JCT - Volume 4 Issue 9 (September 2015)

 


Sr/No/ Title Pdf
1 Progressive Packet Arrival Handling In DTNs through General Policies
M.Sanghamitra Devi, K.Sundara Rao, Kalyanapu.srinivas

Abstract- In this paper we propose some generalpolicies which are used to handling the optimal packet transmission in DTNS when packets arrived progressively at the source. Here we are also using the network- codes for achieving reliable packet transmission in DTNS..

Download
2 Design of Digital Low Power FIR Filter with Serial Architecture (MPF)
Kunal P. Raval

Abstract-This paper is about designing a 15 tap 8-bit FIR filter using Direct form II. Serial Architecture is used for Multiplication and Accumulation. Entire Design is created in structural manner using verilog models from fsa0m a library. Delay b a l a n c i n g technique is u s e d for reducing glitches in Multiplier.

Download
3 Image processing techniques in cancer diagnosis: challenges and perspective
Gowthami.S, Indrani.M

Abstract-Cancer has become one of the biggest threats to human life for many years, and is expected to become the leading cause of death over the next few decades. A mammography exam, called a mammogram, is used to aid in the early detection and diagnosis of breast cancer facilitated with digital mammography can increase survival rate and chances for patient's complete recovery. Cancer detection system based on the analysis of Mammogram, which can be used by the doctors to decide whether further biopsy is needed, or not. The system will act as a decision support system and uses image processing techniques to analyze the mammograms. The application takes the input as mammogram image and reports the presence of suspicious region. This paper poses overview of different algorithms used for different kind of cancers in the human body.br />
Download
4 Design & Development of a Trustworthy and Secure Billing System for Cloud Computing Architecture
G.Manoj Someswar, Firdous Rehana, Mohd Abdul Kareem

Abstract-Cloud computing is an important transition that makes change in service oriented computing technology. Cloud service provider follows pay-as-yougo pricing approach which means consumer uses as many resources as he need and billed by the provider based on the resource consumed. CSP give a quality of service in the form of a service level agreement. For transparent billing, each billing transaction should be protected against forgery and false modifications. Although CSPs provide service billing records, they cannot provide trustworthiness. It is due to user or CSP can modify the billing records. In this case even a third party cannot confirm that the user’s record is correct or CSPs record is correct. To overcome these limitations we introduced a secure billing system called THEMIS. For secure billing system THEMIS introduces a concept of cloud notary authority (CNA). CNA generates mutually verifiable binding information that can be used to resolve future disputes between user and CSP. This project will produce the secure billing through monitoring the service level agreement (SLA) by using the SMon module. CNA can get a service logs from SMon and stored it in a local repository for further reference. Even administrator of a cloud system cannot modify or falsify the data.

Download
5 Adaptive head lighting system with lowering high beam of other road vehicle
Supreeth.A, Hemanth kumar.U, Srivatsav.B.V.S.S, Likith.H.S

Abstract-Present vehicle illumination systems are facing a number of problems such as inadequate visibility, failure in switching over headlights of other road participant and more.one of the main reasons for these problems is that illumination system is static they do not change in relation to varying road, vehicle, traffic conditions. This paper presents feasibility investigation of adaptive head lighting system of vehicle which will switch the headlight of other road participant to low beam. which can be achieved by using IC 555 timer to generate pulse of frequency 1.5Khz, SPDT relay as electrical operated switch, BC547 transistor as a virtual switch, CdS photocell as photometric sensor and thin small outline package (TSOP) for IR remote control.

Download
6 College Campus Radio
Deepak B. Kadam, Vinay S. Suryawanshi, Sameer R.Jahav

Abstract-In this paper we will make College Campus Radio. This radio works like normal radio transmitter. The range of this transmitter kit is less up to 500 meter. This radio is consist of FM transmitter for transmission of audio signal ,usb kit for playing music from pen drive This project is use for small range FM communication for educational purpose ,low power radio station, community radio. fan will be turned ON automatically .As soon as the temperature lowers than the set value the fan will be turned OFF accordingly. The System is fully controlled by ARM7 LPC2148.For wireless controlling the temperature we will use pair of XBee one at the sensor end and other at the user end. The current temperature will be continuously displayed on users PC screen and user will be able to change the threshold value for control purpose using serial communication.

