IJCST Logo

International Journal of Computer Science and Technology Vol 5.4-2 Oct to Dec, 2014

S.No. Research Topic Paper ID Download
31

Neighbour Verification and Discovery in MANETS

Yerramsetti.Venkata Sairam, Medikonda Asha Kiran, Dr. G.S.Bapi Raju

Abstract
Information about the Location is an important resource in mobile networks. This information may be used by a number of protocols and applications. The knowledge of location of the nodes in a MANET is essential in these applications. The correctness of node locations is an important issue in mobile networks, and it becomes particularly challenging in the presence of adversaries aiming at harming the system. In Mobile ad-Hoc networks, routes may be disconnected due to dynamic movement of nodes. Such networks are more vulnerable to both internal and external attacks due to presence of adversarial nodes. These nodes affect the performance of routing protocol in ad-hoc networks. It is essential to identify the neighbours in a MANET. The proposed scheme identifies a neighbour and verifies its position effectively. Full Paper
IJCST/54/4/A-0349
32

Partitioning and Load Balancing in Cloud Computing

Esha Sarkar, Ch. Sekhar

Abstract
Cloud computing implementations offer seemingly infinite pooled computing resource over the network. In cloud computing resources are provided virtually for computation, storage and developing application. Cloud computing is built on the base of distributed computing, grid computing and virtualization. Cloud computing is defined as a large scale distributed computing paradigm. While resource are under sharing, load balancing play a critical role. An efficient load balancing scheme will definitely improve the performance of the cloud. There are many load balancing algorithms, such as Round Robin, Equally Spread Current Execution Algorithm, and Ant Colony algorithm. This paper discuses various load balancing schemes for the better improvement of the performance of the cloud. Full Paper
IJCST/54/4/A-0350
33

Data leakage and Prevention

Deepika Sonal, Ms. D.Kiranmay

Abstract
Data leakage can cause an organization untold public embarrassment and loss of revenue. Decision-makers therefore need to address the risk of data loss or leakage in light of a broader security strategy and roadmap. Data leakage incidents are not necessarily the result of malicious intent. Employees may transfer sensitive data to external media devices such as iPods, smartphones, USB memory sticks or e-mail to home addresses, just so that they can continue working from home. Data leakage can be caused by internal and external parties, either intentionally or accidentally.
As the capabilities of data transmission and storage are being continuously improved nowadays, data leakage incidents may result in more significant damages, diminishing organisations’ value and reputations. A comprehensive and effective DLP solution requires the commitment from both the management and general users to carefully determine the system specifications, functional requirements and data coverage, so that the solution can best fit in the university’s existing IT infrastructure and operational process and would not introduce inefficiencies and incompatibilities. Full Paper
IJCST/54/4/A-0351
34

Heterogeneous Wireless Sensor Networks – Redundancy Mnagement of Multipath Routing

Pandi P Pradeep Kumar, K.Vinay Kumar

Abstract
Un reliable and malicious nodes create a lot of problems in wireless sensor networks. To address the energy consumption and QoS gain in reliability, delay and security with the goal to maximize the lifetime of a clustered Heterogeneous Wireless Sensor Networks while satisfying application QoS requirements in the context of multipath routing. The proposed research is a highly scalable cluster-based hierarchical trust management protocol for wireless sensor networks (WSNs) to effectively deal with selfish or malicious nodes. The proposed work consider multidimensional trust attributes derived from communication and social networks to evaluate the overall trust of a sensor node. The protocol relies on a new multipath constructions paradigm that is defined specifically for heterogeneous WSN. The approach leverages a reasonable increase in the network lifetime and a higher resilience and fault tolerance. Full Paper
IJCST/54/4/A-0352
35

An Ensure Outsourcing System for Crystallizing (LE) Linear Equations using Cloud Computing

Spandana Kala, Kota Venkata Narayana Rao

Abstract
By means of concentrating memory, bandwidth and processing cloud computing permits for additional resourceful computing and to preserve the data the internet was used by the technology.
Cloud is kind of centralized database where numerous clients accumulate their data, recover data and possibly adjust data and it is a representation where user is made available services by Cloud Service Provider on the basis of pay per use. For the past few years, the technology of cloud computing has the extreme growth sections in the field of infrastructure and permits the consumers to make usage of applications devoid of installation and by means of internet access the personal files. Even if the utilization of cloud computing has rapidly improved; the safety of cloud computing is still considered the most important issue in the environment of cloud computing. By taking a variety of kinds of data in the data safety, the difficulty of checking correctness of data is still became a challenging one. The examining from the approaches of existing and the computational practicality inspires to plan secure method of outsourcing linear equations by means of an entirely different method of iterative approach, where the explanation is extracted by means of finding consecutive estimations to the elucidation until the necessary accuracy is attained. When measured to direct method, method of iterative merely demands moderately simpler operations of matrix-vector with O(n2) cost of computation, which is greatly easier to put into practice in practice and extensively adopted for large-scale linear equations. Full Paper
IJCST/54/4/A-0353
36

