Showing papers in "Information Technology Journal in 2011"

A K-Means and Naive Bayes Learning Approach for Better Intrusion Detection

Abstract: Intrusion Detection Systems (IDS) have become an important building block of any sound defense network infrastructure. Malicious attacks have brought more adverse impacts on the networks than before, increasing the need for an effective approach to detect and identify such attacks more effectively. In this study two learning approaches, K-Means Clustering and Naive Bayes classifier (KMNB) are used to perform intrusion detection. K-Means is used to identify groups of samples that behave similarly and dissimilarly such as malicious and non-malicious activity in the first stage while Naive Bayes is used in the second stage to classify all data into correct class category. Experiments were performed with KDD Cup '99 data sets. The experimental results show that KMNB significantly improved and increased the accuracy, detection rate and false alarm of single Naive Bayes classifier up to 99.6, 99.8 and 0.5%.

Adaptive image steganography based on optimal embedding and robust against chi-square attack

Abstract: A real-life requirement motivated this case study of secure covert communication. Steganography is a technique used to transfer hidden information in an imperceptible manner. We proposed a novel approach of substitution technique of image steganography. The proposed method is flexible on size of secret message and allows us to embed a large amount of secret messages as well as maintaining good visual quality of stego-image. Using this method, message bits are embedded into uncertain and higher LSB layers, resulting in increased imperceptible and robustness of stego-image. Results show that the proposed algorithm provides large embedding capacity without losing the imperceptibility of the stego-image. The algorithm is also robust against Chi-square attack.

High watermarking capacity based on spatial domain technique

Abstract: Watermarking capacity refers to the amount of information we are able to insert into the image. Low signal to noise ratio is a phenomenon of watermarking channels, which severely limits the capacity. The aim of this study is to develop a digital watermarking model, which can find out the possibility to embed maximum amount of data in an image without degrading the quality of watermarked image. In this approach, the host image will be partitioned into non-overlapping blocks and passing an imaginary plane in the three critical pixels. The characteristics of this plan should not be changed after embedding message; then the same characteristics will be used to evaluate the embedded capacity in the extracting module.

A study on radar emitter recognition based on spds neural network

Abstract: With the rapid development of new type radar, it's more difficult to recognize radaremitter signal. Trying to improve radar emitter recognition rate and reduce the processing time are the core of the research in this area. A new efficient radar emitter recognizer using Single Parameter Dynamic Search (SPDS) algorithm is proposed in this study. The SPDS algorithm is a modified algorithm of BP network and it only permits one of all parameters in the network to change during each epoch of searching step for parameters which guarantees to carry out the exact one-dimensional search. This algorithm can overcome the giant limits of BP algorithm such as local minimum and long training time. The effectiveness of SPDS algorithm is shown in simulation results. Compared with the classical neural network algorithms, the radar emitter signal can be recognized more accurately with the SPDS algorithm and the learning speed is improved greatly. The recognition rate is close to100% under the condition of enough training times and uncomplicated data. 2011 Asian Network for Scientific Information.

ISB watermarking embedding: A block based model

Abstract: Many watermarking methods have been developed with different methodological complexity levels. Each of these methods tries to reduce exposure in different attack. In this study, the ISB watermarking method was implemented based on average of block of pixels together in order to improve the watermarking method to be more resistant against attacks than a single pixel. The results show that the quality of the images is suitable for the application of the proposed method, based on any size of block. In additional to that the robustness has been improved by increasing the size of the block for all the attacks, including the geometric transform attacks, although they were not improved when the method was applied based on only one pixel.