International Journal of Intelligent Engineering and Systems

6

Authors: Emansa Hasri Putra, Risanuri Hidayat, Widyawan, I Wayan Mustika

Routing process in Wireless Multimedia Sensor Networks (WMSNs) has its own challenges to pass multimedia applications from source nodes to a sink. The WMSNs have specific characteristics consisting of multi-hop network, restricted bandwidth, and limited power resources in the WMSNs. Additionally, multimedia applications require specific Quality of Service (QoS). This paper proposes a routing optimization based on the Ant Colony Optimization (ACO) for multimedia transmission in the WMSNs. The routing optimization can deliver multimedia data within an average throughput of 164.65 kbps and provide better QoS than DSDV and AODV routing protocols.

8

Authors: M. LawanyaShri, Balamurugan Balusamy, S. Subha

Cloud computing empowers sharing of computing resources dynamically to a wide range of end users. The workload on the cloud resources is amassed immensely with the advancement of new applications. To increase the utilisation of the virtual machine in the cloud data center, an efficient load balancing technique is predominantly crucial. Load balancing is the heart of the data centre in cloud environment which makes all the virtual machines accomplish the same amount of workload and helps the virtual machines deliver the services with minimal time delay. To ensure optimum resource usage and fast processing time, M/G/1 is developed with vacation queuing model (VQM-LB) for adjusting the workload of the under-loaded virtual machine in a cloud environment. The proposed system uses vacation and threshold policy to efficiently control the workload level of each virtual machine in the data center , and reduces the energy consumption and cost accordingly. The result obtained in our proposed model assuredly bounces optimum solutions and apparently indicates the triumph of an efficient load balancing of tasks on minimal energy consumption and cost.

9

Authors: Salem Mohammed, Khelfi Mohamed Fayçel, Bachir Bouiajra Rochdi

Harmony search (HS) is a new meta-heuristic optimization method inspired from improvisation process in jazz music searching for a perfect state of harmony. HS has been improved based on parameters settings updating and hybridization. In this paper, HS and four of its variants are compared via Wilcoxon and Friedman nonparametric tests using benchmark functions. Statistical analysis highlights the Global best harmony search as the best variant. Harmony Search algorithm and its variants are also applied to adjust the gains of PID controller of an inverted pendulum nonlinear system giving remarkable results when compared to genetic algorithm.

9

Authors: Rajesh Kaluri, Pradeep Reddy Ch

Sign language is the most communal language used by the deaf people among themselves for any kind of communication. People who are not familiar with Sign Language face difficulty in interacting with the deaf people. Hence, an effective system should be established to attain and discriminate the sign language. In our proposed system, we developed a framework for extracting and recognizing the sign gesture language and which is classified into four stages like noise removal using median filter, Segmentation using Modified Region Growing Algorithm (MRGA), feature extraction and recognition using Adaptive Genetic Fuzzy Classifier (AGFC). We have used Genetic algorithm with Fuzzy classifier to find out the optimal rules generated by Fuzzy classifier.

10

Authors: M. M. Gowthul Alam, S. Baulkani

Nowadays data growth is directly proportional to time and it is a major challenge to store the data in an organised fashion. Document clustering is the solution for organising relevant documents together. In this paper, a web clustering algorithm namely WDC-KABC is proposed to cluster the web documents effectively. The proposed algorithm uses the features of both K-means and Artificial Bee Colony (ABC) clustering algorithm. In this paper, ABC algorithm is employed as the global search optimizer and K-means is used for refining the solutions. Thus, the quality of the cluster is improved. The performance of WDC-KABC is analysed with four different datasets (webkb, wap, rec0 and 7sectors). The proposed algorithm is compared with existing algorithms such as K-means, Particle Swarm Optimization, Hybrid of Particle Swarm Optimization and K-means and Ant Colony Optimization. The experimental results of WDC-KABC are satisfactory, in terms of precision, recall, f-measure, accuracy and error rate.

10

Authors: M. Siva, R. Balamurugan, L. Lakshminarasimman

This paper presents the application of Water Wave Optimization Algorithm (WWOA) for solving economic dispatch problems including practical generator constraints. WWOA is inspired by the shallow water wave theory. The efficiency of the WWO Algorithm for solving economic dispatch problems is demonstrated by implementing it on three test systems having three, six and fifteen generating units with non-linear characteristics of the generator such as ramp-rate limits, prohibited operating zones including the system transmission losses. The results of the proposed approach have been compared with existing results obtained by other solution techniques. The test results reveal the capability of the proposed algorithm as an effective tool for solving various economic dispatch problems in a power system.

10

Authors: Rainu Nandal, Rahul Rishi

The rapidly rising and widespread use of database technology - including heterogeneous, geo-referenced and multidimensional data - there is a growing interest in developing new techniques for extracting knowledge from data. These techniques are the subject of the emerging field of Knowledge Discovery in Databases (KDD). In this paper, an algorithm to mine the intelligence rules from the input data using gravitational search (GS) algorithm is proposed. Initially, the significant patterns are selected from the input database with the use of measures like density ratio, sequence index, and density index. Subsequently, R-tree is modelled from the reduced dataset and is solved using the recently developed optimization algorithm. The rules obtained from the GS algorithm are useful for knowledge discovery. The experimentation is carried using the datasets to evaluate the performance in terms of mined rules and the computation time. The Performance metrics employed are a number of optimized rules, computation time and memory usage. From the analysis, it can be seen that the number of optimized rules decrease with increase in threshold and support.

10

Authors: Sasikumar Karuppusamy, