IJCST Logo

International Journal of Computer Science and Technology
Vol 6.2 ver – 3 (April to June, 2015)

S.No. Research Topic Paper ID Download
51

A Review of Miss Penalty Reduction Techniques in Cache and Introduction of Extra Free Cache Lines

Miraj Gul, Noorul Amin, Faisal Bahadur

Abstract

Improving miss penalty and cache replacement policy has been a hotspot for researches. The miss penalty may be defined as the number of cycles microprocessor is stalled for a memory access is called miss penalty and it is sum of
• Cycles to replace a block in the cache.
• Cycles to deliver the block to the processor.
The miss penalty problem reduces the overall efficiency of the processing environment. Several efforts have been made to improve the performance of the cache by reducing the miss penalty rate. In this paper, we discuss the work done by different researchers to reduce the miss penalty problem and propose a new idea that can be combined with the existing techniques to improve the performance and reduce the miss penalty problem.
Full Paper

IJCST/62/1/A-0492
52

Performance Analysis of Filter based Techniques for Image Denoising

Bhawna, Sukhjeet Kaur Ranade

Abstract

An Image has been often corrupted by different kinds of noises when it is processed, compressed and stored which degrades its quality and becomes an obstacle to image analysis and extraction of image feature. Efficient noise suppressing image de-noising technique is still a big challenge in image analysis and processing. In spatial domain filtering techniques filters can tackle the denoising problems without affecting the significant features of an image and gives us better results in restoring an image. This paper presents practical aspects of various types of filters that have been used for suppression of noise. Experiment was performed on 12 standard test images that have been corrupted by different Gaussian noise levels and results has been compared in terms of peak signal-to-noise ratio(PSNR), mean squared error(MSE) to quantify the performance of denoising techniques.

Full Paper

IJCST/62/1/A-0493
53

Energy Aware Intelligent Resource Provisioning in Heterogeneous Clouds

Rohit N. Raje, Nishigandha Rane, Nikhil Kawishwar, Parul Parwal, Ruhi Patankar

Abstract

Data centers consume tremendous amounts of energy in terms of power distribution and cooling.Also, data centers in particular comprise of heterogeneous machines with different capacities and energy consumption characteristics. Meanwhile, production cloud workloads typically consist of diverse applications with different resource requirements, with different priorities. Despite extensive studies of the problem, existing systems fail to fully consider the machine and workload heterogeneity found in production environments. Heterogeneity oblivious systems lead to both suboptimal energy savings and long scheduling delays. To resolve this we present an intelligent heterogeneity aware resource provisioning scheme in clouds. It is a promising approach which utilizes learning algorithms for reducing scheduling delay and energy consumption by selecting optimal machines to match resource demands. We then compare our system with another system employing a random scheduler and find that our heterogeneity aware scheme performs better in terms of energy costs.

Full Paper

IJCST/62/1/A-0494
54

Performance Improvement in Trust Based Routing Protocol Using Stability Election Distance Vector Protocol in Manet

Neha Yadav

Abstract

A wireless ad-hoc network is a collection of mobile host association that open a concise circle with no aid of any centralized administration. Due to restricted distribution of group of each cell host’s in packet transmission, this category of environment is cooperative for a single mobile host to connect together and support other nodes in forwarding a packet to its destination. Security in MANETs plays a vital role in network environment. However there are five security parameters due to which various types of attacks can be detected. These parameters are Authenticity, Integrity, Nonrepudiation, Availability and Scalability. In this paper a trust based AOTDV(Ad-hoc on demand Trust Based distance vector)[5] and stable election AOSEDV( Ad-hoc on demand Stability Election distance vector) routing protocol have been designed for MANETs to inhibit the attacker from copying response through the packet . Moreover, a trust prediction model for securable network which also increases the network lifetime has been described in the paper. This model can evaluate the trust worthiness of nodes, based on the generic behaviors of nodes.

Full Paper

IJCST/62/1/A-0495
55

Authentication for Data Storage in Clouds with Anonymous Decentralized Access Control

Sandhya.Medavarapu, Madhira Srinivas

Abstract

We propose a replacement suburbanized access management theme for secure knowledge storage in clouds, that supports anonymous authentication. Within the projected theme, the cloud verifies the credibility of the ser while not knowing the user’s identity before storing knowledge. Our theme additionally has the supplemental feature of access management during which solely valid users square measure able to decipher the hold on data. The theme prevents replay attacks and supports creation, modification, and reading knowledge hold on within the cloud. We tend to additionally address user revocation. Moreover, our authentication and access management theme is suburbanized and sturdy, incontrast to different access management schemes designed for clouds that square measure centralized. The communication, computation, and storage overheads square measure similar to centralized approaches.

