP2, dc offset, and noise are identified and estimation procedures for these characteristics are obtained. For example, design guidelines for the optimal noise performance are given. Measured results support all the analysis results, and they are very useful in the practical BHMIX design.