Nonlinear black-box modeling in system identification: a unified overview

A nonlinear black-box structure for a dynamical system is a model structure that is prepared to describe virtually any nonlinear dynamics. There has been considerable recent interest in this area, with structures based on neural networks, radial basis networks, wavelet networks and hinging hyperplanes, as well as wavelet-transform-based methods and models based on fuzzy sets and fuzzy rules. This paper describes all these approaches in a common framework, from a user's perspective. It focuses on what are the common features in the different approaches, the choices that have to be made and what considerations are relevant for a successful system-identification application of these techniques. It is pointed out that the nonlinear structures can be seen as a concatenation of a mapping form observed data to a regression vector and a nonlinear mapping from the regressor space to the output space. These mappings are discussed separately. The latter mapping is usually formed as a basis function expansion. The basis functions are typically formed from one simple scalar function, which is modified in terms of scale and location. The expansion from the scalar argument to the regressor space is achieved by a radial- or a ridge-type approach. Basic techniques for estimating the parameters in the structures are criterion minimization, as well as two-step procedures, where first the relevant basis functions are determined, using data, and then a linear least-squares step to determine the coordinates of the function approximation. A particular problem is to deal with the large number of potentially necessary parameters. This is handled by making the number of ‘used’ parameters considerably less than the number of ‘offered’ parameters, by regularization, shrinking, pruning or regressor selection.     

Human chromosome 5 sequence primer amplifies Alu polymorphisms on chromosomes 2 and 17

Members of the Alu family of repetitive elements occur frequently in the human genome and are often polymorphic. Techniques involving Alu element mediated polymerase chain reactions (Alu PCR) allow the isolation of region-specific human DNA fragments from mixed DNA sources. Such fragments are a source of region-specific Alu elements useful for the detection of Alu-related polymorphisms. A clone from human chromosome 5, corresponding to locus D5F40S1, was isolated using Alu PCR differential hybridization. Alu elements within this clone were investigated for the presence of potentially polymorphic 3' polyA tails. Primers were devised to amplify the 3' polyA tail of an Alu element present within the clone. One primer, D5F40S1-T, was specific to the DNA flanking the 3' end of the Alu element, and the other primer was homologous to sequences within the element. When these primers were used in PCR reactions, products from chromosomes 2 and 17 (loci D2F40S2 and D17F40S3) were amplified in addition to the expected product from chromosome 5. The most likely explanation for this nonspecific amplification is that the D5F40S1-T primer is located within a low-copy repetitive element that is 3' of the Alu element. This phenomenon presents a potential problem for the identification of region-specific Alu polymorphisms.     

Absorbing surface impedances of lossy layers in the finite difference time-domain method

In this paper the time-domain surface impedances of an homogeneous absorber layer, are given for the vertical and horizontal polarizations, or respectively for the electric field perpendicular or parallel to the incidence plane. It turns out that the application of the concept in finite difference time-domain (FDTD) in absorbing surface impedances boundary conditions, gives results in good agreement with analytical Fresnel reflection coefficients.     

Use of the cell division protein FtsZ as a means of differentiating among Bartonella species

Genes coding for homologs of the highly conserved cell division protein FtsZ were isolated from Bartonella henselae and Bartonella quintana, the causative agents of cat scratch disease and trench fever, respectively. DNA fragments coding for the ftsZ open reading frames (ORFs) were cloned into Escherichia coli following PCR amplification with primers based on the ftsZ sequence of the closely related species Bartonella bacilliformis. The amino acid sequences predicted from the cloned B. henselae and B. quintana ftsZ ORFs are 81 to 83% identical to the corresponding protein in B. bacilliformis. Like the FtsZ protein of B. bacilliformis, the B. henselae and B. quintana homologs are about twice as large as the FtsZ proteins reported in most other organisms. Localized sequence differences within the C-terminal coding regions of the Bartonella ftsZ genes were used as the basis for species-specific identification of these organisms at both the DNA and protein levels. Oligonucleotide primers which permit the amplification of an ftsZ fragment from each of the Bartonella species without amplifying DNA from the other two species were designed. Anti-FtsZ antisera raised in rabbits against synthetic peptides corresponding to the relatively divergent C-terminal regions were shown via Western blot analysis to react only with the FtsZ protein from the cognate Bartonella species. These observations raise the possibility that the differences in ftsZ sequences can be used as the basis for diagnostic tests to differentiate among these closely related pathogens.     

