Volume 2 Number 4 March 2013
1 |
Simulation and Analysis of GFF at WDM Mux Bandwidth of 13GHz Abstract: Gain Flattening Filter (GFFs), also known as gain equalizing filters, are used to flatten or smooth out unequal signal intensities over a specified wavelength range. This unequal signal intensity usually occurs after an amplification stage (e.g., EDFA and/or Raman). In this simulation model an attempt is made theoretically to flatten the gain of the EDFA which is the vital issue in the DWDM long haul communication. In our proposed design we optimized the GFF to achieve 23.73 dB gain in all associated channels. From the simulation study it is quite clear that the transmission spectrum of the GFF is complimentary to the ASE of the EDFA. The uneven gain of the EDFA is 0.003 dB due to the ASE of the EDFA. |
2 |
Comparative Analysis of Skin Segmentation Methods in Image Mining Abstract: Data mining is the process of analyzing data from different perspectives and summarizing it into useful information. It uses sophisticated mathematical algorithms to segment the data and evaluate the probability of future events. Several techniques are used in developing data mining projects such as clustering, classification, association rule mining, outlier analysis, etc. Some of the research areas in data mining are web mining, text mining, data streams, image mining, sequence mining and multimedia mining and so on. Image mining can be defined as the nontrivial process of finding out valid, original, potentially useful and ultimately understandable information from large image sets or image databases. Image mining is also an interdisciplinary research area which contains digital image processing, database, image understanding, artificial intelligence, pattern discovery, face recognition and so on. Face detection and localization is the task of checking whether the given input image contains any human face, and if so, the location of the human face in the image is determined. In this research work, the performance of segmenting the skin region and time taken to detect skin region are analyzed by Skin segmentation using HSV color and Gradient Vector Flow techniques. The skin regions that are segmented from the facial images are used for face detection. The performances of both segmentation methods are analyzed by applying performance factors and from the experimental results, it is observed that the skin segmentation using HSV color method works more efficient than Gradient Vector Flow method. |
3 |
Significant Feature Set Driven, Optimized FFN for Enhanced Classification Abstract: Neural Networks augmented with back propagation learning is one the extensively used data classification tools. In this paper, a novel classification scheme is elaborated. The method evolved has two steps: In the first step, significant feature selection is made by using decision tree and GA-CFS (genetic algorithm based correlation based feature selection). In the second step, the connection weights of feed forward network (FFN) are optimized using Particle swarm optimization (PSO) and GA. To convalidate the efficacy of the method, it was applied to four benchmark datasets namely diabetes, iris, ionosphere and heart statlog. PSO showed best classification accuracy in the range of 86%-97% for all the datasets considered when compared with BPN and GA based networks. The topology of the PSO optimized FFN was also modest, with a few neurons in the hidden layer. |
4 |
Enhancement of Cascaded Multilevel Inverter for Solar Power Applications Abstract: In this paper different modulation methods have been reviewed for the control of Cascaded Multilevel Inverter (CMLI). The main function of CMLI is to synthesize the desired AC output based on the series connection of DC sources (Solar Panels) with reduced power quality issues. Nowadays, it has been recognized as important alternative resource in low/medium voltage solar powered inverter market. CMLI with reduced common mode voltage rejection and reduced number of power switches are also investigated in this paper. To reduce the counts of DC source, a concept of Z-source interfaced CMLI system is introduced. This paper provides a general survey of modulation techniques and further development of CMLI topology. |
5 |
An Improved Enhancement of Decision- making Analysis to increase the sales promotion using Hyper ETL in Data Mart Abstract- The multiplication in the number of corporations looking for data mart solutions, with the aim of adding major business gains, has created the need for a decision-aid has come near in preferring the right data mart system. Due to the indistinct concept often represented in decision-making procedure, to facilitate decision matrix analysis, with consideration given to both technical and managerial criteria. This paper illustrates the item- wise and place- wise analysis of sales promotion in sales data mart using Hyper ETL (Extract, Transform and Load) tool. We have used Laplacian method for ranking, which enables to make efficient decision- making in order to increase the sales promotion. The main objective of this paper is to determine the alternative courses of action (the movement of sales quantity and also the movement particular item in number of places) from which the ultimate choice to be made. This approach supports the business goals and requirements of an organization and to identify the appropriate attributes or criteria for evaluation. This improvement communicates information quickly and enlarges the aggregation process and helps to take an efficient decision. |
6 |