Download
7 Security of Payment Gateway in Online Transactions
D. Madhavi

Abstract-In this paper, Security protocols of payment gateways are discussed in online transactions such as E-Commerce operations. The goal is to achieve the security through secure payment gateway protocols. To achieve this security the Ecommerce web designers not only uses payment gateways but also uses security protocols in order to perform the transaction of payment successful by the end users or customers. Generally Certificates are used by the payment Gateways to maintain security of transactions. How those certificates are used in the payment gateways and which protocols are used are mentioned.

Download
8 Design and Analysis of Message Authentication for Source Privacy in WSN by Using Elliptic Curve Cryptography
Mopada Vishnu Priya Darshini, P.Chaitanya Vandavasu

Abstract-Network is an interconnection of several nodes which are connected all together in order to communicate with each other. Generally networks are categorized into two types based on the usage like wired networks and wireless networks. Wireless sensor networks are the nodes which are spatially distributed to monitor the physical and environmental changes. All wireless sensor networks are built of "nodes" from a range of few to several hundreds or even thousands, where each node is connected to one sensor. There were some issues while using wsn like data mislead or un-authorized data access, this was happened because of its open nature. So in this paper we have implemented a new concept called as Hop-by-Hop message authentication where server will generate a message authentication code for the data which is to be sent as well as for the users who participate in routing. Almost there was many message authentication schemes implemented in wsn by using public key or private key cryptography, but they all failed in achieving the security for communication data. So in this paper we have implemented a scalable authentication scheme based on elliptic curve cryptography (ECC). In this algorithm there is no limit of any threshold function, it can allow any node to transmit an unlimited number of messages without suffering the threshold problem. As an extension we have implemented MD5 algorithm in order to generate the hash codes for the each and every individual node as well as routers which participate in the routing. This MD5 will allow a data to be received only by the valid nodes, if any unauthorized node try to access the data illegally it can’t be view in their node. By conducting various experiments on our proposed system we finally came to a conclusion that this will give high level of security for the data which is transferred in WSN especially source node can achieve high level of data security.

Download
9 Design and Analysis of Pattern Classifying Methods for a Secure Message Communication
Sekharamahanti S Nandini, B.Shalini

Abstract-Now a day’s security plays a very important role in almost all fields like health, banking, shopping, business, and government so on. As security has highest privilege there were a lot of adversary users who try to access information of secured users illegally through various hacking codes. Pattern classification systems are commonly used in many of the security applications, like biometric authentication systems, Intrusion detection systems, and spam filtering systems, in which all the input data can be purposely modified by human beings in order to break the current operation. As this was available till now there was no classical methods adopted this new pattern classification. So in this paper we take pattern classification into consideration and try to evaluate the following open issues like: how to evaluate the security of pattern classifiers at design phase. In this proposed paper we try to implement the earlier three security levels of pattern classification approach as well as additional concept like Monitoring the activities of users, like their login and logout details ,availability of network IP address and the details like their upload and download also. So by implementing this new concept as an enhancement we try to prove that our proposed application is very secure compared with all classical systems. Our simulation results clearly tells that by using this new framework an user can be free from adversary attack during his data storage as well as data retrieving.

Download
10 Performance Analysis of IVC Networks for a Secure Multihop Broadcast Solution
Gudepu Lavanya, P.Chaitanya Vandavasu

Abstract-Now a day’s it is very important challenge for efficient data delivery for vehicular networks. This is mainly due to continuous network disruption that occurs in the vehicular networks, which in turn leads to change of topology dynamically. So in order to achieve this efficient data delivery we came with a new concept called as Inter Vehicular Communication (IVC) to significantly contribute to traffic safety and potency. To this extent, there are several potential IVC applications that share the common information like speed of vehicle, direction of vehicle, and position of vehicle in multihop message propagation. Along with this primary information it should also capable to identify various security attacks that occur during data exchange and it should able to provide counter measure for those attacks. In this IVC ,a plenty of malicious vehicles forever attempt to inject incorrect information into the inter vehicle wireless links, that ends up in plenty of cash losses or to any other style of adversarial stinginess in terms of mis-routing the vehicles. So in this paper, we mainly developed a fast and secure multihop intervehicular network for communication no data loss. Our Simulation results on this proposed system we finally came to a conclusion that this is the first time to implement such a new system which can give counter measure quick and fastly for the adversary vehicles which reside in IVC.

