P. Thangavel

Bifurcation and Controlling of Chaotic Delayed Cellular Neural Networks

Bifurcation and chaos in one-cell cellular neural network with delay is investigated. Further controlling of chaotic behavior with suitable additional parameter is studied. Later, the study has been extended by adding one additional cell without delay. Further the dynamics of two coupled homogeneous and hetrogeneous cells with delay is also investigated. The numerical global bifurcaiton analysis for all the systems have also been performed.

Parallel algorithms for addition and multiplication on processor arrays with reconfigurable bus systems

Addition and multiplication operations are very important functions in many computer applications; they play an important role in the processing time. They are presented in the paper

Hysteretic Hopfield network with dynamic tunneling for crossbar switch and N-queens problem

An efficient hysteretic Hopfield network with dynamic tunneling is proposed. The hysteretic activation function is used for training. The dynamic tunneling approach is employed to detrap the network from local minima. The network gives better convergence results for the selected problems namely crossbar switch problem with exclusive switching and concurrent switching, and N-queens problem.

Training feedforward networks using simultaneous perturbation with dynamic tunneling

Abstract An efficient technique, namely simultaneous perturbation with dynamic tunneling for training single hidden layer feedforward network, is proposed. A sigmoidal hidden neuron is added to the single hidden layer neural network after training. Then the cascaded network is trained again using simultaneous perturbation. The dynamic tunneling technique is employed to detrap the local minima in training. The proposed technique is shown to give better convergence results for the selected problems, namely neuro-controller, encoder, adder, demultiplexer, XOR and L–T character recognition problem.

Genetic algorithm based watermarking in double-density dual-tree DWT

Watermarking methods in transform domains are usually achieved using the discrete cosine transform or the discrete wavelet transform. In this paper, we develop a technique for optimizing the image watermarking using the genetic algorithm applied to the double-density dual-tree discrete wavelet transform which improves the quality of the watermarked image and the robustness of watermark. We employ, genetic algorithm based embedding schemes namely surrounding mean and zero tree embedding approach. The detection process can be performed without using the original image. The experimental results demonstrate that the watermark is robust to attacks.

Watermarking in Contourlet Transform Domain Using Genetic Algorithm

Contourlet transform is a new two-dimensional extension of the wavelet transform using multiscale and directional filter banks. In this work, we focus on optimizing the image watermarking using the genetic algorithm applied to the contourlet transform which improves the quality of the watermarked image and the robustness of watermark. We employ, genetic algorithm based embedding schemes namely surrounding mean and zerotree embedding approach. The detection process can be performed without using the original image. The experimental results demonstrate that the watermark is robust to attacks.

Simultaneous perturbation for single hidden layer networks — cascade learning

Abstract A simultaneous perturbation approach for cascade learning of single hidden layer neural network is presented. A sigmoidal hidden neuron is added to the single layer of hidden neurons after training until the error has stopped decreasing after a certain limit. Then, the cascaded network is again trained using simultaneous perturbation. Perturbation employed on the weights connecting to hidden neurons are changed to detrap the local minima in training. The proposed technique gives better convergence results for the selected problems, namely neuro-controller, XOR, L–T character recognition, two spirals, simple interaction function, harmonic function and complicated interaction function.

Noise Removal Using Hopfield Neural Network in Message Transmission Systems

In this paper, a novel approach of two-tier encryption is proposed with the removal of noise generated during transmission based on Hopfield neural networks (HNN). The proposed system reduces the complexity of recognition of characters due to external distortion or diffusion. Though there are many error correction and detection codes, these codes request retransmission when there is an error. If the error rate is high the number of retransmissions are high which causes a delay in the process of communicating the information. Moreover the error correction systems can prevent the systems from data loss but will not help in recognition of letters if diffused. When HNN is added to the existing system, the learning ability enables the network to understand or even remember the pattern. But when images of larger size are stored, the network fails to recognize, which leads to further research in this area.

Fragile Watermark for Tamper Detection using Structural Distortion Measure

Verifying the content of digital images or identifying forged regions would be obviously useful when digital pictures are presented as evidence. We present a curvelet logo watermark based on the image fusion approach. The forgery is determined as the one which lacks the logo watermark. We have used structural distortion approach for integrity verification which we have tested on examples of real forgeries. We have also investigated the effect of attacks such as lossy compression, brightness, etc. Region of interest influences our approach to verify image integrity.

A PARALLEL ALGORITHM TO GENERATE N-ARY REFLECTED GRAY CODES IN A LINEAR ARRAY WITH RECONFIGURABLE BUS SYSTEM

A simple parallel algorithm for generating N-ary reflected Gray codes is presented. The algorithm is derived from the pattern of N-ary reflected Gray codes. The algorithm runs on a linear processor array with a reconfigurable bus system. A reconfigurable bus system is a bus system whose configuration can be dynamically changed. Recently processor arrays with reconfigurable bus systems were used to solve many problems in constant time. There already exists experimental reconfigurable chips.