IJEA - Volume 4 Issue 5 (May 2015)


Sr/No/ Title Pdf
1 Design, analysis of high directivity hexagonal cut I shaped antenna and simple patch antenna using DNG based metamaterial
Hitesh Joshi, Vivek Bajpai, Umesh Barahdia

Abstract-Now days technology has evolved constantly over the year and antenna play an important role in wireless digital technology .several type of antenna is available in market and different antenna have different uses like monopole, yagi-uda antenna and micro strip patch antenna MPA is one of them its low profile ,low cost, high configuration easy fabrication and it is designed with artificial meta material based structure .these These advance material depicts unusual electromagnetic properties which are not available in nature, such as negative permittivity (εr) and permeability (μr). Due to its amazing feature, it has the capacity to enhance antenna characteristic rapidly. We have design two different type of antenna which is operate at 5.2GHz frequency and result has been simulated on CST studio 2010, 5.2 GHz is large frequency range which can be cover more band area and directivity of simple patch antenna is 7.176DBi next we designed hexagonal cut I shaped antenna the directivity of this antenna is improved as 9.146 DBi. Return loss of SRMPA is -37.48db ,proposed a hexagonal cut I shaped antenna reduce return loss 19.19 db this antenna take less area then other antenna and so directivity and return loss of proposed antenna is 20% to 25% batter then SRMPA.VSWR of SRMPA and proposed antenna is also (<1). Design has been made of low cost FR4 material its 1.6 mm from ground plane.

2 Detecting Temporary and Permanent Faults in NOC Using FTDR Algorithm
Manjula G, Kayalvizhi N

Abstract:-Decreasing size of integrated circuit leads to increase in susceptibility to temporary and permanent faults. This project proposes a fault - tolerant solution for a buffer less network-onchip, with on-line fault-diagnosis mechanism is used to detect both temporary and permanent faults. Hybrid automatic repeat request of forward error correction and link-level error control scheme are used to handle temporary faults and a reinforcementlearning based fault-tolerant deflection routing (FTDR) algorithm is used to tolerate permanent faults without deadlock and live lock. A hierarchical routing-table-based algorithm (FTDR-H) is also presented to reduce the area overhead of the FTDR. By backtracking method, Reconfigurable FTDR algorithm is used with to clear the seek structures and compute the new path with dummy node. Routing table is updated to analyze the data under real application workloads. BCH code is used to detect the multibit errors. Modelsim simulator is used to verify an existing and proposed method.

3 Web search by inferring user interest based on Personalized search and implicit indicators
A.Sangeetha, T.Nalini

Abstract:-In the context of the personalized information retrieval, this paper for inferring the user’s interests based on personalizes web search .This paper aims to collect implicitly the relevant documents, during the user’s search session by inference of the degree of interest based on a combination of some relevant implicit indicators. The main advantage of our approach is that it does not require any log file. Indeed, calculates the degree of interest in real time and decides if the web page is quite relevant to feed the search history dimension.

4 An efficient Prototype for Securing Image using Chaotic Map Approach
Prathibha M, Manpreet Bagga, Baljit Singh

Abstract:-Image security is one of the active research topics from last decade. Our investigated has shown that although there are various cryptographically-based methods exists to safeguard the image either during transmission or for storage purpose, but none of them are 100% robust. All existing cryptographic techniques have advantages as well as limitations, which is more inclined towards complexities and vulnerable key management system. Hence, this paper attempts to introduce a non-cryptographic approach for performing encryption using chaos theory. The system takes the input of original image and forged image, where our permutation techniques results in formation of confusion matrix. Each steps being monitored using correlational and histogram analysis, the multi-level encryption is performed using the secret key from the unit components of the standard logistic map. The outcomes of the system are evaluated for various case studies to show that it is highly resilient against attacks.

5 Analyzing Breast Cancer data using different Classifiers
Ch. Bindu Madhuri

Abstract:-The maximum common malignancy among women in all over the world is Carcinoma of the breast. The major killer cancer disease in females is Breast cancer. Globally in emerging regions, where the early diagnoses of cancer facilities are not available, prediction is even worse. The main objective of this study is to create awareness among people in World Wide about this disease and the risk factor of the disease, which gives indication to early detection and thereby decrease the morbidity and mortality. Breast cancer awareness is very low among men, even in metropolitan city like Mumbai, Bangalore, etc. In the absence of breast cancer screening clinics, there is a dire need to take measures to improve breast cancer awareness in both women and men so that they can play a role in early detection of this disease and thereby improve the outcome in this disease. There are a number of risk factors associated with the disease. There is an inter-individual variability in breast cancer risk that attracts that is of anthropological interest. Machine Learning has become a popular technology in current research and for medical domain applications. This paper gives the current overview of use of different classification techniques namely ART1, LVQ are the methods which are the Machine Learning techniques on breast cancer data. This paper presents a diagnosis system which detects the cancer stage using clustering techniques called Adaptive Resonance Theory (ART1), which is a special case of unsupervised learning Algorithm. A vigilance parameter in ART1 used for the stopping condition and its help in manipulating the accuracy of trained network. In this paper we develop a system for diagnosis, prognosis and prediction of breast cancer using Artificial Neural Network (ANN) models. This will assist the doctors in diagnosis of the disease. We implement one of the models of Neural Networks namely learning Vector Quantization & Experimental results show that LVQ shows the best performance in the testing data set. The high accuracy of the LVQ against ART1 indicates its better ability for solving the classificatory problem of Breast Cancer diagnosis.