Download
11 Design and Analysis of a Novel Packet Hiding Mechanism to Avoid Jamming Attacks
B.Sameera, M.V.A.Naidu

Abstract-Now a day’s security plays an awfully vital role within the field of data technology. This is often owing to as a result of all the info that is to be transferred from one location to alternative is transferred through network. As security plays an important role plenty of users attempt to gain the precious info illicitly through varied hacking techniques. One amongst the fraud techniques is electronic countermeasures. Electronic countermeasures is that the external thread model for accessing the data within the network while not knowing to the network users and check out to misuse the info in middle of information communication. This principally happens owing to the explanation like wireless networks square measure open supply in nature. In this paper we've got enforced a brand new conception to avoid the info access from jammers called channel encryption wherever the plain text that ought to be transferred are bornagain within the type of binary format and interleave mechanism that shuffles the positions of packets bits. Thus by exploitation this combined mechanism we will} able to forestall the packets from jammers. By conducting varied experiments on our projected technique we have a tendency to finally come to a conclusion that this projected technique achieves high level of security on network information compared with primitive scientific discipline algorithms.

Download
12 Fast Dynamic Algorithm for Sequence Based Alignment in Detection of Genes
Nancy Anurag Parasa, Dr. G.Lavanya Devi

Abstract-The difference recognition in the genomics is the major problem that is widely encountered in bioinformatics. Here sequence alignment plays a major role to sort out this problem of pattern recognition. For this any DNA, RNA or protein sequences are used as source for the difference identification. This uses many algorithms present in computational biology called bioinformatics algorithms. Here we come up with a new implemented algorithm called Fast Dynamic Algorithm for Sequence Based Alignment (FDASA) which resolves the problem of genome difference identification. In this paper FDASA is implemented by building scoring matrices which is represented by rows and columns of M and N rows. Which is calculated as M*N where M represents first sequence order, N represents second sequence order. This matrix is filled up with three main diagonals then the unused data is left which reduces the memory and space. Thus making the algorithm efficient. Finally this brings out an optimal solution which results in high performance and execution time is reduced to its optimal level. By conducting various experiments on these 3 algorithms by taking various input species, we finally came to a conclusion that FDASA is utmost understood and easy to identify the genes of various species when compared to Needleman – Wunsch and Smith – Waterman algorithms.

Download
13 An Approach for Iris Plant Classification by Exploitation of Neural Network
R. Nagalakshmi, Dr. G. Lavanya Devi

Abstract-In this paper, the difference of network weights using Particle Swarm optimisation (PSO) was planned as a mechanism to boost the performance of Artificial Neural Network Algorithm (ANN) in finding the species of plant from IRIS dataset. We know that classification is one type of machine learning technique that will predict cluster membership for information instances. To change the matter of classification neural networks square measure being introduced. In this we mainly concentrate on IRIS plant classification exploitation Neural Network. Classification of IRIS information set would be discovering patterns from examining floral leaf and floral leaf size of the IRIS plant and the way the prediction was made up of analysing the pattern to make the category of IRIS plant. By exploitation this pattern and classification, in future forthcoming years the unknown information are often foreseen additional exactly. Artificial neural networks are with success applied to issues in pattern classification; perform approximations, optimisation, and associative reminiscences. During this work, Multilayer feed- forward networks square measure trained exploitation back propagation learning algorithmic program. By conducting various experiments on the proposed PSO we finally came to an conclusion that this proposed algorithm is best suited to find out the species name of plant with the help of width and heights of sepal and petal of a plant. For this purpose we have taken a set of petal and sepal width and length values on a data set called as IRIS.

Download
14 Design and Analysis of a Secure Cloud Storage by Using Single Instance Storage
Kothapalli Sameer, Mr. Shafiulilah Shaik