Atrazine, alachlor, and total inorganic nitrogen concentrations of winter wind-eroded sediment samples

Virus-specific CD8+ effector T cells (eCTL) are enriched in the lungs of mice with primary influenza pneumonia, though later detection of memory T cells (mCTL) in the mediastinal lymph nodes (MLN) or spleen by peptide-based staining protocols is at the limits of flow cytometric analysis. Respiratory challenge with an H3N2 virus months after H1N1 priming induces a massive recall response, which reduces virus titers 2-3 days earlier than in nave controls. Influenza-specific mCTL produce interferon-gamma within 6 hr, but still take 4-5 days to localize to the infected respiratory tract. The delay reflects that the recall response develops first in the MLN, which contains relatively few mCTL. The response to a subdominant epitope is less obvious after secondary challenge.     

Measurement of pulmonary resistance and dynamic compliance with airway obstruction

We compared four algorithms by using least squares regression for determination of pulmonary resistance and dynamic elastance in subjects with emphysema, normal subjects, and subjects with asthma before and after bronchoconstriction. The four methods evaluated include 1) a single resistance and elastance, 2) separate resistances and elastances for each half breath, 3) separate inspiratory and expiratory resistances with a single elastance, and 4) separate inspiratory and expiratory resistances, an expiratory volume interaction term, and a single elastance. All methods gave comparable results in normal and asthmatic subjects. We found expiratory resistance was larger than inspiratory resistance in normal and asthmatic subjects during control conditions, but inspiratory resistance was higher than expiratory resistance in subjects who experienced severe bronchoconstriction in response to methacholine. In subjects who are flow limited, method 2 gives a higher inspiratory resistance than would be computed by assuming that the elastic pressure-volume curve passes through the zero-flow points. Methods 1 and 3 overestimate dynamic elastance and inspiratory resistance. Method 4 appears to identify flow limitation and dynamic hyperinflation and gives a good measure of inspiratory resistance and dynamic elastance.     

An orphan nuclear receptor activated by pregnanes defines a novel steroid signaling pathway

PURPOSE: To determine the effects of 28 d of creatine supplementation during training on body composition, strength, sprint performance, and hematological profiles. METHODS: In a double-blind and randomized manner, 25 NCAA division IA football players were matched-paired and assigned to supplement their diet for 28 d during resistance/agility training (8 h x wk[-1]) with a Phosphagen HP (Experimental and Applied Sciences, Golden, CO) placebo (P) containing 99 g x d(-1) of glucose, 3 g x d(-1) of taurine, 1.1 g x d(-1) of disodium phosphate, and 1.2 g x d(-1) of potassium phosphate (P) or Phosphagen HP containing the P with 15.75 g x d(-1) of HPCE pure creatine monohydrate (HP). Before and after supplementation, fasting blood samples were obtained; total body weight, total body water, and body composition were determined; subjects performed a maximal repetition test on the isotonic bench press, squat, and power clean; and subjects performed a cycle ergometer sprint test (12 x 6-s sprints with 30-s rest recovery). RESULTS: Hematological parameters remained within normal clinical limits for active individuals with no side effects reported. Total body weight significantly increased (P < 0.05) in the HP group (P 0.85 +/- 2.2; HP 2.42 +/- 1.4 kg) while no differences were observed in the percentage of total body water. DEXA scanned body mass (P 0.77 +/- 1.8; HP 2.22 +/- 1.5 kg) and fat/bone-free mass (P 1.33 +/- 1.1; HP 2.43 +/- 1.4 kg) were significantly increased in the HP group. Gains in bench press lifting volume (P -5 +/- 134; HP 225 +/- 246 kg), the sum of bench press, squat, and power clean lifting volume (P 1,105 +/- 429; HP 1,558 +/- 645 kg), and total work performed during the first five 6-s sprints was significantly greater in the HP group. CONCLUSION: The addition of creatine to the glucose/taurine/electrolyte supplement promoted greater gains in fat/bone-free mass, isotonic lifting volume, and sprint performance during intense resistance/agility training.