Volume 3 Number 2 September 2013


JSRA: Robust Joint Scheduling and Resource Allocation in IEEE 802.16 WiMAX Wireless Networks
P Kavitha, R Uma Rani

Abstract: The WiMax (IEEE 802.16) standard provides a mechanism for creating multi-hop mesh, which can be deployed as a high speed wide-area wireless network. The scheduling and resource allocation has become a significant issue in WiMax wireless networks. In the proposed work, the Joint Scheduling and Resource Allocation (JSRA) Algorithm is introduced to maximize the network throughput and provides the robust resource allocation in WiMax. The proposed methodology supports unicast mechanism which can be applied to any kind of wireless networks. At first, the route request (RREQ) is composed from the receiver. Consequently, the route is discovered and further the data rate is evaluated. The time slot is allotted for discovering the route. The proposed method provides the time slot for each and every authenticated user using JSRA algorithm. Therefore, the waiting time is reduced for the user and thus the user satisfaction is enhanced. The experimental analysis achieves higher network throughput to bandwidth consumption ratio than the existing greedy approaches.


An Enhanced Technique on workload Management for Increasing the Return on Investment in Cloud R-Code Image Steganography
T Ambika, S Karthik

Abstract: With the advent of cloud computing remote and wide spread cloud users are provided various services on demand. Workloads with cloud providers comprises of both transactional applications and long-running batch jobs. With an existing methodology the heterogenic nature of workloads is dealt with a technique of consolidating the varied workload types within the same machine. The infrastructure cost and power consumption will be reduced when compared with the techniques where a separate machine is allocated for each workload type. For each workload a scheduling mechanism chosen and performance goals of each workload are to be satisfied. Traditionally, for the purpose of resource provisioning with virtualization the Virtual Machine (VM) size is estimated for individual machines i.e., one VM at a time. Today the need for energy and resource consumption is drastically increasing for a better return on investment for cloud providers. In this paper, we present the concept of resource provisioning with consolidating multiple VMs could be combined along with the workload consolidation. This paper reduces the underutilization of resources as the unused portion of every virtual machine is utilized. The proposed consolidating technique is more effective in energy consumption and resource utilization when compared with other techniques as consolidation is done at two stages. The heterogeneous workload consolidation depending on the performance goal of each workload type is done at one stage for collocating them at the same machine and the consolidation is also done at the resource provisioning time where the unused portion of the VMs is done at the other stage.


A Novel Routing Scheme Based On Protection Mechanism To Discover Unconventional Disjoint Path In Ad Hoc Wireless Networks
S S Suthaharan

Abstract: The performances of the routing protocols are important since they compute the primary path between source and destination. In addition, routing protocols need to detect failure within a short period of time when nodes move to start updating the routing table in order to find a new primary path to the destination. Meantime, loss of packets and end-to- end delays will increase thereby reducing throughput and degrading the performance of the network. This paper proposes a new algorithm, DBRT (Determined Backup Routing Table), to improve the existing proactive protocols such as DSDV (Destination Sequenced Distance Vector) protocol by creating a backup routing table to provide multiple alternative routes. The DBRT algorithm identifies adjacent nodes for each node in the same range and then selects one of these as a backup next hop according to the available path to the destination. The results show that loss of data packets, throughput and end-to-end delay times between source and destination are improved. The results show that the new protocol does not degrade the network’s performance despite sending extra messages to construct and update the new backup routing table. Simulations (using an NS2 simulator) are undertaken to demonstrate the difference between using a DSDV protocol with or without the proposed schema.


An Efficient Detection And Segmentation Of Pavement Cracks In Remote Sensing Images Using Grow Cut Algorithm
D Napoleon, A Shameena, R Santhoshi

Abstract: Pavement cracking detection and measurement is essential for maintaining a well thoroughfare network. Several algorithms are developed quickly in recent years to detection pavement disasters. Segmentation plays an important role in automated pavement crack detection method. In this proposed work, a binary image has been extracted from gray scale image using auto-tuned threshold value acquired from the Image Euler numbers to identify crack regions. Further, the enhanced image acquired by the auto tuned threshold process is run through with the Grow cut algorithm to get exact crack region. Image quality measurement is imperative for various image processing applications. Assessment of Image quality is intimately related to image similarity appraisal in which quality is based on the dissimilarity between an original image and distorted image. Eminent image quality measure algorithm includes Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR), Normalized Absolute Error (NAE), Normalized Cross-Correlation (NCC), Feature Similarity Index (FSIM) has been assessed between original image and segmented image. The experimental shows that this method is efficient for segmenting pavement crack region from satellite imagery.