6 Policy Based Data Sharing In Encrypted Cloud Storage Services
Gayathri.V.P, Jayakumar.T.P, Karunakaran.S

Abstract:-Data sharing is an important functionality in cloud storage. Data owner maintains the shared data under cloud data centers. Semi-trusted third party servers are used to manage medical records. Personal Health Record (PHR) is used to manage patient personal and diagnosis details. PHR service allows a patient to create, manage and control their personal health data in one place through the web. Patient data can be shared with healthcare providers, family members and friends. Sensitive details are managed in medical records. Data owner decides the access privileges for the medical records. The key aggregation system is divided into five major steps. They are Setup, KeyGen, Encrypt, Extract and Decrypt. Patient controlled encryption scheme is designed using Key Aggregate cryptosystem (KAC). The key aggregate cryptosystem is enhanced with boundary less ciphertext classes. The system is improved with device independent key distribution mechanism. The key distribution process is enhanced with security features to protect key leakage. The key parameter transmission process is integrated with the ciphertext download process.

7 SVM Noise Optimization Algorithm for Video Background Subtraction
Audhavan.S, Amsavalli.K, Rajini Girinath.D

Abstract:-The important feature of detecting the moving objects in videos is Background subtraction. The main process involved in the background is the foreground detection. However, many algorithms usually neglect the fact that the background images consist of different objects whose conditions may change frequently. In this paper, a hierarchical background model is proposed based on segmenting the background images. It first segments the background images into several regions by the Support Vector Machine. Then, a hierarchical model is built with the region models and pixel models. The region model is extracted from the histogram of a specific region which is similar to the kind of a Gaussian mixture model.

8 Application of Image Mining for Medical Imaging
Deepika Kishor Nagthane

Abstract:-Image mining is synonymous to data mining concept. It is important to first understand the data mining concept prior to image mining. Data mining is a set of techniques used in an automated approach to exhaustively explore and establish relationships in very large datasets. It is the process of analyzing large sets of domain‐specific data and subsequently extracting information and knowledge in a form of new relationships, patterns, or clusters for the decision‐making process. The approaches of data mining are association, sequence-based analysis, clustering, estimation, classification, etc. Algorithms and models are then developed based on the dataset type to perform the data mining. These approaches of data mining are applied to images.

9 Skin disease diagnosis using Artificial Neural Network: A Review
Priyanka S.Biradar, S.N.Patil

Abstract:-The artificial neural network is a branch of artificial intelligence which is made up of small processing units called as neurons. These artificial neurons are conceptually similar to that of biological neurons which are the major part of human nervous system. The artificial neurons are having the property to interact in the same way that the biological brain would do this. They have the property to learn patterns of symptoms of particular diseases and therefore they can provide faster diagnosis than a human physician. Here we discuss the applications of artificial neural networks in diagnosis of skin disease. Skin is one of the most difficult terrains to study because of the uneven surface and presence of hair. Human skin consists of definite pigment structure. The slight variation in the structure changes the colour of the skin. Therefore by analysing the colour and texture a lot of information can be obtained regarding various skin diseases. Here we discuss various techniques that deal with skin disease diagnosis. The focus of our work is on skin disease diagnosis using artificial network which provide good accuracy in diagnosing particular disease.

10 Social Networks for p2p Content- based file sharing in Disconnected MANETs
Nisarga Patel N, Mohammad younus, Chaya P

Abstract:-Peer-to-Peer (P2P) file sharing methods in mobile ad hoc networks (MANETs) is classified into three groups: floodingbased, advertisement-based, and social contact-based. The first two groups of methods can easily have high overhead and low scalability and are mainly developed for connected MANETs, in which end-to-end connectivity among nodes is ensured. The third group of methods adapts to the opportunistic nature of disconnected MANETs but fails to consider the social interests of mobile nodes, which can be exploited to improve the file searching efficiency. The proposed approach is a P2P content based file sharing system, namely SPOON, for disconnected MANETs. The system uses an interest extraction algorithm to derive a node’s interests from its files for content-based file searching. For efficient file searching, SPOON groups commoninterest nodes that frequently meet with each other as communities. It takes advantage of node mobility by designating stable nodes, which have the most frequent contact with community members, as community coordinators for intracommunity searching, and highly mobile nodes that visit other communities frequently as community ambassadors for intercommunity searching. An interest-oriented file searching scheme is proposed for high file searching efficiency. Future work will incorporate method to determine appropriate thresholds in SPOON and the affect of the file sharing efficiency, and also how to adapt SPOON to larger and more disconnected networks.