Power, Bandwidth and Distance Optimization in WMAN – Using Cross Layer Approach

Raman Kumar, Manjinder Singh Kahlon

Abstract
Cross layer design is an important technique in the design, performance analysis and simulation study for improve quality of service to the user. The present paper provides the overview of the wireless technologies along with critical and comparative analysis for future need. The study is useful in the design and development for protocols to fulfill the research of future user needs in this field as well as possible a new dimension power, bandwidth & distance using cross layer approach. Our results shows that the task response time for the proposed scenarios produces efficient results as compared to existing work as proposed by previous authors. Full Paper
IJCST/54/4/A-0354
37

Load Balancing Performance in Content Delivery Networks (CDN’s)

Kanthai Srinivasa Rao, Manjula Srinivas, Molli Srinivasa Rao

Abstract
It is an interesting issue of describing and executing an effective law for load balancing in Content Delivery Networks (CDNs). A CDN system conceded out through the exploitation of a fluid flow model classification of the network of servers. Beginning from such categorization, we develop and demonstrate a lemma about the network queues balance. The result is then leveraged in mandate to create a novel distributed and time-continuous algorithm for load balancing which is also reformulated in a timediscrete form. The isolated devising of the proposed balancing law is ultimately deliberated in terms of its definite application in a realworld consequence. Lastly, the complete method is authenticated by means of imitations. Full Paper
IJCST/54/4/A-0355
38

To Do Comparitive Study by Implemennting Cryptography Algorithms on the Bases of RSA and EL-GAMAL

Shaina Arora, Pooja

Abstract
This paper is based on the combination of different types of asymmetric key cryptography. Here we will going to combine two different algorithms which are RSA algorithm and El-Gamal algorithm and hence form a new algorithm named as RSAELGAMAL algorithm. The results will be implemented with the help of JAVA, Such that the combined RSA-ELGAMAL algorithm will be written in JAVA code hence run on it. We will be going to find the time of different bit length of a same text massage, such as “hello”. By increasing the bit length of the same massage will going to increase the complexity of the massage, due to which the hacker will be failed to crack the massage. We were having one more opposition to increase massage size instead of bit length but it is more complex and confusing because with the increase in massage size the memory space will increase the whole process will be very complicated and confusing.
Our motive is to confuse the hacker to crack the massage not to confuse our self. So we are going to implement easy coding, just increase the bit length only not massage by this the massage will be easy and more secure. At the end the graphical representation will be done on MATLAB.
Cryptography basically meant for secure communication, its motive to send or receive the massage, ought to be secure, confidential, authenticated and non-repudiate. This paper relies on the mixture of RSA AND El-Gamal Cryptography with their security problems.
The RSA Cryptography depends on the integer factorization problem (IFP), whereas the El-Gamal Cryptosystem depends on distinct logarithm problem (DLP). This model works on the idea of merging between IFP with DLP which provides a good speed of computation for uneven cryptosystem supported troublesomeness of the solutions of 2 difficult issues, Hence the computation of the proposed system is additional economical compared to El- Gamal algorithm and RSA algorithm. Cryptography may be a acknowledge for confidentiality , security and privacy .Here during this paper we tend to area unit functioning on the implementation numerous |of varied| of assorted} asymmetric key cryptography techniques by modifying their architecture design as well as combining of varied algorithms along and so to try and do the comparative study beside various complexness analysis. Full Paper
IJCST/54/4/A-0356
39

WSN Secure Coverage of Large Scale WSN Using Scalable Key Management Scheme

Avala Venkata NookaRaju, Renuka B, Molli srinivasa Rao

Abstract
One of the foremost apprehensions when designing a key management scheme is the network scalability. Indeed the protocol should support a huge number of nodes to allow a large scale distribution of the network. An improved united-based key predistribution arrangement providing high network scalability and good key distribution probability approximately lower confined by 1 − e¯ ≈ 0.632. Results display that the proposed approach improves the network scalability while providing high secure connectivity coverage and complete better-quality performance.
Furthermore, for an equivalent network size, our solution diminishes significantly the storage overhead compared to those of existing solutions. Full Paper
IJCST/54/4/A-0357
40

