Feature extraction by supervised independent component analysis based on category information

Independent component analysis (ICA) is a technique of transforming observation signals into their unknown independent components; hence, ICA has often been applied to blind signal separation problems. In this application, it is expected that the obtained independent components extract essential information of independent signal sources from input data in an unsupervised fashion. Based on such characteristics, ICA is currently utilized as a feature extraction method for images and sounds for recognition purposes. However, since ICA is an unsupervised learning, the obtained independent components are not always useful in recognition. To overcome this problem, we propose a supervised approach to ICA using category information. The proposed method is implemented in a conventional three‐layered neural network, but its objective function to be minimized is defined not only for the output layer but also for the hidden layer. The objective function consists of the following two terms: one evaluates the kurtosis of hidden unit outputs and the other evaluates the error between output signals and their teacher signals. The experiments are performed using several standard datasets to evaluate performance of the proposed algorithm. It is confirmed that a higher recognition accuracy is attained by the proposed method as compared with a conventional ICA algorithm. © 2007 Wiley Periodicals, Inc. Electr Eng Jpn, 161(2): 25–32, 2007; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.20522     

A new blind deconvolution algorithm based on the probability distribution method

Blind deconvolution is a method of recovering transmitted signals from only received signals. The probability distribution method is one of the blind deconvolution methods. This method has two problems: it has slower convergence and its reliability is lower. In this paper, we propose a new algorithm for solving these two problems. The proposed algorithm is as follows. (1) It is based on the adaptive processing with each sample. (2) Kurtosis is adaptively estimated by recovered signals with each sample. (3) Cost function is decided by kurtosis. (4) Transmitted signals are recovered by received signals using decided cost function on the sample time. We confirm the validity of the new algorithm by computer simulation. © 2007 Wiley Periodicals, Inc. Electr Eng Jpn, 162(1): 56–65, 2008; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.20317     

Synthesis of soluble poly(9,10-dihydrophenanthrene-2,7-diyl)s. A new class of luminescent poly(p-phenylene)s with ethylene type bridges

Poly(9,10-dihydrophenanthrene-2,7-diyl)s with -OSi(R)₂(R′) groups at the 9,10-positions were synthesized by dehalogenative polycondensation of the corresponding monomers by using a zerovalent nickel complex. They showed number average molecular weights (M_n's) of 9800-69,000 and high quantum yields (62%-quantitative) in photoluminescence. Palladium catalyzed copolymerization of 2,7-dibromo-9,10-dihydrophenanthrene having -OCH₃ or -OSi(R)₂(R′) groups at the 9,10-positions with diethynyl- or diboronic-aromatic compounds also gave photoluminescent polymers with high quantum yields.     

Cesium-adsorption capacity and hydraulic conductivity of sealing geomaterial made with marine clay,bentonite, and zeolite

Great amounts of soil and waste contaminated with radioactive cesium have been generated due to the decontamination work after the Fukushima Daiichi Nuclear Power Plant accident. The aim of this study is to develop a sealing geomaterial for use at the disposal facilities of the soil and waste constructed in the maritime environment. The geomaterial consists of marine clay, bentonite, and zeolite. The hydraulic conductivity and cesium-adsorption performance of the geomaterial were examined through laboratory tests with different proportions of bentonite and zeolite added to marine clay. It was concluded that the hydraulic conductivity could be reduced to the required level by increasing the amount of bentonite and that the cesium-adsorption capacity could be enhanced by increasing the amount of zeolite.     

Functional analysis of genes encoding putative oxidoreductases in Aspergillus oryzae, which are similar to fungal fructosyl-amino acid oxidase

We found 11 genes (FAO1-11) encoding putative oxidoreductases in the Aspergillus oryzae genome, which are similar to fungal fructosyl-amino acid oxidases. The cDNAs corresponding to the genes were cloned and expressed in Escherichia coli. rFao2 had fructosyl-amino acid oxidase activity, whereas rFao1 did not show any enzyme activity, even though the deduced amino acid sequence of Fao1 is identical to that of one of the fructosyl-amino acid oxidase isozymes from Aspergillus oryzae. rFao7 and rFao8 showed oxidase activity toward sarcosine, L-pipecolate, and L-proline. rFao10 was active toward only sarcosine, of the substrates tested. The functions of the other proteins were also predicted from a phylogenetic analysis.     

