Research Publications

@proceedings{323,

title = {Combined Off-Line Signature Verification Using Neural Networks},

author = {D. R. Kumar and Ravi Kumar and K. B. Raja and R. K. Chhotaray and S. Pattanaik},

doi = {10.1007/978-3-642-15766-0_99},

isbn = {18650929 (ISSN); 978-364215765-3 (ISBN)},

year  = {2010},

date = {2010-01-01},

journal = {Communications in Computer and Information Science},

volume = {101},

pages = {580-583,},

abstract = {In this paper, combined off-line signature verification using Neural Network (CSVNN) is presented. The global and grid features are combined to generate new set of features for the verification of signature. The Neural Network (NN) is used as a classifier for the authentication of a signature. The performance analysis is verified on random, unskilled and skilled signature forgeries along with genuine signatures. It is observed that FAR and FRR results are improved in the proposed method compared to the existing algorithm. © Springer-Verlag Berlin Heidelberg 2010.},

note = {3},

keywords = {CSE},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{324,

title = {Seeding cluster centers of K-means clustering through median projection},

author = {L. Suresh and J. B. Simha and R. Velur},

doi = {10.1109/CISIS.2010.133},

isbn = {978-076953967-6 (ISBN)},

year  = {2010},

date = {2010-01-01},

journal = {CISIS 2010 - The 4th International Conference on Complex, Intelligent and Software Intensive Systems},

pages = {217-222, 5447429+},

abstract = {K-means Clustering is an important algorithm for identifying the structure in data. K-means is the simplest clustering algorithm. This algorithm uses predefined number of clusters as input. The original algorithm is based on random selection of cluster centers and iteratively improving the results. However there are two major limitations in this approach. First, the need for number of clusters in advance, is difficult since the underlying structure is not known. Second selection of cluster centers randomly in local optima. In addition most of the K-means implementations are memory based structures limiting the data size. In this work, a novel approach to seeding the clusters with the latent data structure is proposed. This is expected to minimize: The need for number of clusters apriory, thereby reducing time for convergence by providing near optimal cluster centers. In addition the implementation of the algorithm is done in SQL, to provide the disk based solution, to handle large data sets, which cannot fit into memory. The proposed solution was tested on both row store and column store databases. The results are promising and the work is under progress to test in different domains. © 2010 IEEE.},

note = {4},

keywords = {CSE},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{337,

title = {Routing analysis and flexibility of ROADMs in all Optical networks},

author = {T. S. Indumathi and T. Srinivas and K. Rajaiah and V. Rajappa},

doi = {10.1109/WAINA.2010.92},

isbn = {978-076954019-1 (ISBN)},

year  = {2010},

date = {2010-01-01},

journal = {24th IEEE International Conference on Advanced Information Networking and Applications Workshops, WAINA 2010},

pages = {75-78, 5480874+},

abstract = {Reconfigurable Optical Add / Drop Multiplexer is a subsystem of All Optical Networks. In this paper we analyze the routing performance of different ROADMs at dynamic traffic using Bernoulli, Poisson and Pascal distribution models. A theoretical routing power model is considered and through simulation results, the routing power and blocking probability have been evaluated to achieve better performance and flexibility. © 2010 IEEE.},

note = {2},

keywords = {OTHER},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{339,

title = {Hybrid fingerprint matching using block filter and strength factors},

author = {D. R. Kumar and Kiran Kumar and K. B. Raja and R. K. Chhotaray and S. Pattnaik},

doi = {10.1109/ICCEA.2010.98},

isbn = {978-076953982-9 (ISBN)},

year  = {2010},

date = {2010-01-01},

journal = {2010 2nd International Conference on Computer Engineering and Applications, ICCEA 2010},

volume = {1},

pages = {476-480, 5445783+},

abstract = {Fingerprint verification is a socially accepted biometric method for identification of a person. In this paper we propose Hybrid Fingerprint Matching using Block Filter and Strength Factor (HFMBFS). The minutiae and ridge based methods are combined to verify the fingerprint matching using strength factors Alpha (α) and Beta (β). For minutiae and ridge extraction Block Filter and Hough Transform are used respectively. It is observed that the matching percentage of two different fingerprints is improved compared to the existing algorithms. © 2010 IEEE.},