Improving Security For Dynamic Groups in Cloud Using Bell-Lapadula Model

Sridevi K

Abstract
A public cloud provides a scalable access to the computing resources & IT resources. In addition to these services it should also posses the level of security when it comes to dynamic groups We propose a dynamic group system to implement Multi-Level Security using Bell-Lapadula Model to distinguish the users in the dynamic group by their roles in the organization. This technique allows the user to permit the required privileges for a selected group based on the level of the model and also initializing fault-tolerance by providing back-up facilities to the servers by improving the availability in the cloud.
Full Paper
IJCST/54/4/A-0358
41

Modified History Information-Exchange Optimized Protocol : A Clone Node Detection Protocol in Mobile Wireless Sensor Networks

Avneet Kaur, P. S. Mann

Abstract
Wireless Sensor Networks (WSNs) are frequently deployed in unfriendly environments, they are exposed to physical capture and cooperation. In a node replication attack or clone attack, an adversary physically arrest a sensor node, removes its cryptographic confidences and distributes a large number of imitations of captured node all over the network. Previous node clone attack detection scheme are mainly concentrated on either centralized or distributed solutions and have severaldrawbacks. In this paper, we propose a clone attack detection scheme, namely the Modified History Information-Exchange Optimized Protocol (Modified HOP), which is based on both centralized as well as distributed solution and its simulation result shows that this protocol is more efficient and provides better detection capability. Full Paper
IJCST/54/4/A-0359
42

Energy Balanced Clustering and Aggregation for Wireless Sensor Networks

M.Umadevi

Abstract
Wireless Sensor Network is a large group of dedicated equipments with a communication infrastructure intended to supervise and document conditions at diverse locations. Sensor nodes are limited in power. It is difficult to recharge the battery of sensor. Energy preservation is the main issue of WSN. Clustering and Aggregation are the energy efficient methods. To deal this issue, EBCA(Energy Balanced Clustering and Aggregation for Wireless Sensor Networks) is proposed. In the proposed algorithm the network is divided into subnets called as clusters and head for the cluster is selected based on the number of neighbors, residual energy and distance of a node from the centre of the cluster.
And also a mobile sink node is used which travels across the network to collect data. Simulation results show that the proposed algorithm achieves higher energy efficiency and increases network lifetime.
Full Paper
IJCST/54/4/A-0359
43

Improving Transliteration Candidate Ranking Using Phonetic String Distance

Mohamed Afifi, Gouda Ismail, Khaled Badran

Abstract
Many transliteration methods are applied to different language pairs, but still a lot of them didn’t reach the accuracy needed. We propose a method that can be applied to any of these techniques that can be used to improve the transliteration result. We create the phonetics for both the input and the resulting candidates then we calculate the edit distance between them using a modified Levenshtein distance. Then we order them according to their least distance. As an example we considered the romanized Arabic to Arabic transliteration in chat text. A top(1) accuracy of 55.6% and a top(5) accuracy of 86% was achieved using this technique. When considering the least distance phonetics to be all correct due to variations in the result, the top(1) accuracy reached 75%.
Full Paper
IJCST/54/4/A-0360
44

Study of Spatial and Temporal Variations in Surface Air Temperature Using Spatial Data Mining

Harpinder Singh, R.K. Setia, P.K. Litoria, Brijendra Pateriya

Abstract
Identification of interesting and unexpected climate change patterns from vast spatial and temporal datasets is an intricate task.
The spatial and temporal variations in the surface air temperature of Indian Punjab State was studied using the data of 101 years (1901- 2002). Spatial data mining was used for multivariate spatial clustering and geovisualization of the data for identifying clusters with similar trends of temperature change. We used this approach, for the first time, to predict the changes in temperature over the years in Indian Punjab. The least change in surface temperature is in central Punjab, but maximum change is in north and north eastern regions of Punjab.
Full Paper
IJCST/54/4/A-0361
45

Action Recognition Using AMI and Support Vector Machine

V.K.Swarnambigai

Abstract
Human activity recognition is an important area of computer vision research.Its applications include surveillance systems, patient monitoring systems , and a variety of systems that involve interactions between persons and electronic devices such as humancomputer interfaces. The goal of human activity recognition is to automatically analyze ongoing activities from an unknown video. This paper provides a detailed overview of recognition of human actions. We first define the Accumulated Motion Image (AMI) by using the technique of frame differencing. The Energy Histograms for horizontal and vertical directions are computed from AMI and those features are extracted for further process.The Discrete Fourier Transform is Computed from Energy Histograms and those features are also extracted. A trained Multi-class SVM(Support Vector Machine) is used to recognize the various actions from all these features. Public dataset is used for Evaluation.
Full Paper
IJCST/54/4/A-0362
46

