Volume 5, Issue 3, March 2014 Edition
Publication for Volume 5, Issue 3, March 2014 is inprocess.
Register for IJSER Research Forum
IJSER Xplore Research Paper Database
Pages
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
A ROBUST METHOD OF SELFEVALUATION FOR UNDERWATER ACOUSTIC LOCALIZATION[FullText ] Rohini.S. Nair, V.GopiUnderwater localization (UWL) is one of most important technologies in Underwater Wireless Sensor Networks (UWSNs) since it plays a critical role in many applications. The variable speed of sound and the long propagation delays under water in addition to bandwidth limitations all pose a unique set of challenges for localization in UWSN. The nodes assigned are anchor and Unlocalized (UL) nodes, these nodes are permanently moving due to ocean current or self motion. In this paper, a new sequential algorithm for time synchronization and localization named as partial markov decision process. The time synchronization is calculated by analyzing the packet exchange between anchor and UL node. The localization accuracy is improved by self evaluating of UL node using Accelerometer and Compass. The complexity in localization can be improved by self evaluation of nodes. In the simulation part, the performance can be evaluated by comparing the algorithm with two benchmark localization methods and is done by using Cramer Rao bound (CBR) method. The results show that the algorithm attains localization accuracy and time synchronization using only two anchor nodes.

A heuristric based Artificial Bee Colony algorithm for Optimmu Placement of PMU for complete Observability of Power System[FullText ] R. Nafeena, C. Muthu SeralathanThis paper presents a novel approach to optimal placement of Phasor Measurement Units (PMUs) for state estimation. At first, an optimal measurement set is determined to achieve full network observability using heuristic approach during normal conditions. An Artificial Bee Colony algorithm is used as an optimization tool to obtain the minimal number of PMUs and their corresponding locations while satisfying associated constraint. The integer based artificial bee colony optimization method and heuristic method are tested on IEEE 14bus, 30bus, 57bus and 118bus systems.

Trafic Vibrations on Historical sites of Ahmedabad[FullText ] Vishvam Pancholi, Janak Shah, Ronak PatelImpact assessment of the historical sites due to traffic vibration will be considered in the Present study. Historical sites are very valuable from cultural point of view. Sometimes vibrations induced by traffic cause Architectural and Structural damage to nearby historical sites. Therefore, it is necessary to predict and assess the vibration effect on historical sites.it is also necessary to mitigate the effects of traffic vibrations along with remedial measures. In this paper we will discuss major sites surrounded by heavy traffic vehicular movements by using vibration analyzer instruments on historical sites like Astodiya gate, Raipur gate.We have considered for the objective determined the effect of traffic vibrations on it and give level of traffic vibration generated by road. A comparative result shows the level of traffic vibrations

Classification of Hepatitis C Virus using Modified Particle Swarm Optimization and Least Squares Support Vector Machine[FullText ] Omar S.Soliman, Eman Abo ElhamdHepatitis C Virus is one of the most dangerous diseases all over the world. It affects millions of people every year and could takes man's life. Many classification algorithms have been applied for its diagnoses and treatment. This paper proposes a hybrid classification system for HCV diagnosis, using Modified Particle Swarm Optimization algorithm and Least Squares Support Vector Machine. Principle Component Analysis algorithm is employed to extract features vector. As LSSVM algorithm is sensitive to the changes of values of its parameters, ModifiedPSO Algorithm was used to search for the optimal values of LSSVM parameters in less number of iterations. The proposed system is implemented and evaluated on the benchmark HCV data set from UCI repository of machine learning databases. It was compared with another classification system, which utilized PCA and LSSVM. The experimental results showed the superiority of the proposed system that was able to obtain classification accuracy of 98.86% versus 96.12% of the other system.

Design and Implementation of 12C Communication Protocol onFPGA for EEPROM[FullText ] Radha R C, Ravuri Aneesh KumarThe 12C protocol was given by Philips Semicounductors in order to allow faster to communicate with slower devicies and also allow devicies to communicate with each other over a serial data bus without data loss.

A Research on Improved Controllers to Stabilize the Frequency in MultiArea Interconnected Power Systems[FullText ] NgocKhoat Nguyen, Qi Huang, and ThiMaiPhuong DaoThis work deals with a stability problem of the frequency in multiarea interconnected power systems resulting from load disturbances at generation stations by using different controllers. Traditionally, an integral controller can be used to obtain the zerosteadystate frequency deviation with the poor control performances, i.e., high overshoots and long settling times. In order to achieve the better control performances, improved control strategies must be investigated by applying advance controllers, such as PID, fuzzy logic, artificial neural network, etc. For the most practical and efficient application, fuzzy logic controllers are used to make sure that the frequency in multiarea power system tends to the nominal values as soon as possible after the appearance of load disturbances. Simulation results of this control technique are carried out for different multiarea interconnected power systems in comparison with the conventional control strategy.