Abstract-Cloud Computing is one among the apply of employing a network of remote servers hosted on internet to store,access,retrive knowledge from remote machines not from native machines. As the knowledge is going to be kept on remote server, the user can retrieve the information from that server at the time of want. So for that knowledge storage and access the user ought to pay the amount on rental basis, that is the most reason why the cloud is additionally called PAUZ (I.e. Pay As you Uze).As knowledge is keep in remote system we'd like to deploy a facility/service to avoid the duplicate knowledge to not be reside on it server systems, if this is often not avoided the user ought to pay excess quantity for that same duplicated storage. So inorder to avoid this knowledge duplication we'd like to use a replacement principle referred to as knowledge Deduplication. This is one among the simplest knowledge compression technique that was used for eliminating the duplicate copies of perennial knowledge and this was wide employed in recent cloud storage. As this knowledge deduplication technique is employed inorder to avoid the duplicate records to not be keep onto the information base, we need to guard the information confidentiality conjointly. Inorder to guard the confidentiality of sensitive knowledge, the confluent cryptography technique has been projected inorder to cipher the information. In this paper we've enforced cryptography technique to form plain text into cipher so store onto the cloud server. The non-public cloud server can generate the priveleges for the user like scan, write, update, modify and delete regardless of the user need to possess access privelege.If the user gets the access priveleges then only solely he can operate the hidden information, as well as inorder to access the information he can get associate degree token id from the cloud service supplier. We have conducted numerous experiments on this projected design and incurs bottom overhead compared to traditional operations.

Download
15 CICP: Performance Analysis of Sensitive Data Storage by Using Captcha Using Image Click Points Technique
Koppula Sri Siri, S. Neeraja, Ramesh Challagundla

Abstract-Now a day’s security plays a prominent role in each and every part of human life. As security plays a major role in all domains like health, banking, schools, finances, shares and so on. There were a lot of third party persons who try to access the sensitive information of others illegally. So this type of problem leads to a major problem like attack or threat. So till now there was many security primitives available inorder to give security for the sensitive data. So in this paper we have implemented CAPTCHA as main security primitive for login purpose as well as 2 Click algorithms in order to give more security for the data which is to be hidden safely. This 2 click algorithm is majorly used on image for choosing randomly 2 click points in which each and every point takes 2 values like one from xcoordinate and other from y-coordiante.So these two clicks give more security for the proposed system. So by applying these two concepts parallel we call these both as Captcha as Image Click Passwords (CICP) .CICP is termed with the facility like captcha along with Image click points scheme. As an extension we have also implemented another concept called as secret key in which user chooses a secret key randomly by his own during the time of registration and the user should substitute the same secret key at the time of login, this will be failed if the user substitutes it wrongly. And as an extension we have also implemented forget password, forget key and also an additional functionality like Auto Block of the user accounts (I.e. if the user gives invalid inputs for more than 3 times, the account will be automatically blocked).By conducting extensive experiments on our proposed CICP we finally came to an conclusion that this was the technique which was not yet implemented so far in providing security for the sensitive data which was stored on remote servers.

Download
16 Effect of Machining Parameters on Surface Roughness for Aluminium 5083 in CNC Milling Using Multiple Regression Analysis
Metta Jagadish Kumar, N. Satya Kranthi Kiran

Abstract-The main objective of this work is to investigate the effect of machining parameters on surface roughness of Aluminum 5083 in CNC milling machine with carbide tool. The three machining parameters namely the spindle speed, feed rate and depth of cut are selected as input parameters. The effect of each parameter and interaction effect of these parameters on surface roughness is analyzed by parametric Analyses of Variance (ANOVA). Surface roughness prediction model is established with Multiple Regression Analysis for predicting the minimum surface roughness of Aluminium 5083 in CNC milling operation. Surface roughness prediction model is tested with the experimental surface roughness values (Ra). The average percentage deviation of models is 11.17% and is found to be in acceptable limit.

Download
17 A Novel Attribute Based Encryption Mechanism for Secure Data Storage in the Cloud
Kusuma Kari, Meduri Kamalakar, Ramesh Challagundla

Abstract-Cloud computing is one of the practices of using a network of remote servers hosted on internet to store and retrieve data from remote servers not from local machines. As the data will be stored on remote server, the user will retrieve the data from that server at the time of need. As the cloud server is used to store the data remotely on its own server rather than storing it on users system, data should be shared securely for the cloud users. Till now there was no facility of giving data securely through the cloud server as the data is been saved in plaintext only rather than storing in an encrypted manner. So in this paper for the first time we have implemented the encryption of data before it is saved into the server and the same data can be viewed in plain text if the authorized user downloads the file. If any un-authorized user downloads the file without having the permission from the cloud admin, he/she can’t able o access the file in a plain text manner. In order to provide such a security for the data we have implemented a new key aggregate system for encrypting the data and we have implemented ABE algorithm in order to assign aggregate keys during data sharing. Generally aggregation is a process of joining or combining more than one keys into a single key, so in this paper the user have the ability to give the public key of his own during the file upload and the cloud server will generate a secret key dynamically at the time of uploading the data into the cloud. So if any user who want the retrieve the data from the cloud server he need to join that public key as well as secret key which is generated dynamically for each and every file upload. If both the keys are aggregated into a single key then only the data will be decrypted if not the data will remain in encrypted manner. As an extension we have also implemented this proposed concept on Drop Box, where as a proof of concept we have used drop box as the live cloud server for storing and retrieving the data. Also as an extension the aggregation of key is also monitored by drop server only dynamically.