A Review of Multicasting Routing Protocols in MANETs

Nagendra Sah, Nagendra Pal, Paritam Soni

Abstract
Mobile ad-hoc network (MANET) is a network of wireless mobile nodes (MN) which communicate with each other without any centralized control or established infrastructure. Routing is the process of selecting paths in a network along which data is to be sent. Routing is a critical task in MANET where the nodes are mobile. Dynamic and reliable routing protocols are required in the ad-hoc wireless networks, as they have no infrastructure (base station) and their network topology changes. There are various protocols for handling the routing problem in the ad-hoc wireless network environment. The objective of this paper is to review the current state of the art of existing routing protocols for MANETs, and compare different approaches. There are three main classes of routing protocols for MANETs: reactive, proactive and hybrid.
Finally, the comparison of two protocols MAODV (Tree based) and ODMRP (Mesh based) are presented under some specified condition using Qualnet-5.0.
Full Paper
IJCST/54/4/A-0363
47

Hybrid Optimization for Plant Leaf Classification

C. S. Sumathi, A.V. Senthil Kumar

Abstract
In agriculture and medicinal plant recognition/classification has great application prospective and is especially significant to biology diversity research. The shape of plant leaves is an important feature for the visual characterization of various plants.
As many plants face extinction, it is necessary to start a plant protectiondatabase. Various methods automate leaf recognition for plant classification. In this paper, a proposed Differential Evolution along with GA optimization algorithmis optimized and the percentage of classification accuracy is determined.
Full Paper
IJCST/54/4/A-0364
48

Software Cloning and Its Detection Methods

Meena Bharti, Rajan Goyal, Manish Goyal

Abstract
Software Systems are evolving by adding new functions and modifying existing functions over time. As we do evolution of software, the structure of software is becoming more complex [5] and so the understandability and maintainability of software systems is deteriorating day by day. Copy and paste programming is a common activity but it introduces a negative point to reuse by creating Clones. Several studies show that about 5% to 20% of software systems can contain duplicated code, which are basically the results of copying existing code fragments and using then by pasting with or without minor modifications. The main disadvantage of duplication of code is that if a bug is detected in a code fragment; all the other fragments similar to it should be investigated to check the possible existence of the same bug in the similar fragments. Restructuring of such code is another prime issue in software maintenance although several studies claim that refactoring of certain clones is not desirable and there is a risk of removing them. Detection of duplicate fragments increases the efficiency of software maintenance process and thus decreases maintenance cost.
Full Paper
IJCST/54/4/A-0365
49

Energy Aware Cloud Computing Using Dynamic Voltage Frequency Scaling

Pooja Chauhan, Manjeet Gupta

Abstract
The increasing demand of cloud infrastructure has severely increased the energy consumption of datacenters, which has become a critical issue. High energy consumption not only results in high operational costs, but also leads to high CO2 emissions which contribute to growing environmental issue of global warming significantly. In this paper, we proposed a scheduling algorithm for the cloud datacenter with dynamic voltage frequency scaling technique. And we simulated it by Green Cloud Simulator. Our proposed method can efficiently maximize the resource utilization and results in significant reduction in power consumption for executing the jobs. Experimental results show that our proposed scheme reduces more energy consumption than Non DVFS scheme.
Full Paper
IJCST/54/4/A-0366
50

Classifier Comparison For Bilingual Gujarati-Roman Printed Documents

Shailesh A. Chaudhari, Dr. Ravi M. Gulati

Abstract
In an OCR system for processing a bilingual document, primary importance is given to recognition of bilingual script in image of a document page. In a multi-lingual country like India, in most of the official papers, school text books, magazines, it is observed that English words intersperse within the Indian regional languages. In this paper authors discussed feature extraction and classification for segmented Gujarati and Roman scripts for printed bilingual documents. Here authors propose the line-wise script identification.
The spatial spread of pixels on Upper and Lower parts associated with Gujarati and Roman are used to identify the script. Authors have used horizontal projection for line distinction belonging to different script. Two classifier k-nearest neighbour (KNN) and support vector machine (SVM) scheme have been proposed for analyzing error rate for recognition. It has been observed that SVM classifier outperform as compared to KNN classifier.
Full Paper
IJCST/54/4/A-0367
51