Cryptographic key generation from multiple fingerprints Abstract: This research deals with new innovative model for biometric Automated Teller Machines (ATMs) for joint account without the remembrance of a Personal Identification Number (PIN). Since the banks in India are not providing ATM card for “joint and other account” system. In this research cryptographic key is generated from the fused biometric fingerprint of the joint account holders which can be used as a PIN at the ATM terminal. Among all the biometrics, fingerprint based identification is highly scalable and proven technique. Proposed model provides high security in authentication for joint account holders. At the time of transaction fingerprint images are acquired from the account holders at the ATM terminal using high resolution fingerprint scanner and the extracted images are preprocessed for enhancing the image, then the minutiae points (ridge ending and bifurcation) are extracted from the preprocessed fingerprint image. The extracted features from the account holders are compared with the existing image in the database for authentication. If there is a match, then the extracted features are fused at the feature level to attain the fused fingerprint model. Finally a 256-bit cryptographic key is generated from the fused fingerprint image. The generated cryptographic key can be used as a PIN at the ATM terminal for transaction. This model reduces complexity with authentication as “authentication is always with you” with high security. It also saves time, cost and efforts compared with the traditional ATMs. |
7 |
RVN-Chord: A Novel P2P Lookup Algorithm for Dynamic Grid Abstract: Grid computing is the next generation distributed system. Grid is an environment that allows sharing of resources that are managed by diverse, independent administrative organizations that are geographically distributed. An efficient resource discovery mechanism is one of the fundamental requirements for grid computing systems, as it aids in resource management and scheduling of applications. Among various discovery mechanisms, Peer-to-Peer (P2P) technology witnessed rapid development and the key component for this success is efficient lookup applications of P2P. Chord is a P2P structural model widely used as a routing protocol to find resources in grid environment. Since improvement on Finger table plays a vital role in efficient Grid Resource Discovery, our work contributes to reconstruction of finger table with new updates. In this paper, we proposed RVN-Chord (Recently Visited Node) which modifies the original Chord’s finger table by adding a new column which stores the ID of Recently visited node. Every new lookup uses that ID to find the successor of the key and the search is minimal if the key matches with RVN.ID. We use NSC_SE simulator and Wireshark packet analyzer in our simulation. Theoretical and experimental analysis shows that RVN-Chord can reduce the number of hops per peer, message per peer, average communication time and memory consumed. |
8 |
Scheduling Homology Modeling in Grid Environment Abstract: Grid computing discipline involves the actual networking services and connections of a potentially unlimited number of ubiquitous computing devices within a grid. This research shows how homology modeling works for given protein sequences in grid environment. The quality of homology modeling is dependent on the quality of the sequence alignment and template structure. In First Come First Served (FCFS) strategy, the protein sequence is scheduled to the resource on first come first serve order and processed until the particular process comes to a completion. On average it takes more time to search for number of sequences and users have to wait for a long time to submit their queries and get the results. To overcome this time delay, FCFS is implemented in Grid. Here the time taken to process the protein sequences gets minimized. The scheduling is done based on the size of the protein sequence such that the system that takes the minimum time to process the particular protein sequence is found out initially in the Grid environment and it is allotted for further processing. |
9 |
Decision Making in Spectrum Allocation in Cognitive Radio using Fuzzy Logic Abstract: Cognitive radio is considered as a software-defined platform that evolves a fully reconfigurable wireless transceiver which automatically adapts its communication parameters to network and user demands. By analyzing the radio environment as well as the primary and secondary user’s environment and have to make a decision for which secondary user can be the best secondary user to access the spectrum from the primary user. Fuzzy logic approach is taken for indicating the environments for accessing the available spectrum by the secondary users. By using this fuzzy logic, possibility for every secondary user can be computed and the available spectrum accessing decision will be based on this possibility. So that, the secondary user who is having the highest possibility among all the secondary users will be considered and permitted to access the primary users spectrum. Thus, it will make more utilization of the available spectrum of the primary users without causing interference. |
10 |
Efficient Memory Utilization for Resource Discovery in a Grid Environment Abstract: One of the fundamental requirements of grid computing is efficient and effective resource discovery mechanism. Resource discovery involves discovery of appropriate resources required by user applications. In this regard various resource discovery mechanisms have been proposed during the recent years. Simple matchmaking rules are used to identify each resource as a part of a certain technical category and the distance travelled in hops is calculated for a certain request. This paper deals with the reduction in the time taken for matching the resources based on the user’s requirements. In this paper Matchmaking algorithm, Flooding algorithm, Swamping algorithm, Random pointer jump algorithm were applied for hops calculation and for efficient memory utilization of the routers. From the simulation results it is found that swamping algorithm gives better result when compared to Flooding, Random pointer jump and matchmaking algorithms. To lessen the burden on the routers, partial information of virtual organization is stored in the routers. This results in efficient memory utilization on the routers. Comparative graphs are also given to show the efficient utilization of memory space and hops calculation of the resource of the routers. |