note = {1},

keywords = {CSE},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{315,

title = {Voice transmission over local area network using bluetooth},

author = {A. Biradar and R. C. Thool and R. Velur},

doi = {10.1109/TENCON.2009.5395970},

isbn = {978-142444547-9 (ISBN)},

year  = {2009},

date = {2009-01-01},

journal = {IEEE Region 10 Annual International Conference, Proceedings/TENCON},

pages = {5395970+},

abstract = {Bluetooth is a low cost, low power; short-range radio technology intended to transfer data between cell phones, PDAs and other portable devices. Bluetooth is also used as a medium for voice communication between cell phones. This paper aim is to develop an application that allows a user to make use of mobile phones enabled with Bluetooth and PC's connected over Local Area Network (LAN) in order to communicate with other user who is out of Bluetooth's range. ©2009 IEEE.},

note = {1},

keywords = {OTHER},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{321,

title = {Analysis and prediction server with column store database - A case study in telecom churn},

author = {L. Suresh and J. B. Simha and R. Velur},

doi = {10.1109/TENCON.2009.5395982},

isbn = {978-142444547-9 (ISBN)},

year  = {2009},

date = {2009-01-01},

journal = {IEEE Region 10 Annual International Conference, Proceedings/TENCON},

pages = {5395982+},

abstract = {One of the major concepts in business analytics is to identify the anomalies over time, also called as trend analysis. This can be easily done in pivot tables by using time as one of the dimensions, usually across columns. However, the trending information itself is insufficient to make any quick and insightful observations. Ranking the time series to identify the similar units of information can accelerate the analysis process. Similarly projection of the time series to the future will help the decision maker to proactively build alternate plans for differing scenarios. For doing such an analysis and modeling it becomes necessary to have aggregated data on demand. Current breed of row store database have limited capabilities to provide the response time required for such an analysis and modeling. Hence, column store database are expected to be better alternatives, for these types of problems. In this work the attempts made by the authors to develop such a system named as 'rePivot' are presented. The proposed frame work consists of three modules namely - a column store database to provide quick access to data, a time series ranking module and a probabilistic forecasting module. A case study of the proposed frame work in churn analysis and modeling in telecom has been carried out to test the suitability of framework for industrial applications. Application of the framework has shown promising results. Work is under progress to develop additional modules for survival analytics of individual entities in the database. ©2009 IEEE.},

note = {0},

keywords = {CSE},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{330,

title = {rePivot - Developing analysis services with column store database},

author = {J. B. Simha and L. Suresh and R. Velur},

doi = {10.1109/ICETET.2009.191},

isbn = {978-076953884-6 (ISBN)},

year  = {2009},

date = {2009-01-01},

journal = {2009 2nd International Conference on Emerging Trends in Engineering and Technology, ICETET 2009},

pages = {301-306, 5394999+},

abstract = {one of the major concepts in business analytics is to identify the anomalies over time, also called as trend analysis. This can be easily done in pivot tables by using time as one of the dimensions, usually across columns. However, the trending information itself is insufficient to make any quick and insightful observations. Ranking the time series to identify the similar units of information can accelerate the analysis process. Similarly projection of the time series to the future will help the decision maker to proactively build alternate plans for differing scenarios. For doing such an analysis and modeling it becomes necessary to have aggregated data on demand. Current breed of row store database have limited capabilities to provide the response time required for such an analysis and modeling. Hence, column store database are expected to be better alternatives, for these types of problems. In this work the attempts made by the authors to develop such a system named as 'rePivot' are presented. The proposed frame work consists of three modules namely - a column store database to provide quick access to data, a time series ranking module and a probabilistic forecasting module. A case study of the proposed frame work in churn analysis and modeling in telecom has been carried out to test the suitability of framework for industrial applications. Application of the framework has shown promising results. Work is under progress to develop additional modules for survival analytics of individual entities in the database. © 2009 IEEE.},

note = {0},

keywords = {CSE},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{342,

title = {Hardware in Loop implementation and analysis of a neural augmented fault tolerant flight controller for a high performance dynamic fighter aircraft model on a target digital signal processor},

author = {N. Ramrao and T. V. R. Murthy},

doi = {10.1109/TENCON.2009.5396191},

isbn = {978-142444547-9 (ISBN)},

year  = {2009},

date = {2009-01-01},

journal = {IEEE Region 10 Annual International Conference, Proceedings/TENCON},

pages = {5396191+},

abstract = {The high performance fighter aircraft achieves a high angle of attack (up to 90°) and superb maneuverability because of the presence of leading edge flaps. In this paper we make an attempt to achieve reconfigurable control of this high performance fighter aircraft during auto-landing[1], using Extended Minimal Resource Allocating Network (EMRAN) augmented controller [2] with additional faults and validate the same through software simulation using MATLAB® Simulink® and also hardware implementation of the controller on a target Digital Signal Processor (DSP) using Hardware-in-the-loop simulation technique. A high fidelity aircraft model [1] with seven control surfaces, including leading edge flaps (LEF) in comparison to the low fidelity model that employs only five control surfaces is employed in the simulation as well as hardware implementation. Also while reconfiguring the aircraft model will be made dynamic where in multiple faults can occur at different instances. ©2009 IEEE.},

