A Hybrid Forward Algorithm for RBF Neural Network Construction

This paper proposes a novel hybrid forward algorithm (HFA) for the construction of radial basis function (RBF) neural networks with tunable nodes. The main objective is to efficiently and effectively produce a parsimonious RBF neural network that generalizes well. In this study, it is achieved through simultaneous network structure determination and parameter optimization on the continuous parameter space. This is a mixed integer hard problem and the proposed HFA tackles this problem using an integrated analytic framework, leading to significantly improved network performance and reduced memory usage for the network construction. The computational complexity analysis confirms the efficiency of the proposed algorithm, and the simulation results demonstrate its effectiveness

A novel continuous forward algorithm for RBF neural modelling

A continuous forward algorithm (CFA) is proposed for nonlinear modelling and identification using radial basis function (RBF) neural networks. The problem considered here is simultaneous network construction and parameter optimization, well-known to be a mixed integer hard one. The proposed algorithm performs these two tasks within an integrated analytic framework, and offers two important advantages. First, the model performance can be significantly improved through continuous parameter optimization. Secondly, the neural representation can be built without generating and storing all candidate regressors, leading to significantly reduced memory usage and computational complexity. Computational complexity analysis and simulation results confirm the effectiveness.

Physical modelling of millimetre wave signal reflection from forward biased PIN diodes

This paper examines the DC power requirements of PIN diodes which, with suitable applied DC bias, have the potential to reflect or to permit transmission of millimetre wave energy through them by the process of inducing a semiconductor plasma layer in the i-region. The study is conducted using device level simulation of SOI and bulk PIN diodes and reflection modelling based on the Drude conduction model. We examined five diode lengths (60–140 µm) and seven diode thicknesses (4–100 µm). Simulation output for the diodes of varying thicknesses was subsequently used in reflection modelling to assess their performance for 100 GHz operation. It is shown that substantially high DC input power is required in order to induce near total reflection in SOI PIN diodes at 100 GHz. Thinner devices consume less DC power, but reflect less incident radiation for given input power. SOI diodes are shown to have improved carrier confinement compared with bulk diodes.

The diamond pyramid structure in electroless copper deposit, its atomic model and molecular dynamics simulation

In this seminar, I will talk about the discovery of the diamond pyramid structures in the electroless copper deposits on both epoxy and stainless steel substrates. The surface morphology of the structure was characterized with scanning electron microscopy (SEM). According to the morphological feature of the structure, an atom model was brought forward in order to describe the possible mechanism of forming such structure. Molecular dynamics simulations were then carried out to investigate the growing process of the diamond pyramid structure. The final structures of the simulation were compared with the SEM images and the atomic model. The radial distribution function of the final structures of the simulation was compared with that calculated from the X-ray diffraction pattern of the electroless copper deposit sample.

A novel forward gene selection algorithm for microarray data

This paper investigates the gene selection problem for microarray data with small samples and variant correlation. Most existing algorithms usually require expensive computational effort, especially under thousands of gene conditions. The main objective of this paper is to effectively select the most informative genes from microarray data, while making the computational expenses affordable. This is achieved by proposing a novel forward gene selection algorithm (FGSA). To overcome the small samples' problem, the augmented data technique is firstly employed to produce an augmented data set. Taking inspiration from other gene selection methods, the L2-norm penalty is then introduced into the recently proposed fast regression algorithm to achieve the group selection ability. Finally, by defining a proper regression context, the proposed method can be fast implemented in the software, which significantly reduces computational burden. Both computational complexity analysis and simulation results confirm the effectiveness of the proposed algorithm in comparison with other approaches

Modeling and simulation of bulk gallium nitride power semiconductor devices

Bulk gallium nitride (GaN) power semiconductor devices are gaining significant interest in recent years, creating the need for technology computer aided design (TCAD) simulation to accurately model and optimize these devices. This paper comprehensively reviews and compares different GaN physical models and model parameters in the literature, and discusses the appropriate selection of these models and parameters for TCAD simulation. 2-D drift-diffusion semi-classical simulation is carried out for 2.6 kV and 3.7 kV bulk GaN vertical PN diodes. The simulated forward current-voltage and reverse breakdown characteristics are in good agreement with the measurement data even over a wide temperature range.