Providing Preventive Schemes Like EAACK to Secure Routing in MANETS

M.U.V.S.Deepthi, Syed Shaheen

Abstract
The mobility and scalability carried by wireless network ended it probable in many applications. Surrounded by all the contemporary wireless net- works Mobile Ad hoc NETwork (MANET) is one of the most significant and exclusive applications. On the different to traditional network architecture MANET does not entail a fixed network infrastructure every single node mechanism as both a transmitter and a receiver. Nodes communicate directly with each other when they are both inside the similar communication range. If not they rely on their neighbours to communicate messages. The self-configuring capability of nodes in MANET completes it admired amid essential mission applications like military use or emergency recovery. In latest years safety has turn out to be a most imperative service in Mobile Adhoc Network. Compared to other networks MANETs are more defenceless to variety types of attacks. In this paper a qualified learns of Secure Intrusion- Detection Systems for determining malicious nodes and attacks on MANETs are presented. Due to some special characteristics of MANETs prevention mechanisms alone are not sufficient to handle the secure networks. In this case detection should be determined as another part before an attacker can injure the structure of the system.
Full Paper
IJCST/54/4/A-0368
52

An Approach to Extract Information Based on Tree-Based Association Rules (TARS) Mined Rules From XML Documents

K.S.P.Keerthi, B.Kishore

Abstract
The database research field has determined on the Extensible Markup Language (XML) as a supple hierarchical model appropriate to represent huge amounts of data with no complete and fixed schema and a probably uneven and imperfect structure. There are two main approaches to XML document access keyword-based search and query-answering. The first one comes from the tradition of information reclamation where most searches are carried out on the textual content of the document this means that no advantage is imitative from the semantics conveyed by the document structure.
As for query-answering since query languages for semi structured data depend on the document structure to convey its semantics in order for query formulation to be effectual users need to know this structure in advance which is often not the case. When users state queries without knowing the document structure they may fail to retrieve information which was there but under a different structure. This restraint is a critical issue which did not come out in the context of relational database management systems.
Full Paper
IJCST/54/4/A-0369
53

Energy Conservation Technique Using Disposable Low Cost Mobile Relays

B. Srinivasa Rao, I.S. Srinivas Rao

Abstract
Data-intensive WSNs is to convey all the data generated within an application’s lifetime to the base station despite the fact that sensor nodes have limited power supplies. The approach differs from previous work in two main aspects. First it does not need complex motion planning of mobile nodes so it can be executed on a number of low-cost mobile sensor platforms. Second we combine the energy consumption due to both mobility and wireless transmissions into a holistic optimization framework. Our framework consists of three main algorithms. The first algorithm calculates an optimal routing tree assuming no nodes can move.
The second algorithm improves the topology of the routing tree by greedily adding new nodes exploiting mobility of the newly added nodes. The third algorithm improves the routing tree by relocating its nodes without changing its topology. We present efficient distributed implementations for each algorithm that require only limited localized synchronization. Because we do not necessarily compute an optimal topology our final routing tree is not necessarily optimal.
Full Paper
IJCST/54/4/A-0370
54

A Systematic Approach for Safe Data Forwading in Cloud

N.V.S.Sowjanya, S.Rama Sree

Abstract
A cloud is a collection of storage servers maintained by the cloud service provider. It is providing on-demand self service, resource pooling, rapid elasticity and measured service. But data confidentiality is the main problem for the users when they are storing the information in cloud. But cloud minimizes the investment cost of the individual users. Security can be provided by building a secure storage system along with two identity-based secure distributed data storage schemes. This scheme can capture the following properties: (1) The file owner can decide the access permission independently without the help of the private key generator (2) For one query, a receiver can only access one file, instead of all files of the owner. The proposed method is a proxy reencryption scheme which is integrated with a decentralized erasure code so that a secure distributed storage system is developed.
Full Paper
IJCST/54/4/A-0371
55

Comparative Synthesis of Speech Signal in Linear and Quadratic Discriminant Analysis With Radial Basis Function Mapping Algorithm

Jeevant Singh Gaharwar, Samta Gajbhiye

Abstract
Many wireless communication devices are available now days. Since Signal generated from wireless devices when transferred from sender to receiver, sometimes signals are noisy signal. In order to know how much noise is present in the signal and what amount is affected classification is necessary. Linear and Quadratic discriminant analysis is performed in order to classify noise from accurate signal. Since there are various frequency bands available so at the time of classification care must be taken so that data must not be lost. Both these classification strategies such as (LDA and QDA) are used with Radial Basis function mapping Algorithm in order to get the accurate signal on the receiver side.
Full Paper
IJCST/54/4/A-0372