Full Paper

IJCST/62/1/A-0496
56

Implementation and Performance Analysis of Face Recognition Using MATLAB

Vimal, Virender Kadyan

Abstract

A face recognition system is a computer application for automatically identifying or verifying a person from a digital image or a video frame from a video source. Face recognition is the most efficient and sophisticated method for the security systems. It is the part of the Biometric Technology, which has been widely used in forensics, secured access, human computer interaction and prison security.

Full Paper

IJCST/62/1/A-0497
57

De-duplication with Authorization in Hybrid Cloud Approach for Security

Sevitha.Narne, Madhira Srinivas

Abstract

Data deduplication is one in every of necessary information compression techniques for eliminating duplicate copies of continuation information, and has been wide employed in cloud storage to scale back the quantity of space for storing and save information measure. To guard the confidentiality of sensitive information whereas supporting deduplication, the merging encoding technique has been projected to encode the info before outsourcing. To raised defend information security, this paper makes the primary conceive to formally address the matter of approved information deduplication. Completely different from ancient deduplication systems, the differential privileges of users square measure additional thought-about in duplicate check besides the info itself. We have a tendency to conjointly gift many new deduplication constructions supporting approved duplicate sign on a hybrid cloud design. Security analysis demonstrates that our theme is secure in terms of the definitions laid out in the projected security model. As a symptom of conception, we have a tendency to implement an epitome of our projected approved duplicate check theme and conduct test bed experiments exploitation our epitome. We have a tendency to show that our projected approved duplicate check theme incurs to kenish overhead compared to traditional operations.

Full Paper

IJCST/62/1/A-0498
58

Prediction of Cardio-Vascular Disease Using Classification Algorithms in Data Mining

Shilpa, Mamta Dhanda

Abstract

The field of data mining involves working on the huge data stored in databases to extract useful information from that data which is otherwise not evident. We perform a research here to use the data from Cleveland database which contains the health records of 300 patients to create a methodology to predict the heart disease in a human being on base of his health statistics.The database contained 73 attributes which were later reduced to 15. We use optimization techniques like simulated annealing and GA to reduce the number of attributes to 5. The effectiveness of decision tree method is further improved by employing local heuristic search algorithm like GA, Tabu Search or simulated annealing before that.

Full Paper

IJCST/62/1/A-0499
59

Vampire Attacks: Improved and Unload Ontology in Wireless Sensor Networks

S.Pallavi, C.Yosupu

Abstract

Ad-hoc is small wireless networks is modify different methods direction in many different ways Provide security in our network in the domain main focused on different taken nodes in the network routing the work is taken different locations dictions attacks at the routing different layer which permanently disconnections networks is efficiently changed the nodes different battery power These “Vampire” attacks are not specific to different specific We find the different networks is taken protocols are susceptible to Vampire attacks which are finding difficult errors and are easy to take other locations using as few small insider sending only network protocol different sizes a single Vampire can increase network-range size usage by a factor of O(N) where N in the number of network nodes We take different methods to uses these types of attacks is finding a new algorithms models these are dictions leakage data Vampires working to finding the data forwarding phase to used two different algorithms these are Linkstate algorithm and Distance vector routing algorithms.

Full Paper

IJCST/62/1/A-0500
60

Efficient Ant Colony Based Load Balancing in Cloud Computing

Kamalpreet Kaur, Suveg Moudgil

Abstract

Load balancing is moderately new method that facilitates webs and resources by bestowing a maximum throughput alongside negligible reply time. Splitting the traffic amid servers, data can be dispatched and consented lacking main wait. Disparate assortments of algorithms are obtainable that can aid traffic loaded amid obtainable servers. Burden balancing resolutions usually apply redundant servers that assist an enhanced allocation considering the contact traffic so that the locale supply is conclusively settled. Burden Balancing is an vital facet of large Groundwork cloud computing environments. Burden balancing has to seize into report two main jobs, one is the resource provisioning or resource allocation and supplementary is task arranging in distributed environment.

Full Paper

IJCST/62/1/A-0501
61

Facilitating Document Annotations Using Content and Querying Value

Uminvad Sampath, C.Yosepu

Abstract

A large no of company’s every day papered and give the information textual format of the products, services, and actions these are collection in of textual data information’s efferent and given structured information While data extraction algorithms used the extraction of structured relevant data they are modify expensive and insufficient model then operating on top to bottom text that does not contain any other information’s of the targeted structured information We present a novel different models that sharing and generated of the structured data about data is finding documents that are likely to data information in online devisees and this information is going to be taken useful for statements the database We studied how the fuzzy search and proximity ranking will improve efficiency of searching.