A Study on Clustering the Protein Interaction Networks using Bio-Inspired Optimization
R Gowri, R Rathipriya

Abstract: The gist of the paper is to provide an insight about the various clustering using bio-inspired optimization for PIN. The major idea behind the clustering protein-protein interaction network is to identify dense sub-graphs that show significant functional modules in protein-protein interactions. A set of proteins that interact with each other are actors of a specific cellular process is termed as significant functional module which helps to recognize the structure and functional dynamics of the cell. In order to find those modules we have to cluster them based on their interaction. It is very difficult to define the boundaries among clusters and unable to identify the overlapping clusters. The bio-inspired optimization techniques are helpful in overcoming those difficulties.


A Hatchet Way of Cropping the Chops of Crocodiles to Measure the Angle Using Mathematical Concepts for the Geometrical Feature Extraction Using Distance Measure Algorithm
D Napoleon, K Ragul

Abstract: Images are the more powerful tool for effective communication in which image processing aims to extract valid information from an image. Feature Extraction is a process of extracting and generating information to assist the image classification task. Considering the feature as a key role which works well to determine and extract the geometrical features from an image and also to measure the distance between the picture elements. Initially the conversion processes are to be carried out in order to reduce the noise from the samples. The pixel element of the particular region is partitioned in name of segmentation. Morphing technique has been used to crop the chops of the crocodile region for the angle measurement. In order to hit upon the exact picture element, binary conversion techniques have been implemented for the region which is cropped. Extracting Algorithm is used to measure the angle with the distance calculation by means of the mathematical concepts. This method is extremely suitable for measuring the angle to find the distance between the positioned regions.


An Approach to Virtual Machine Placement Problem in a Datacenter Environment Based on Overloaded Resource
T Thiruvenkadam, V Karthikeyani

Abstract: The arrival of the clouds has introduced very strict initiative over the physical resources. A typical resource management system receives queues and finally matches user job requirements with the characteristics of the offered hardware. Server overload is the basis of resource insufficiency and spoils the performance of applications which leads to affect the QoS. Lively consolidation of Virtual Machines is an efficient way to get better use of resources and power efficiency in Cloud data centers. Identifying when it is best to move VMs from a congested host is an aspect of Lively VM consolidation that directly influences the resource consumption and Quality of Service (QoS) delivered by the system. This paper focus on the importance of the detection of server overload and compares the various scheduling algorithms currently used for scheduling virtual machines and also proposes the design methodology of a new algorithm that helps to improve the resource utilization and at the same time energy efficiency.


SAR Image Segmentation Using Superpixel Merging Algorithm
C Karthick, C Saraswathy

Abstract: Image segmentation is an important tool in satellite image processing and serves as an efficient front end to sophisticated algorithm and thereby simplify subsequent processing. It used to extract the meaningful objects lying in the image. The aim of the paper is to obtain the segmentation of the Synthetic Aperture Radar (SAR) image with minimum run time of the algorithm. The algorithm used for the segmentation is named as superpixel merging algorithm. In this paper instead of a pixel, the superpixels are used as basic operation units. The preprocessing stage consists of formation of superpixel. The superpixel merging algorithm is used to merge the superpixel based on the analysis of superpixel. In this edge detection, feature extraction is obtained from the final segmented output. It will use less running time for the superpixels which are not present at the boundaries of different patterns of the image. The proposed algorithm is effectively reducing the process of segmentation and computational complexity.


Memory Based FIR Filter Design on FPGA using Distributed Arithmetic and OBC Coding Technique
M Mahalingam, S Govindaraju

Abstract: This paper provides an efficient implementation of FIR filter without using multipliers. Area complexity in an algorithm of finite impulse response (FIR) filter is mainly caused by multipliers. Among the multiplierless implementation of FIR filter, Distributed Arithmetic (DA) is most efficient technique. In Distributed Arithmetic inner products are precomputed and stored in Look Up Table (LUT), than this precomputed values are added and shifted with number of times equal to the precision of input samples. If filter order increases than Look Up Table size also increases in its basic structure, makes it inefficient for many applications. In order to eliminate exponential growth of LUT with the order of filter we use memory partitioning (slicing) technique. We presented 16-tap FIR filter, with different size of memory partitioning of LUT and combine with OBC Coding. Implementation and synthesis result shows drastic improvement in performance in terms of speed as well as saving in area, with more number of slices.


Building Trust for Web Service Security Patterns
V Prasath

Abstract: Security is a very important aspect for web service technology. Many people find the idea of creating security metrics to be a daunting task. Secure web service discovery aims at finding the best component services that satisfy the end-to-end security requirements between service consumer and service provider. The term "metrics or criteria" refers to specific objectives that have defined security measurement. It helps to select the most suitable security configuration according to a consumer business process and different levels of trust. In this paper, we presents a new pattern methodology for web service to address the security issues and propose a scalable security computation based on a heuristic approach which decomposes the complex problem into smaller sub problems that can be solved more efficient than the original problem. We define quality of service (QoS) in security as a set of security requirements a service provider guarantees. We identified several security parameters classified under different criteria to evaluate web services vulnerabilities. Metrics evaluation can be done through heuristic approach where in the security parameters are assigned prioritized weights which meliorates rank of web services.