Download
18 Epilepsy Detection by Computing Wavelet Coefficients and Neural Network Model
Deepali Khandekar, Prof . Patil S.P

Abstract-Electroencephalogram is a very complex signal. It is one of the important domains and source of information which is often is used to study brain function with neurological disorders [1]. Evolution of the crisis is a dynamic and not fixed process. The EEG signals are composed of multiple frequencies. Detection of epilepsy takes longer due to the long recording of EEG. Activity detection registered epileptic EEG segments is crucial for the localization and classification of epileptic seizures. The problem of detection of attack is basically a classification among the signs of normal EEG and convulsions. The whole process usually further divided in a series of small segmentation, extraction of characteristics and classification of the signal parameters. This document illustrates the use of characteristics of wavelet based for the classification between the signals of normal EEG and convulsions. They are then classified feature vector belonging to signal segments separated by a competitive neural network. Normal signals and convulsions were classified using a neural network with high accuracy classifier. The document provides a comparison of classification results using different methods of extraction of features most appropriate for the detection of EEG signal components.

Download
19 A New Method for Preventing Posting of Offensive Messages in any OSN Networks
Sura Srikanth, Ganta Santosh Kumar

Abstract-Today, Online Social Networks (OSNs) is ubiquitous in interactive medium for communication, sharing and publishing huge information. As the usage of social network became popular, the major setbacks in OSNs are increased. One of the primary issues of these days in OSNs is to avail the users for preventing redundant posts on their private space. A novel concept of filtering protocol is proposed for this problem to unswerving control on the redundant posts on their walls, for OSN users. This protocol was put into action by means of automatic identification of abusive words from the entire message by using a Machine Learning (ML) based on soft classifier algorithm which blocks the content by labelling messages based on the category.

Download
20 Research Issues in Knowledge Management System (KMS) and Framework of the Research Study in IT Based Competency Building
Bechoo Lal, Chandrahauns R.Chavan

Abstract-In this research paper, the researcher identified the current research issue of knowledge management system in IT based competency building, the research also focused on research issues on knowledge creations, storage and retrieval, transfer, and management initiative in IT based competency building. The researcher designed a framework of research study which is significantly predicted to the dependent variables of business optimization services. There are also some moderating variables which is in between dependent and independent variables makes a bridges between the controlling factors of knowledge management towards the business optimization.

Download
21 A Novel Search Based Face Annotation (NSBFA) Algorithm for Annotating Weakly Labeled Images
SDasari Viswanadham, Pydipala Laxmikanth

Abstract-Data mining is that the method of extracting valuable data from an over-sized information supply. Currently a day’s web has gained additional and additional attention of users with its wealthy interfaces and surplus quantity of knowledge on the market on net. As this has earned plenty of user’s interest in extracting plenty of helpful data it's still restricted with a number of the resources extraction like frail tagged facial pictures. This paper mainly investigates a unique framework of search-based face annotation by mining frail tagged facial pictures that are freely available on the web. Whereas by using this new technique, there occurs one major limitation i.e. how effectively can we perform annotation by exploiting the list of most similar facial pictures and their weak labels that are usually vociferous and incomplete? So as to resolve this drawback, we have a tendency to propose an efficient unsupervised label refinement (ULR) approach for refinement the labels of net facial pictures mistreatment machine learning techniques. We have developed the training drawback as a broken-backed improvement drawback and develop a effective improvement algorithms to resolve the large-scale learning task with efficiency. To any speed up the planned theme, we have a tendency to additionally propose a clustering-based approximation algorithmic rule which might improve the quantifiable significantly. As an extension for this proposed paper we've enforced a replacement search supported image search I.e. Image is taken as input not any text keyword and also the output is additionally retrieved within the sort of image, if the input image is matched with any of the set of pictures in image sound unit. Also we've enforced a graph so as to indicate the performance of assorted users United Nations agency has searched completely different classes of knowledge.

