Developing pseudo random number generator based on neural networks and neurofuzzy systems

Pseudo random number generators (PRNGs) are one of the most important components in security and cryptography applications. We propose an application of Hopfield Neural Networks (HNN) as pseudo random number generator. This research is done based on a unique property of HNN, i.e., its unpredictable behavior under certain conditions. Also, we propose an application of Fuzzy Hopfield Neural Networks (FHNN) as pseudo random number generator. We compare the main features of ideal random number generators with our proposed PRNGs. We use a battery of statistical tests developed by National Institute of Standards and Technology (NIST) to measure the performance of proposed HNN and FHNN. We also measure the performance of other standard PRNGs and compare the results with HNN and FHNN PRNG. We have shown that our proposed HNN and FHNN have good performance comparing to other PRNGs accordingly.

Developing pseudo random number generator based on neural networks and neurofuzzy systems

Pseudo random number generators (PRNGs) are one of the most important components in security and cryptography applications. We propose an application of Hopfield Neural Networks (HNN) as pseudo random number generator. This research is done based on a unique property of HNN, i.e., its unpredictable behavior under certain conditions. Also, we propose an application of Fuzzy Hopfield Neural Networks (FHNN) as pseudo random number generator. We compare the main features of ideal random number generators with our proposed PRNGs. We use a battery of statistical tests developed by National Institute of Standards and Technology (NIST) to measure the performance of proposed HNN and FHNN. We also measure the performance of other standard PRNGs and compare the results with HNN and FHNN PRNG. We have shown that our proposed HNN and FHNN have good performance comparing to other PRNGs accordingly.

Proposal of an Adaptive Neurofuzzy System to Control Flow Power in Distributed Generation Systems

Systems of distributed generation have shown to be a remarkable alternative to a rational use of energy. Nevertheless, the proper functioning of them still manifests a range of challenges, including both the adequate energy dispatch depending on the variability of consumption and the interaction between generators. This paper describes the implementation of an adaptive neurofuzzy system for voltage control, regarding the changes observed in the consumption within the distribution system. The proposed design employs two neurofuzzy systems, one for the plant dynamics identification and the other for control purposes. This focus optimizes the controller using the model achieved through the identification of the plant, whose changes are produced by charge variation; consequently, this process is adaptively performed. The results show the performance of the adaptive neurofuzzy system via statistical analysis.

Thermal Error Modelling of the Spindle Using Neurofuzzy Systems

This paper proposes a new combined model to predict the spindle deformation, which combines the grey models and the ANFIS (adaptive neurofuzzy inference system) model. The grey models are used to preprocess the original data, and the ANFIS model is used to adjust the combined model. The outputs of the grey models are used as the inputs of the ANFIS model to train the model. To evaluate the performance of the combined model, an experiment is implemented. Three Pt100 thermal resistances are used to monitor the spindle temperature and an inductive current sensor is used to obtain the spindle deformation. The experimental results display that the combined model can better predict the spindle deformation compared to BP network, and it can greatly improve the performance of the spindle.

Towards Utilization of Neurofuzzy Systems for Taxonomic Identification Using Psittacines as a Case Study

Demonstration of the neurofuzzy application to the task of psittacine (parrot) taxonomic identification is presented in this paper. In this work, NEFCLASS-J neurofuzzy system is utilized for classification of parrot data for 141 and 183 groupings, using 68 feature points or qualities. The reported results display classification accuracies of above 95%, which is strongly tied to the setting of certain parameters of the neurofuzzy system. Rule base sizes were in the range of 1,750 to 1,950 rules.

An Intelligent Spatial Proximity System Using Neurofuzzy Classifiers and Contextual Information

In this paper, we propose a novel approach to reasoning with the concepts of spatial proximity. The approach is based on contextual information and uses a neurofuzzy classifier to handle the uncertainty aspect of proximity. Neurofuzzy systems are a combination of neural networks and fuzzy systems and effectively incor porate the advantages of both techniques. Although fuzzy systems are focused on knowledge rep re sen ta tion, they do not allow for the estimation of membership functions. Conversely, neuronal networks use powerful learning techniques but are not able to explain how results are obtained. Neurofuzzy systems ben e fit from both techniques by using neuronal network training data to generate membership functions and by using fuzzy rules to represent expert knowledge. Moreover, contextual information is collected from a knowl edge base. The neurofuzzy classifier is used to compute the membership function parameters of the spatial relations fuzzy quantifiers. The complete solution that we propose is integrated in a geographic information sys tem (GIS), enhanced with proximity-reasoning. Our approach is used in the telecommunication domain and particularly in fiber optic monitoring systems. In such systems, a user needs to qualify the distance between events reported by sensors and the surrounding objects of the environment, in order to form spatiotemporal pat terns. These patterns are defined to help users making decisions pertaining to operations, such as optimizing the assignment of emergency crews.

An Intelligent Spatial Proximity System Using Neurofuzzy Classifiers and Contextual Information

In this paper, we propose a novel approach to reason with spatial proximity. The approach is based on contextual information and uses a neurofuzzy classifier to handle the uncertainty aspect of proximity. Neurofuzzy systems are a combination of neural networks and fuzzy systems and incorporate the advantages of both techniques. Although fuzzy systems are focused on knowledge representation, they do not allow the estimation of membership functions. Conversely, neuronal networks use powerful learning techniques but they are not able to explain how results are obtained. Neurofuzzy systems benefit from both techniques by using training data to generate membership functions and by using fuzzy rules to represent expert knowledge. Moreover, contextual information is collected from a knowledge base. The complete solution that we propose is integrated in a GIS, enhancing it with proximity reasoning. From an application perspective, the proposed approach was used in the telecommunication domain and particularly in fiber optic monitoring systems. In such systems, a user needs to qualify the distance between a fiber break and the surrounding objects of the environment to optimize the assignment of emergency crews. The neurofuzzy classifier has been used to compute the membership function parameters of the contextual information inputs using a training data set and fuzzy rules.