A Neural Network Approach for Signal Detection in Digital Communications

In this paper, we propose a new approach for signal detection in wireless digital communications based on the neural network with transient chaos and time-varying gain (NNTCTG), and give a concrete model of the signal detector after appropriate transformations and mappings. It is well known that the problem of the maximum likelihood signal detection can be described as a complex optimization problem that has so many local optima that conventional Hopfield-type neural networks fail to solve. By refraining from the serious local optima problem of Hopfield-type neural networks, the NNTCTG makes use of the time-varying parameters of the recurrent neural network to control the evolving behavior of the network so that the network undergoes the transition from chaotic behavior to gradient convergence. It has richer and more flexible dynamics rather than conventional neural networks only with point attractors, so that it can be expected to have much ability to search for globally optimal or near-optimal solutions. After going through a transiently inverse-bifurcation process, the NNTCTG can approach the global optimum or the neighborhood of global optimum of our problem. Simulation experiments have been performed to show the effectiveness and validation of the proposed neural network based method for the signal detection in digital communications.     

Artificial neural networks for coordination and control: The portability of experiential representations

It is time to locate connectionist representation theory in the new wave of robotics research. The utility of representations developed in artificial neural networks (ANNs) during learning has been demonstrated in cognitive science research since the 1980s. The research reported here puts learned representations to work in a decentered control task, the disembodied arm problem, in which a mobile robot operates an arm fixed to a table to pick up objects. There is no physical linkage between the arm and the robot and so the robot's point of view must be decentered. This is done by developing a modular Artificial Neural Net system in three stages: (i) a classifier net is trained with laser scan data to output transformationally invariant position classes; (ii) an arm net is trained for picking up objects; (iii) an inter net is trained to communicate and coordinate the sensing and acting. The completed system is shown to create new nonsymbolic transformationally invariant representations in order to perform the effective generalization of decentered viewpoints.     

高氧化态（Ⅱ─Ⅳ）三核金属簇合物构型的人工神经网络判别方法

本文运用一典型的人工神经网络模型─“反向传播”模型，对高氧化态（Ⅱ─Ⅳ）三核金属簇合物的构型分布进行了分析，得到了较好的分类、预报结果为化合物结构分析提供了新的工具。     

Image processing using cellular neural networks based on multi-valued and universal binary neurons

Multi-valued and universal binary neurons (MVN and UBN) are the neural processing elements with the complex-valued weights and high functionality. It is possible to implement an arbitrary mapping described by partially defined multiple-valued function on the single MVN. An arbitrary mapping described by partially defined or fully defined Boolean function, which can be non-threshold, may be implemented on the single UBN. The quickly converging learning algorithms exist for both types of neurons. Such features of the MVN and UBN may be used for solving the different problems. One of the most successful applications of the MVN and UBN is their usage as basic neurons in the Cellular Neural Networks (CNN). It opens the new effective opportunities in nonlinear image filtering and its applications to noise reduction, edge detection and solving of the super resolution problem. A number of experimental results are presented to illustrate the performance of the proposed algorithms.An erratum to this article can be found at     

Validation of a Method of Determination of Apparent Diffusivity Versus Composition in Solids

A method is presented for computing the values of apparent diffusivity in solids with respect to the concentration of the diffusing substance (water or sodium chloride). This method does not require any assumption upon the mathematical relationship between diffusivity and concentration. It can be applied to experimental measurements of local concentration versus position within the solid (profiles) with relatively few measurements (circa 10) and a mathematical smoothing of the experimental data by using an artificial neural network model. The method was first validated on simulated data obtained by using a constant diffusivity value and on experimental profiles when the relation between diffusivity and concentration was given. It was then applied to original experimental moisture profiles obtained by putting gelatin gels with different initial moisture contents into contact for up to 14 days. The method was also successfully applied to five sets of experimental moisture and sodium chloride profiles taken from the literature and obtained from different food products. Apparent diffusivities calculated by our method were found in agreement with those obtained by authors using different numerical methods to compute the diffusivity values.     

Hierarchical intelligent control with flexible AC transmission systems application

A novel hierarchical intelligent controller configuration is proposed using an artificial neural network as a control-mode classifier in the supervisory level and a set of pre-designed controllers in the lower level. Controller outputs are modified nonlinearly by the classifying signals in a structure resembling one artificial neuron with adaptively changed weights. The lower-level local controllers are implemented using neural networks. An illustrative example of this approach is based on the transient stabilization of a single-machine infinite-bus system studied in Flexible AC Transmission Systems (FACTS) research.     

Bioprocess applications of model-based estimation techniques

The last decade has seen the development of a number of approaches for estimating those variables which are difficult to measure on-line in industrial process situations. Whilst a range of techniques is available, a common element is the use of process knowledge in the form of a system model. In the case of bioprocess systems, although a large range of models has been presented in the literature, their use in estimation schemes on an industrial scale has been limited. A number of reasons can be identified for their low level of utilisation. Of particular significance is the uncertainty which exists in quantifying system performance and the process-model mismatch which inevitably results. The level of ‘pre-defined model’ uncertainty, together with the knowledge gained during the course of the fermentation, serves to dictate estimator structure. The paper considers a range of estimation strategies and contrasts, through industrial applications, their performance characteristics and utility.     

Evolving Fault-Tolerant Neural Networks

In this paper, genetic algorithm is used to help improve the tolerance of feedforward neural networks against an open fault. The proposed method does not explicitly add any redundancy to the network, nor does it modify the training algorithm. Experiments show that it may profit the fault tolerance as well as the generalisation ability of neural networks.     

Artificial Neural Network Modeling of Osmotic Dehydration Mass Transfer Kinetics of Fruits

Artificial neural network (ANN) models were developed for the prediction of transient moisture loss (ML) and solid gain (SG) in osmotic dehydration of fruits using process kinetics data from the literature. ANN models for ML and SG were developed based on data over a broad range of operating conditions and ten common processing variables: temperature and concentration of osmotic solution, immersion time, initial water and solid content of the fruit, porosity, surface area, characteristic length, solution-to-fruit mass ratio, and agitation level. The trained models were able to accurately predict the outputs with associated regression coefficients (r) of 0.96 and 0.93, respectively, for ML and SG. These ANN models performed much better than those obtained from linear multivariate regression analysis. The large number of process variables and their wide ranges considered along with their easy implementation in a spreadsheet make them very useful and practical for process design and control.     

人工神经网络A／D变换器

本文对［５］的４位人工神经网络Ａ／Ｄ变换器进行了研究．指出按其方法推广实现多位Ａ／Ｄ时存在突触电阻大小以至无法实现且抗干扰能力也不好的问题并给出了解决办法．同时对如何改善神经元转移特性的硬限幅曲线作了简单讨论．