Complex-valued B-spline neural networks for modeling and inverting Hammerstein systems

Many communication signal processing applications involve modelling and inverting complex-valued (CV) Hammerstein systems. We develops a new CV B-spline neural network approach for efficient identification of the CV Hammerstein system and effective inversion of the estimated CV Hammerstein model. Specifically, the CV nonlinear static function in the Hammerstein system is represented using the tensor product from two univariate B-spline neural networks. An efficient alternating least squares estimation method is adopted for identifying the CV linear dynamic model’s coefficients and the CV B-spline neural network’s weights, which yields the closed-form solutions for both the linear dynamic model’s coefficients and the B-spline neural network’s weights, and this estimation process is guaranteed to converge very fast to a unique minimum solution. Furthermore, an accurate inversion of the CV Hammerstein system can readily be obtained using the estimated model. In particular, the inversion of the CV nonlinear static function in the Hammerstein system can be calculated effectively using a Gaussian-Newton algorithm, which naturally incorporates the efficient De Boor algorithm with both the B-spline curve and first order derivative recursions. The effectiveness of our approach is demonstrated using the application to equalisation of Hammerstein channels.

Root elongation rate is correlated with the length of the bare root apex of maize and lupin roots despite contrasting responses of root growth to compact and dry soils

Background We investigated interacting effects of matric potential and soil strength on root elongation of maize and lupin, and relations between root elongation rates and the length of bare (hairless) root apex. Methods Root elongation rates and the length of bare root apexwere determined formaize and lupin seedlings in sandy loam soil of various matric potentials (−0.01 to −1.6 MPa) and bulk densities (0.9 to 1.5 Mg m−3). Results Root elongation rates slowed with both decreasing matric potential and increasing penetrometer resistance. Root elongation of maize slowed to 10 % of the unimpeded rate when penetrometer resistance increased to 2 MPa, whereas lupin elongated at about 40 % of the unimpeded rate. Maize root elongation rate was more sensitive to changes in matric potential in loosely packed soil (penetrometer resistances <1 MPa) than lupin. Despite these differing responses, root elongation rate of both species was linearly correlated with length of the bare root apex (r2 0.69 to 0.97). Conclusion Maize root elongation was more sensitive to changes in matric potential and mechanical impedance than lupin. Robust linear relationships between elongation rate and length of bare apex suggest good potential for estimating root elongation rates for excavated roots.

Calcium-mediated binding of DNA to 1,2-distearoyl-sn-glycero-3-phosphocholine-containing mixed lipid monolayers

The calcium-mediated interaction of DNA with monolayers of the non-toxic, zwitterionic phospholipid, 1,2-distearoyl-sn-glycero-3-phosphocholine when mixed with 50 mol% of a second lipid, either the zwitteronic 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine or neutral cholesterol was investigated using a combination of surface pressure-area isotherms, Brewster angle microscopy, external reflectance Fourier transform infrared spectroscopy and specular neutron reflectivity in combination with contrast variation. When calcium and DNA were both present in the aqueous subphase, changes were observed in the compression isotherms as well as the surface morphologies of the mixed lipid monolayers. In the presence of calcium and DNA, specular neutron reflectivity showed that directly underneath the head groups of the lipids comprising the monolayers, DNA occupied a layer comprising approximately 13 and 18% v/v DNA for the 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine and cholesterol-containing monolayers, respectively. The volume of the corresponding layer for 1,2-distearoyl-sn-glycero-3-phosphocholine only containing monolayers was ∼15% v/v DNA. Furthermore regardless of the presence and nature of the second lipid and the surface pressure of the monolayer, the specular neutron reflectivity experiments showed that the DNA-containing layer was 20–27 Å thick, suggesting the presence of a well-hydrated layer of double-stranded DNA. External reflectance Fourier transform infrared studies confirmed the presence of double stranded DNA, and indicated that the strands are in the B-form conformation. The results shed light on the interaction between lipids and nucleic acid cargo as well as the role of a second lipid in lipid-based carriers for drug delivery.

Probabilistic wind power forecasts with an inverse power curve transformation and censored regression

Forecasting wind power is an important part of a successful integration of wind power into the power grid. Forecasts with lead times longer than 6 h are generally made by using statistical methods to post-process forecasts from numerical weather prediction systems. Two major problems that complicate this approach are the non-linear relationship between wind speed and power production and the limited range of power production between zero and nominal power of the turbine. In practice, these problems are often tackled by using non-linear non-parametric regression models. However, such an approach ignores valuable and readily available information: the power curve of the turbine's manufacturer. Much of the non-linearity can be directly accounted for by transforming the observed power production into wind speed via the inverse power curve so that simpler linear regression models can be used. Furthermore, the fact that the transformed power production has a limited range can be taken care of by employing censored regression models. In this study, we evaluate quantile forecasts from a range of methods: (i) using parametric and non-parametric models, (ii) with and without the proposed inverse power curve transformation and (iii) with and without censoring. The results show that with our inverse (power-to-wind) transformation, simpler linear regression models with censoring perform equally or better than non-linear models with or without the frequently used wind-to-power transformation.