note = {1},

keywords = {ECE},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{308,

title = {MHD flow and heat transfer in the flow of a power law fluid over a non-ISO thermal stretching sheet},

author = {V. Rajappa and K. V. Prasad},

doi = {10.1109/ICETET.2008.273},

isbn = {978-076953267-7 (ISBN)},

year  = {2008},

date = {2008-01-01},

journal = {Proceedings - 1st International Conference on Emerging Trends in Engineering and Technology, ICETET 2008},

pages = {1077-1082, 4580063+},

abstract = {This article presents a numerical solution for the MHD flow of an electrically conducting non-Newtonian power law fluid over a semi-infinite non-isothermal stretching sheet with internal heat generation/ absorption. The flow is caused by linear stretching of a sheet from an impermeable wall. Thermal conductivity is assumed to vary linearly with temperature. The intricate coupled non-linear boundary value problem has been solved by a combination of Keller box method and shooting technique. An important observation of our study is that the velocity boundary layer thickness and thermal boundary layer thickness decreases with the increase of power law index in the presence / absence of variable thermal conductivity. © 2008 IEEE.},

note = {0},

keywords = {OTHER},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{328,

title = {Synchronization of traffic pattern in mobile ad-hoc network with genetic algorithm},

author = {A. Biradar and S. Panda and R. C. Thool},

doi = {10.1109/ICETET.2008.55},

isbn = {978-076953267-7 (ISBN)},

year  = {2008},

date = {2008-01-01},

journal = {Proceedings - 1st International Conference on Emerging Trends in Engineering and Technology, ICETET 2008},

pages = {246-249, 4579904+},

abstract = {MANET is a challenge for every network designer or router manufacturer for the simple reason it is unpredictable for a given space and time. This bottleneck of unproductively designed ad-hoc system can be tracked by the traffic pattern of the system using genetic algorithm. The dynamic nature of the network makes the job difficult to attain accuracy in the routing level. To reduce the cost of calculating the expected traffic in the network in the current proposed system we will keep a database of well defined topologies and their respective traffic load under given number of node count. On finding a topology matching in the MANET system we can get the existing traffic load from the database. For finding a matching topology in the MANET model we will use the basic function of genetic algorithm. The search space of the network topology has been made optimized with the genetic representation of the network in binary bit string representation of 0 and 1. To find the traffic pattern we will use DSRFLOW, which in turn give us the data security in MANET data transfer. The propose solution to use is tracing on the concept of GA for the base of reconciling of the network topology as on a regular interval of sensing the nodes by sending control packet in the network. © 2008 IEEE.},

note = {1},

keywords = {CSE},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{329,

title = {Efficient software test case generation using genetic algorithm based graph theory},

author = {V. Rajappa and A. Biradar and S. Panda},

doi = {10.1109/ICETET.2008.79},

isbn = {978-076953267-7 (ISBN)},

year  = {2008},

date = {2008-01-01},

journal = {Proceedings - 1st International Conference on Emerging Trends in Engineering and Technology, ICETET 2008},

pages = {298-303, 4579914+},

abstract = {In orthodox software testing approach we generally use modeling based testing approach for generating the test cases of a given problem. This leads to confusion of the test input and the expected output for a given test case. More over we also can miss some of the test cases due to lack of clarity in the test paths. To over come such sort of predictive modeling we propose graph theory based genetic approach to generate test cases for software testing. At first we will create a directed graph of all the intermediate state of the system for the expected behavior of the system. Then we will create a population of all the nodes of the graph as the base population of genetic algorithm. From this population we can find a pair of node the parents and perform genetic crossover and mutation on them for the getting the optimum child nodes as the out put. We should continue this process of genetic operation until all the nodes are covered or any of the nodes, which are visited more than once, should be discarded form the population. Then follow the same process for the generation of test case in the real time system. This technique will be more concrete in case of network testing or any of the system testing where the predictive model based tests are not optimized to produced the out put. © 2008 IEEE.},

note = {38},

keywords = {CSE},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{340,

title = {Genetic algorithm based secure authentication protocol with dual central server and token authentication in large scale mobile Ad-Hoc networks},

