Providing proper secure communications is challenging in Mobile ad hoc networks (MANETs) because of unreliable wireless media, mobility of the nodes and lack of infrastructure. Usually, wired and wireless networks use cryptographic techniques for secure communications. Symmetric and asymmetric cryptography have been extensively used in Ad hoc networks and have both advantages and disadvantages. Any cryptography becomes weak if it has a weak key management which also forms the main aspect for security in MANETs. Thus, the network is susceptible to attacks by malicious nodes and packets are dropped in attacks like greyhole attack. In this paper, it is proposed to evaluate the performance of a network under the impact of malicious nodes. Greyhole attack was simulated as it is difficult to identify them in the network and their behaviour is also highly unpredictable. Simulations are conducted using DSR routing protocol to evaluate the performance degradation of MANET due to malicious node activity..

Keywords— Mobile Ad hoc Networks (MANETs), Routing, Dynamic Source Routing (DSR), Malicious Nodes, Performance Degradation Introduction

In today’s era it is a crucial concern that proper encryption decryption should be applied to transmit the data from one place to another place across the internet in order to prevent unauthorized access.  Image Cryptography is a special kind of encryption techniques to hide data in an image for encryption and decryption of original message based on some key value.  Very few algorithms, provides computational hardness and it makes difficult to break a key to find the original message.  Here RSA algorithm is used to encrypt the image files to enhance the security in the communication area for data transmission. An image file is selected to perform encryption and decryption using key generation technique to transfer the data from one destination to another.
Keywords- RSA Algorithm, Images, Symmetric Key, Asymmetric Key, Key Generation, Prime Numbers,  Hex Code

Speech Recognition technology is a fast growing engineering technology now days. Speech Recognition is the ability for a device to recognize individual words or phrases from human speech. These words further can be command the operation of a system. This paper describes the basics of speech recognition including its types. There would be number of factors that can affect and does matters in speech recognition, few of these are discussed in this paper with application in numerous fields.

Keywords - Speech recognition, voice recognition, speaker identification, speaker verification, isolated words.

This proposed work discusses about the breast cancer detection at the earlier stage on mammogram images using k-means algorithm. This approach has been done in three steps. The primary step is pre-processing, which removes noises in the images. Then canny edge detection is used to detect the edges of images. After finding the edges morphological operation is done to get clear mass. Then original image overlapped with the erode image to get clear view of tumor. K-means algorithm is used to classify the tumor level based on the count of pixel values in the mammogram images. Further the level of the tumor has been analysed and classified. In this proposed work identifies tumor level based on the pixel count as well as it also detects the tumor in the earlier stage itself.

Keywords— Mammogram, tumor, Segmentation, Edge detection, Canny, Median Filter, Thresholding.

Detecting the salient regions on feature points in an image is very fundamental and important task to digital image processing. There are numbers of techniques present now days for recognizing the objects in an image. One of most popular techniques include is feature based method. The paradigm that is followed in the dissertation is to detect the object based on features. The feature based techniques include SIFT, SURF, FAST, MSER and so on. This study presents only one method for scale and rotation invariant features descriptors that is SIFT based on transformation techniques DWT. A  Haar wavelet type of wavelet transform is used in review and these are the forms that are used in many methods of discrete wavelet transform and processing.

Keywords-Digital image processing, object detection, SIFT, DWT, Descriptor.

In present scenario, every business need input regarding user’s metadata related to behavioral and situational while accessing information through World Wide Web and also they just wanted to utilize this information in taking crucial and bold decision as quick as possible.
Specially segment like online shopping, airline reservations, online gaming, online food chains are few examples from lot many businesses, where this kind of information play a vital role and affect profitability up to a great extent. When customer or user visit any web site over WWW, web servers and application servers tracks the audit information in logs.
The growth of web is tremendous as approximately one million pages are added daily. Because of it the web log files are growing at a faster rate and the size is becoming huge. These logs can be used to identify the patterns which can be used by any business. Patterns discovery [5] and analysis process required user sessions as input.
This paper suggests, user session identification process can be improved by combining right available techniques to get more effective and accurate results and using distributed file processing system like Hadoop, the overall processing can be speedup to a great extent.
Keywords - Web Mining, Data Preprocessing, Pattern Analysis, Hadoop, Distributed File System.