Download
22 Wound Healing Assessment by Using Granulation Tissue
Malempati Hari Krishna, Kunjam Nageswara Rao

Abstract-Wounds assessment is one of the prominent analysis that do not follow a predictable course of healing within a specified period of time One of the best methods for healing of the wound is the colour of the tissues. Describing the tissues colour change in terms of percentages clinical method are used for assessment of wound The growth of the red granulation tissue defines the beginning of ulcer healing. Granulation tissue appears red in colour due to haemoglobin content in the blood. An approach based on utilizing haemoglobin content in chronic ulcers as an image marker to detect the growth of granulation tissue is investigated. In this paper we Independent perform the component analysis for the extractions to of grey level haemoglobin images from chronic ulcers extracted RGB colour images indicate areas of haemoglobin distribution and detecting regions of granulation tissue. Segmentation and Data clustering techniques are implemented to classify to detect regions of granulation tissue. By using the implemented algorithm specificity and sensitivity is calculated and compare with the assessment. This project was developed in MATLAB implemented under windows platform.

Download
23 Filtering Florescent Noise for Effective Spot Segmentation of Microarray
Kottada Naveen Kumar, Kunjam Nageswara Rao

Abstract-microarray image quality produces lots of errors, especially the noise. A microarray image contains different type noises which produces different types of influence during image processing. For segmentation of features in Microarray gridding method was implemented, and the affection of different noises to automatic gridding has been listed. A new algorithm for noise reduction was implemented, which included two parts as edge noise reduction and highly fluorescence noise reduction. Edge detection was executed on the vertical and horizontal projections of microarray image. Highly fluorescent noise was removed by linear replace with calculated threshold value T which is an easy and fast means. The algorithm was implemented and compared to other common noise reduction methods. Experiment results show the feasibility of the proposed approach. This project was developed using MATLAB and implemented on windows environment.

Download
24 Design and Implementation of a New Four Wheeler Navigation System by Using Global Positioning System (GPS) Connection
P. Umadevi, A.Hari Kumar

Abstract-As there was a tremendous increase of vehicles in the current society, there was a huge traffic problem that occurs with this increase of vehicles. As the vehicles are increased in number almost maximum number of cars or heavy motor vehicles uses navigation system in order to identify the routes. But the major problem with these navigation systems is they can use Global Positioning Systems (GPS) as main source for identifying the routes from google maps, but they fail in identifying the traffic that is available between the roués which we travel. So in this paper for the first time we have implemented Live Traffic Indicator (LTI) based on air index in which all the paths will be available for the driver based on shortest path with no traffic. By conducting various experiments on our proposed LTI system, our simulation model clearly states that this model is the first time to be implemented in car navigation systems that reduces a lot of delay for the drivers who are very new to city.

Download
25 Design and Analysis of various Clustering Algorithms on Text Data Sets
Karetla Santhosh Kumar, Pydipala Laxmikanth

Abstract-Inspection is a process of finding the bugs /defects. In associate inspection, a defect is associate a part of the work on any product that may keep an inspector from approving it. In the process of inspection there will be two mechanisms like entry and exit conditions .This prevents unfinished work product from getting into the inspection method. In the entry condition the document has been initially spellchecked”. Computer/PC inspection may be a branch of knowledge security that examines structured data with relation to plenty of files that are hold on already in some elements of the system. The main aim is to find and analyze patterns of dishonest activities. Information from application systems or from their underlying information bases is spoken as structured data. Analysis of unstructured information is typically spoken as information inspection. In pc inspection analysis, many thousands of files are typically examined. A lot of the info in those files consists of unstructured text, whose analysis by pc examiners is troublesome to be performed. In this paper we try to find the clustering of text data sets with a well known clustering algorithms like (K-means, K-medoids, Single Link, Complete Link, Average Link, and trigonometric function Similarity Measure) applied on a number of the text documents of getting completely different classes like games, sports, living beings, animals, food items, books and also as an extension we are going to find out the outliers from the input data set. By conducting various experiments on our proposed CSPA algorithm on the real time text data sets, our simulation results clearly tell that our approach is best fitted for clustering the unstructured documents.