Full Paper

IJCST/62/1/A-0502
62

A Survey of Feature Based Finger Print Recognition Techniques

Jaspreet Kaur, Navleen Kaur

Abstract

Fingerprints have long been utilized as a reliable biometric feature for confidential identification. Fingerprint association mentions to the setback of allocating fingerprints to one of countless prespecified classes. Automatic association can be utilized as a pre-processing pace for fingerprint matching, cutting matching period and intricacy by restricting the find space to a subset of a normally huge database. Fingerprint matching methods can be mainly categorized as minutiae established and correlation based. Minutiae based method early locates the minutiae points in a given fingerprint picture and matches their comparative placements in a stored template fingerprint. This Paper surveys Feature based and wavelet based fingerprint marching of the methods happening in the research community.

Full Paper

IJCST/62/1/A-0503
63

H.264 AVC Part-10 Visual Video Encoding Standard

Romita Gakhar, Dr. Parvinder Singh

Abstract

H.264/AVC is newest video coding standard of the ITUT Video Coding Experts Group and the ISO/IEC Moving Picture Experts Group (MPEG). The main goals of the H.264/AVC standardization effort have been enhanced compression performance and provision of a “network friendly” video representation addressing conversational (video telephony) and “non-conversational” (storage, broadcast, or streaming) applications. This paper provides an overview of the technical features of H.264/AVC, describes profiles and applications for the standard, and outlines the history of the standardization process.

Full Paper

IJCST/62/1/A-0504
64

A Proposed to Display any One Character of Regional Language for Booting Screen of Linux Based Mobile Operating System Using RELAN Algorithm

Milan S. Bhatt, Prashant M. Dolia

Abstract

There is a no to display a regional language character at the time of booting screen of Linux based mobile operating system in any Linux devices. So Information technology without doubt, has enormous power to improve how village people live and work using mobile!? A revolution is taking place today in the way of regional people, how to access, learn, and interact with information looking booting process in regional language for the Linux based mobile Operating System.

Full Paper

IJCST/62/1/A-0505
65

Performance Evaluation of Collective Communication Based Matrix Multiplication Algorithm on Cluster

Gaurav Gupta, O.P. Gupta, B.K. Sawhney

Abstract

In today’s era of scientific computing and technology, there is a shift in trend of data processing from conventional processing towards parallel processing. The technique of computing of multiple tasks simultaneously on multiple processors is called Parallel Computing. The parallel program consists of multiple processes solving a given problem simultaneously. Parallel computers can be roughly classified as Multi-Processor and Multi-Core. In both these classifications the hardware supports parallelism within single node where each node can have multiple processing elements either in single chip pack or in distinct chips. Parallel hardware design employs array processors, pipelined systems which can be further extended to scalar and super scalar pipelined systems.

Full Paper

IJCST/62/1/A-0506
66

Automatic Test Packet Generation

M.Santhosh, U.Sivaji

Abstract

Networks is generated many complex administrators tools in the time of debug problems We propose the automated and produces 0rinted testing and debugging networks called “Unmanned Quality Container Prompting reads router configurations and generates small devises model This model is used to generate a minimum set of test container to exercise every link in the network changes every rule in the network model Test container are sent to data and detected errors a separate models to localize the fault ATPG can detect both functional and performance problems complements but goes beyond earlier work in static verify fault environment We take our protocol implementation and results it different applications Stanford University’s backbone network and Internet2 We find that a small number of test container each one to test all rules in different networks models

Full Paper

IJCST/62/1/A-0507
67

Survey of Asynchronous Medium Access Protocols for Wireless Sensor Networks

Dishee Agarwal, Arvind Kakria, Dr. Trilok C. Aseri

Abstract

Wireless Sensor Network (WSN) has become an emerging technique providing solution in scenarios such as seismic monitoring, industrial process monitoring, perimeter security, surveillance etc. It consists of large number of inexpensive miniature nodes capable of sensing, computation and communication. WSN MAC protocols are required for energy efficient working of the network. In this paper we discuss and compare various Asynchronous MAC protocols for WSN.