Printed and Handwritten Hindi/Arabic Numeral Recognition Using Centralized Moments[FullText ] Mohamed H. Ghaleb, Loay E. George, and Faisel G. Mohammed Printed and handwritten numerals recognition plays a vital role in postal automation services. The major problem in handwritten recognition is the huge variability and distortions of patterns. The aim of the current research work is to develop fast and efficient method to recognize Hindi printed and free handwritten numerals objects. In this research, the introduced method for extracting features from patterns is based on the relative density distribution of each numeral object; specifically it depends on the centralized moments. This method gives sufficient results to recognize the printed and highly stylized handwritten numeral images. The attained recognition rate is 97.47% for the printed numeral images with total number of samples equal (198) samples and 95.55% for the highly stylized handwritten numeral images with total number of samples equal (90) samples, while, the attained recognition rate is unacceptable when the system is applied for a handwritten numeral samples which have wide differences in their shapes with total number of samples equal (4500) samples. The attained recognition rate is (74.93%). Each tested numeral image is scanned with scanning resolution of 300 dpi.

Random Forest Technique for Email Classification[FullText ] Bhagyashri U. Gaikwad, P. P. HalkarnikarEmail has been an efficient and popular communication mechanism as the number of Internet users increase. Therefore, email management is an important and growing problem for individuals and organizations because it is prone to misuse. The blind posting of unsolicited email messages, known as spam, is an example of misuse. Spam is commonly defined as the sending of unsolicited bulk email that is, email that was not asked for by multiple recipients. The classification algorithms such as Neural Network (NN), Support Vector Machine (SVM), and Nave Bayesian (NB) are currently used in various datasets and showing a good classification result. This paper described classification of emails by Random Forests Technique (RF). RF is ensemble learning technique. A data mining technique called “Ensemble learning” consists of methods that generate many classifiers like decision trees and aggregates the results by taking a weighted vote of their predictions is developed. First the Body of the message is evaluated and after preprocessing the tokens are extracted. Then using a term selection method, the best discriminative terms are retained and other terms are removed. Then iterative patterns are extracted and a feature vector is built for each sample. Finally Random Forest is applied as classifier. If identified category is 0 then it is nonspam otherwise if identified category is 1 then it is spam.

OPTIMAL LOCATION AND PARAMETER SETTINGS OF TCSC UNDER SINGLE LINE CONTINGENCY USING PSO TECHNIQUE[FullText ] S.Raju, G.MadhaviThe Flexible AC Transmission System (FACTS) in a Power System improves stability, reduces the cost of generation, losses and also improves loadability of the system. In order to use FACTS devices in a Power system to maximum extent, optimal location of FACTS devices is an important consideration. This paper presents the optimal location of series FACTS device for eliminating or minimizing the line overloads under single contingency. Among various types of FACTS devices TCSC with reactance control is used to control the power flow in congested lines for different operating conditions. The optimal location of the device is ascertained by using CSI and optimal setting of the device by PSO technique. The effectiveness of this method has been tested on IEEE14 bus system using MATLAB Programming.

An Efficient Approach of Support Vector Machine for Runoff Forecasting[FullText ] Satanand Mishra, Vinayak Choubey, S.K.Pandey, Dr. J.P.ShuklaThis research presents the survey of the one of the most recent and broadly applicable approach of Data mining that is support vector machine. In this research, a hybrid cum combined model approach for runoff forecasting with the help of the technique of chaotic identification, least square support vector machine and wavelet analysis have been considered for rainfallrunoff prediction and accuracy of this new approach evaluated through the statistics of root mean square error (RMSE), mean absolute error (MAE), and correlation (R).

Adaptive elearning using Granulerised Agent Framework[FullText ] Priya Sahai, Manuj DarbariIn this paper we propose a mutliagent approach to the problem of recommending relevant and adequate course work to the student in elearning environment. Our framework illustrates granulerised approach that provides dynamic binding of various course mixes stored in the form of data sets providing the user with relevant data work in real time. A flexibility , adaptability and interactiveness is achieved through agents that autonomously and intelligently uses fuzzy relational models to generate decisions from imperfect input sources. We also suggests a right course selection and pace setting algorithm that agent uses to provide best course mix for the student depending on various parameters such as his/her skill set, learning pace and others.

Prediction of Closing Price of Stock Using Artificial Neural Network[FullText ] H.B.Kekre, Hitesh Makhija, Pallavi N.HalarnkarThis paper analyses the theories used to explain the stock market movements. It uses the Chaos Theory, which essentially states that the stock market is a chaotic system. It then uses Artificial Neural Networks to learn this chaotic system. The Learning algorithm used is the Error Back Propagation Learning Algorithm. The Artificial Neural Network use is the Feedforward Artificial Neural Network. The Neural Network predicts the next day closing price of a stock.

Improved Hierarchical Sparse Method with application to Offline handwritten Arabic character recognition[FullText ] Ramadhan Abdo Musleh Alsaidi, Hong Li, YantaoWei and Rokan KhajiOffline handwriting Arabic character recognition has received increased attention in recent years. In this paper, we introduce a novel technique to enhance the recognition rate of offline handwritten Arabic characters within the low computation time based on developing Hierarchical Sparse Method offered. The developed principles of this algorithm can be specified as scaling and kmean techniques. For scaling level, the main advantage is the avoidance of attributes in greater numeric ranges dominating those in smaller numeric ranges, at the same time, the kmean used to obtain an efficient template selection method included in the derived kernel for developing the object recognition performance. In this paper, the experimental results comprise the developed algorithm matching with other existing techniques.

Pages
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