author = {S. Panda and V. Rajappa and A. Biradar},

doi = {10.1109/GLOCOM.2008.ECP.1061},

isbn = {978-142442324-8 (ISBN)},

year  = {2008},

date = {2008-01-01},

journal = {GLOBECOM - IEEE Global Telecommunications Conference},

pages = {5552-5557, 4698836+},

abstract = {The major challenges in mobile ad-hoc network (MANET) are that, it does not have any predefined topology and channel path for communication. In the wireless communication domain the limitation of channel security is enforced by the current technical limitation Wireless communication can not support the orthodox protocol and topology modeling because of continuous change in network ordering and range with variable number of wireless element present in the network. On the other hand we have constraints of the mobile device capabilities in term of memory, processing speed and usage of power. As a result of such limitations we have challenges to provide the channel security or the authorization of access by the available algorithms. Besides the general security objectives like authentication, confidentiality, integrity, availability and non- repudiation, the ad hoc routing protocols should also address location confidentiality, cooperation fairness and absence of traffic diversion. In this paper we attempt to analyze threats faced by the ad hoc network environment and provide a classification of the various security mechanisms. We analyzed the respective strengths and vulnerabilities of the existing routing protocols and suggest a comprehensive solution using dual authentication server for the trust management in a ad-hoc network. We propose the ad-hoc routing protocol must start with a node as server and from that onward when ever a new connection is established it should assign a node number generated from the server and maintain a table entry for that. When ever the second node comes in the network it will be an image server for the main server node for re-authentication any new node comes we will take it for a new entry in the routing table and find a trusted network value for that node. To find the new node value the genetic algorithm is used. And to maintain the natural path we follow ANT based routing protocol. © 2008 IEEE.},

note = {0},

keywords = {CSE},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{341,

title = {Novel and efficient clustering algorithm using structured query language},

author = {L. Suresh and J. B. Simha},

doi = {10.1109/ICCCNET.2008.4787767},

isbn = {978-142443595-1 (ISBN)},

year  = {2008},

date = {2008-01-01},

journal = {Proceedings of the 2008 International Conference on Computing, Communication and Networking, ICCCN 2008},

pages = {4787767+},

abstract = {Clustering becomes an indispensable requirement while dealing with immense volume of data. Since Database Management tools does not provide an inbuilt mechanism to cluster datasets of higher magnitude it inevitably requires an external module to perform the same. This external module should be devised specifically to deal with the data extracted from the data source. There exists myriad of techniques for clustering and the conventional approaches deficits like excessive time and computational complexity. Computational complexity becomes a factor of consideration when data is extracted voluminously and the process of clustering and filtration is performed as a subsequent separate operation. The concept proposed is crafted with an object of eliminating the inherent redundancies in the accustomed practice. The proposed solution relies on exploiting the processing power of the Database Management tool by streamlining the SQL used for data extraction. And hence a procedure is formulated that can combine data retrieval and clustering to one single operation and leave it to DBMS without letting it to dissipate to the adjoining tiers. © 2008 IEEE.},

note = {0},

keywords = {CSE},

pubstate = {published},

tppubtype = {proceedings}

}

@proceedings{400,

title = {A Literature Survey—Application of Image Processing for the Effective Diagnosis of Rod Cone Dystrophy Using OCT},

author = {K. Aswini and S. Gopal and Husna Tabassum and D. Loganathan and Smitha Kurian and S. Sarumathi},

url = {https://link.springer.com/chapter/10.1007/978-981-97-7616-0_17},

doi = {10.1007/978-981-97-7616-0_17},

isbn = {978-981-97-7616-0},

journal = {Proceedings of the First International Conference on Data Engineering and Machine Intelligence},

pages = {243-256,},

publisher = {Springer Nature Singapore},

address = {Singapore},

abstract = {Rod and cone cell function is impacted in Rod-Cone Dystrophy (RCD), an inherited retinal condition that causes progressive vision loss. To stop further damage to retina, early detection is crucial. The development of Optical Coherence Tomography (OCT) is a useful tool for the early diagnosis of RCD. However, by employing OCT to diagnose RCD accurately and effectively needs a sophisticated image processing method. In order to diagnose RCD at early stage using OCT, this research reviews the state-of-the-art image processing approaches. The study discusses several pre-treatment procedures, segmentation algorithms, and feature extraction strategies that have been applied to Oct image analysis in RCD.},

keywords = {ISE},

pubstate = {published},

tppubtype = {proceedings}

}