Download
26 Design and Implementation of MAC-Aware Routing for Achieving Quality of Service in Mesh Networks
S.G. Ramanujan, B. Kishore

Abstract-Networking is a process of connecting one or many systems together for communicating between each other. The performance of any network primarily enthusiastic about topology what we've chosen at the time of network deployment. Usually within the real time surroundings virtually mesh is employed for his or her digital communication. Generally in the mesh networks there will be a lot of problems occur frequently due to improper assignment of nodes to the routers. Generally each and every router will have maximum and minimum levels for handling the users; whenever we deploy a network we must check the node allocation properly. Another important reason is while deploying the network all the systems may not be homogenous in nature, so by specifying static bandwidth rules for data resource allocation and sharing it will lead to packet loss. These are the major causes why the quality of service is not maintained properly in wireless or mesh networks. So in this paper we have implemented a new topology clear programming like TTS where this will achieve QoS while routing. Also we have implemented two sub-modules like BWE and BWA in order to justify the bandwidth allocation to achieve Best Effort (BE).As an extension we have also implemented this paper on multi-router configuration where the multiple routers provide facility like no delay during routing between client and server nodes. By conducting various experiments of this planned new mechanism we've finally came to a conclusion that this new approach provides QoS routing in unintentional networks.

Download
27 Monitoring Data Leakage Using Concept Drifts in Process Mining
Sudamalla Venkata Krishna, G M Padmaja, Ramesh Challagundla

Abstract-Although most commercepracticestransform over time, modern-daypractice mining techniques have a tendency to scrutinize these practices as if they are in a steady state. Processes may change unexpectedly or progressively. The drift may be episodic (e.g., because of seasonal persuades) or oneand- only (e.g., the effects of new legislation). For the process management, it is decisive to ascertain and comprehend such perception drifts in processes. A nonspecificscaffold and specific techniques to distinguish when a process changes and to confine the parts of the process that have changed. Diverse features are projected to characterize relationships among deeds. These descriptions are used to ascertain differences between successive populations. The approach has been implemented as a plug-in of the ProM process mining scaffold and has been assessed using both simulated event data demonstrating controlled perception drifts and real-life event data from a Dutch municipality.

Download
28 Location based Search using Keyword with Attributes
Adireddi Jagadeesh, N. Sagar Pavan, Ramesh Challagundla

Abstract-Unadventurous spatial queries, such as range explore and nearest neighbour repossession, involve only circumstances on objects geometric properties. Nowadays, numerous contemporary applications call for novel forms of queries that aspire to find objects satisfying both a spatial predicate, and a predicate on their allied texts. For example, as an alternative of allowing for all the restaurants, a nearest neighbour query would instead solicit for the restaurant that is the neighbouring among those whose menus enclose “steak, spaghetti, brandy” all at the similar time. At present the unsurpassed solution to such queries is based on the IR2-tree, which, as shown in this paper, has a hardly any deficiencies that critically impact its effectiveness. Motivated by this, we build up a new access method called the spatial inverted index that broadens the unadventurous inverted index to hack it with multidimensional data, and comes with algorithms that can come back with nearest neighbour queries with keywords in authentic time. As demonstrated by experiments, the anticipated techniques outperform the IR2-tree in query retort time significantly, often by a feature of orders of enormity.

Download
29 Design and Analysis of a Denail of Service Attack in WSN using Group Testing (GT) Based Approach
Mohanrao Karaka, Y.Satish Kumar

Abstract-In current days almost all parts of the world is concentrating more and more on their security levels. As there was a tremendous increase of electronic devices almost each and every user try to store, retrieve, access the data to and fro via electronic devices. As the security plays a very important role even though there was a lot of hackers or intruders who try to hack or attack the sensitive data of others during data transmission. One among the several types of attacks is Denail of Service attacks where this attack will not physically cause damage to the sensitive data, but it creates some delay while transferring from one PC to other PC within a network. Till now there was a lot of literature regarding minimizing the routing delay in wired and wireless networks but still there was a lot of DOS occurs during data transfer. So in this paper we have implemented a new pair of two algorithms like Group Testing Approach (GT) and Novel Multivariate Correlation Analysis Algorithm (NMCAA) in order to identify the denial of service attack that occurs either request or response within the data communication. Here the GT based approach is used for identifying whether the denial of service attack occurred and NMCAA is used to identify at which level the attack has occurred and what will be delay time during data request. Here as an extension we have also implemented a concept like in which mode the attack has occurred. By conducting various experiments on a group of systems which are connected on LAN, we finally came to a conclusion that this proposed NMCA algorithm is best for reducing DOS attacks compared with various classical approaches.