Full Paper

IJCST/62/1/A-0508
68

On Selection of Efficient Protocol for Mobile IP Enabled VANET through Realistic Channel

Richa Sharma, Jyoteesh Malhotra

Abstract

Vehicular Ad Hoc network is a versatile mobile wireless Ad-Hoc network targeted to support traffic monitoring, vehicular safety and many more applications. For the robust and reliable services in the VANET there is need for selection of efficient protocol to check the performance under frequent handovers in Mobile IP to prevent packet loss. Mobile IP is an interface that helps to track the mobile nodes and deliver messages even if vehicles are out of the coverage area of home node. In order to find the achievable performance bounds in terms of throughput, packet drop, collision rate and packet broadcast rate, extensive simulations have been done. A realistic city scenario has been proposed here by using the rician channel simulator, mobile IP enabled IEEE 802.11b OBUs and RSUs. The transmission speed levels of vehicles have been varied to obtain the optimum performance through realistic conditions. Simulations are performed using NCTUns6.0 (National Chiao Tung University Network Simulator) in mobile IP interface.

Full Paper

IJCST/62/1/A-0509
69

Survey on Mining Web Graphs for Large Scale Meta Search Engine Results

Bandi Krishna, Dr. V.B.Narasimha

Abstract

MSE is a search tool that sends user requests to several other search engines/databases, aggregates the results, merges/re-ranks them into a single list and displays them to user by using web graphs. MSEs enable users to enter search criteria once and access several search engines simultaneously. This also may save a lot of time to the user from having to use multiple search engines separately by initiating the search at a single point. Most of today’s MSEs employ only a small number of general purpose search engines. Building large-scale MSEs using numerous specialized search engines is another area that deserves more attention.

Full Paper

IJCST/62/1/A-0510
70

Software Effort Estimation Using Bayesian Methodologies

Anjali Sharma, Sonia Sharma

Abstract

Software Effort Estimation is an issue that is important software project management. The costs increase with increasing the complexity of the software and Efforts. Therefore the accuracy of the estimation in early stages of the project is extremely favorable. In fact, the cost and finishing that is timely of projects are among decisive factors of prosperity of a software development company. Other strategic issues such as timing and manpower that is experienced also effective in the success of the project. The major objective of software engineer is to guarantee to deliver software that is high-quality time and within budget. The size and complexity of software is increasing so quickly that the cost is often out of control as the development of software technology and the rapid extension of application areas. However, few groups or researchers have proposed an method that is effective help project manager make reasonable project plan, resource allocation and improvement actions.

Full Paper

IJCST/62/1/A-0511
71

Efficient Web Log Mining Using Ant Clustering on Tree Structured Data

Amita Kulhare, Hare Ram Shah

Abstract

Application of data mining techniques to the World Wide Web is referred to as web mining. Web mining is a technology to find out implicit pattern, from vast web document structures and muster. With the rapid development of Internet, web data mining, especially weblog mining plays an important role in many fields, including personalized information service, improving designs,services of websites and so on.

Full Paper

IJCST/62/1/A-0512
72

Energy Efficient Distance Oriented Clustering Protocol for WSN

Shanky Rani, Rohit Chahal

Abstract

WSN is a power constrained system, since nodes run on limited power batteries shorten lifespan of the system. Prolonging the network lifetime depends on efficient management of sensing node energy resource. Hierarchical routing protocols are best known in regard to energy efficiency. By using a hierarchical clustering routing protocols Technical greatly minimize energy Consumed in collecting and Disseminating data. Low Energy Adaptive Clustering Hierarchy (LEACH) is one of the Fundamental protocols in this class. It forms node clusters based on the received signal strength and uses local cluster heads thesis as routers to the BS. As the Lifetime of the System is Generally low and Amount of packets sent to base station after aggregation are very low. We have worked on the same problem and proposed a Efficient WSN that utilizes both energy and distance for Selection and Data aggregation in Cluster head Selection process.

Full Paper

IJCST/62/1/A-0513
73

Challenges and Prospects of Internet Banking in Bangladesh: Banker’s (Service Providers) Point of View

Md. Monirul Islam

Abstract

Information technology has become a necessary tool for today’s organizations and banking industry is not an exception. Internet banking is modern technology based computerized system of providing banking systems to client with low cost and quick services. Though Bangladesh was comparatively a late adoptee but now almost all the banks are providing internet banking to their customers. But what is striking is the lack of comprehensive study of internet banking covering organizational and operational sides and exhaustive study as to problem and prospects. Considering the importance of internet banking this study was taken up. The main objectives of this study were to evaluate internet banking structure, operations and to examine the bankers’ (Service Providers) perception towards internet banking in Bangladesh. There are 25 banks were selected for intensive study.

Full Paper

IJCST/62/1/A-0514