Estimation of concentrations of antifungal agents allowed as food additives in foods and their daily intake based on official inspection results in Japan in fiscal year 1998

The mean concentrations and daily intake of four antifungal agents were estimated based on the results of an analysis of 7,005 samples of food obtained in official inspections by Japanese local governments in fiscal year 1998. The mean concentration of diphenyl was 0.0004% of the allowable limit, and those of imazalil, o-phenylphenol, and thiabendazole were 14.0%, 3.5%, and 5.7%, respectively. The daily intakes of these antifungal agents per person, estimated from their concentrations and the daily consumption of the foods, were 0.000326, 1.89, 11.5, and 23.3 micrograms, respectively, and assuming a body weight of 50 kg, the amounts of these antifungal agents consumed were 0.000013%, 0.15%, 0.12%, and 0.47% of the acceptable daily intake, respectively. These values are similar to the values obtained on the basis of the results of the official inspection in fiscal years 1994 and 1996, except that the amount of diphenyl is much lower (1/100).     

Effect of the atmosphere on chemical composition and electrochemical properties of solid electrolyte interface on electrodeposited Li metal

Many efforts have been paid to realize the superior anodes for future Li batteries in either the dry Ar atmosphere or the dry air atmosphere. In this work, in order to clarify the effects of such atmospheres, the most reactive anodes of Li were freshly electrodeposited under the dry Ar or under the dry air condition. The Solid Electrolyte Interface (SEI) formed during the electrodeposition of Li anodes is revealed to have a different chemical composition and protective feature. The Li deposited under the dry air was revealed to have longer cycle life in the electrolyte than that deposited in Ar, even in the electrolyte containing ionic liquid. From the XPS results, the SEI formed in dry air is proved to be different from that formed in Ar gas atmospheres, that is, the SEI formed in dry air consists of Li₂CO₃ and Li nitride. In order to improve the performance of the anodes, the atmosphere for the initial preparation of the anode/electrolyte interface should be tuned.     

Hydrogen storage properties of lithium silicon alloy synthesized by mechanical alloying

A lithium silicon alloy was synthesized by mechanical alloying method. Hydrogen storage properties of this Li-Si-H system were studied. During hydrogenation of the lithium silicon alloy, lithium atom was extracted from the alloy and lithium hydride was generated. Equilibrium hydrogen pressures for desorption and absorption reactions were measured in a temperature range from 400 to 500 °C to investigate the thermodynamic characteristics of the system, which can reversibly store 5.4 mass% hydrogen with smaller reaction enthalpy than simple metal Li. Li absorbing alloys, which have been widely studied as a negative electrode material for Li ion rechargeable batteries, can be used as hydrogen storage materials with high hydrogen capacity.     

Stochastic model predictive braking control for heavy-duty commercial vehicles during uncertain brake pressure and road profile conditions

When heavy-duty commercial vehicles (HDCVs) must engage in emergency braking, uncertain conditions such as the brakepressure and road profile variations will inevitably affect braking control. To minimize these uncertainties, we propose acombined longitudinal and lateral controller method based on stochastic model predictive control (SMPC) that is achieved viaChebyshev–Cantelli inequality. In our method, SMPC calculates braking control inputs based on a finite time prediction thatis achieved by solving stochastic programming elements, including chance constraints. To accomplish this, SMPC explicitlydescribes the probabilistic uncertainties to be used when designing a robust control strategy. The main contribution of thispaper is the proposal of a braking control formulation that is robust against probabilistic friction circle uncertainty effects.More specifically, the use of Chebyshev–Cantelli inequality suppresses road profile influences, which have characteristicsthat are different from the Gaussian distribution, thereby improving both braking robustness and control performance againststatistical disturbances.Additionally, since theKalman filtering (KF) algorithm is used to obtain the expectation and covarianceused for calculating deterministic transformed chance constraints, the SMPC is reformulated as a KF embedded deterministicMPC. Herein, the effectiveness of our proposed method is verified via a MATLAB/Simulink and TruckSim co-simulation.