Download
30 Design and Analysis of a Bloom Filter for Information Sharing in Mesh Networks
Boddakayala Manikumari, B. Kishore

Abstract-The term information sharing has a very long history in field of IT (Information Technology). In the primitive days, information sharing referred to one-to-one exchanges of data between a sender and receiver. But now a day’s information sharing is almost peer to peer in nature I.e Multiple sender to multiple Destinations. As the current world is running through mobile devices for communication between each other, a lot of internet usage is also increased in order to process the data. So while accessing the mobile devices remotely via internet there occurs a lot of problems during communication like poor bandwidth, battery limitations, frequency and a lot more. As the current mobile devices mainly depend on infrastructure or topology in order to communicate, if a node wants to send data from its location to other it should depend on various intermediate nodes in order to process the data. If any of the intermediate nodes becomes in active while data is been transferred the remaining nodes cant able to take the data of previous node and carry to the destination automatically. This will lead to a major problem called as Data Loss or Attack. So in this paper we have design and implemented a novel network which was created by human that enable the information sharing between mobile devices through direct inter device communication also known as UHUNET. In this we have design a BSUB network for inter-driven information sharing between HUNET nodes. By conducting various experiments on our proposed system, we finally came to a conclusion that this Unique Human Network very efficient and useful for almost all mobile/wireless users for data communication.

Download
31 Design and Analysis of a UORRA Algorithm in Order to Optimize the Resource Allocation Cost in Overlay Networks
Gorle Ramadevi, B.Shalini

Abstract-Routing is a process of sending dedicated packets from valid source to the valid destination through a device called as router. Generally this routing periodically changes if there was any node failure or link failure occurred intermediately during data transmission. If any node or link fails the general topology should be changed in order to send the data from source to destination, which is somewhat a costly process to reconfigure the network topology. Now a day’s overlay routing have achieved more attention of various network users for sending the data through network, this is mainly because there is no need to change the standards of current routing scheme if there was any delay or loss during the data transfer.Eventhough this is very attractive scheme that was opted by various network users, deployment of such a overlay network requires a huge overlay infrastructure. Inorder to build this network it requires huge maintenance cost for usage as well as deployment. As an extension for this paper we have implemented the same concept on network simulator where we can progressively show the performance and cost effectiveness of overlay relay networks. Also we have used AES Algorithm for encrypting the data request and send the data from valid source to destination. Once the data have been received in the Destination system the data should be identified by its node IP address then only the data will be decrypted in the receivers node, if not data will be in encrypted manner only. Our simulation results clearly tells that this was the proposed system on overlay networks which was not yet implemented till now where we can reduce a lot of delay, packets loss and reduce maintenance cost.

Download
32 Secure Data Sharing in Cloud Using Group Key
Midathana Rajeswari, A V D N Murthy, Ramesh Challagundla

Abstract-The nature of low maintenance, cloud computing offers an inexpensive andproficientelucidation for sharing group resource among cloud users. Unfortunately sharing data in a multi-owner comportment while preserving data and identity confidentiality from a untrusted cloud is tranquil a challenging topic, due to the recurrent change of the membership. Dynamic groups in the cloud, any cloud user can incognito share information with others. Meanwhile, the storage overhead and encryption computation charge of our scheme are sovereign with the quantity of revoked users.

Download
33 Comparative Evaluation of Spoofing Defenses
Gonapa Sravya, Vusirikala.Visweswararao, Ramesh Challagundla

Abstract-IP spoofing aggravates loads of security coercion, and tumbling it would significantly boost Internet security. Seven defenses that filter spoofed traffic have been projected to date; three are designed for end-network deployment, while four take for granted some collaboration with core routers for packet marking or filtering. Because each defense has been assessing in a unique setting, the following significant questions remain unrequited: 1) Can end networks efficiently protect themselves or is core support indispensable? 2) Which defense performs unsurpassed assuming sparse exploitation? 3) How to select central part participants to accomplish unsurpassed protection with fewest exploitation points? This paper answers the above difficulties by: 1) formalizing the predicament of spoofed traffic filtering and defining innovative effectiveness measures, 2) observing each defense as selfish or philanthropic and differentiating their performance goals, 3) defining most advantageous core deployment points for defenses that necessitate core support, and 4) evaluating all defenses in a familiar and pragmatic setting. Results offer a precious insight into advantages and limitations of the proposed defenses, and unearth the relationship between any spoofing defense’s performance and the Internet’s topology.

Download