11 Identifying Packet Hackers in Wireless Sensor Network
Kavana M D, Ravi Kumar V G

Abstract:-In wireless sensor system, detecting bundle droppers and modifiers is a complex assignment. Two algorithms are proposed in order to address this problem. Firstly, Node Categorization algorithm to identify nodes that are droppers or modifiers for sure or suspicious droppers or modifiers. Secondly, the sink will periodically run heuristic ranking algorithms to identify most likely bad nodes from suspiciously bad nodes. The proposed system is more useful in detecting attackers such as packet droppers and modifiers in wireless sensor network. It also determines the packets that have been dropped/modified. It enables safe communication by checking authentication of each node that is involved in data delivery. It helps to protect the confidential data from unauthorized access. It is useful in the area of business, military and other applications in which data security is a major issue.

12 A Proactive Approach for Detection and Prevention of DDoS Attacks
Anusha .N, Manjunath S

Abstract:-Distributed Denial of Service (DDoS) is a critical threat to the network users. It is an attack designed to render a computer or network incapable of providing services to the customers. DDoS attack uses many compromised attacker to launch these attacks and it is distinguish by an explicit attempt by attackers to prevent legitimate users of a service from using that service. Detection and prevention of DDoS attacks remain a major problem when it comes from high-bandwidth connection. Detecting this type of attack is a real challenge, and it is need to protect the network users and its valuable infrastructure resources from these attacks. A trivial mechanism is proposed which uses credence score to differentiate users and attackers. The system administrator provides security for the registered customers. The core of the proposed system is intrusion prevention system (IPS). The IPS forms a virtual trapeze around the host they are going to protect. A true traffic generation in the network to show the flooding DDoS attacks is expensive and high–level. So it is shown using file upload and downloads criteria. In the present work scarcity of resources is reduced using enhanced rate limiting mechanism and certain protection rules like subscription rule, allow rule and block rule. By using this technique the attackers can be blocked and normal users can access the resources with risk free state. The proposed rule structure will provide protection for the users from this DDoS attack.

13 Recommendation System Based on Combined Features of Hybrid Algorithm and Keyword Based Filtering
Vasundhara M S, Gururaj K S

Abstract:-Providing efficient recommendations for items in online shopping websites has become a Challenging task. People prefer to go for online shopping rather than going out and shopping for themselves as it provides an easier and quicker way to purchase items of their choice with quick transactions. Recommendation systems are widely used for recommending products to the end users in their interested fields. Providing recommendations based on the browsing history may or may not be of user’s interest and also the quality of the recommended items may not be guaranteed. The main factor of recommending sites is customer satisfaction. The recommendations provided by websites should meet the maximum level of customer satisfaction. This paper aims to find the efficient approach using Data Mining concept called Association Rule Mining with content-based and collaborative filtering in order to recommend the only relevant information to the buyers. The items are recommended for the buyers based on the content of buyers past buying history and opinion of other users in order to find out the quality of the item. Association Rule mining is used for extracting the useful information from the transaction dataset and produce efficient and effective recommendation based on buyer’s interest thus satisfying the buyer in better way. Similarly for music and videos the recommendation is based on the keywords set by the Business Entity using Feature-based recommendation system.

14 Performance Evaluation of MSQM and RED Queue Management in DiffServ-Networks
Goree. Narsimhulu, D. Sreenivasa Rao

Abstract:-In this paper Modified Sized-based Queue Management as a dropping scheme that aims to fairly prioritize and allocate more service to multimedia traffic over bulk data like FTP as the former one usually has small packet size with less impact to the network congestion. In the same time to guarantee that this prioritization is fair enough for both traffic types, on the other hand the total link delay over the congestive link with the attempt to improve this congestion as much as possible at the by function of early congestion notification. The M-SQM scheme has been carried out using network simulator NS2.34. The simulation parameters were link delay and received packets. The simulation parameters compared with other AQMs (RED and RIO) the M-SQM scheme has given better performance than other AQM.

15 Stable and Secure Routing Strategy for Ad hoc Wireless Networks based on DSR
Nirav Bhatt, Dhaval Kathiriya

Abstract:-With ad hoc networks nodes are portable. It is not having any established infrastructure for communication. So the nodes if want to communicate, then ahead data to the other nodes available in the range. So we are going to proposed one routing strategy supports with Dynamic Source Routing with improving performance in terms of energy saving and provides security in terms of trustiness. Because lack of infrastructure in such networks power consumption is measure issue. Nodes are poor in terms of energy as well as security. In such networks nodes can perform all the functions with selfish in nature some times to save energy. So we are trying to modify original DSR and provide some mechanisms for saving residual energy and also trying to provide security mechanisms in terms of trusted routes. In modified protocol routes are established for transmission of data packets are with highest energy level and highest trusted nodes. According to the distance of two consecutive hopes they can also regulate transmission power and then established routes. As during transmission if minimum power is used then residual energy of nodes can be saved. For generating the results we have used NS-2.34 simulators. Through that we are succeed to improve performance compare to original protocol.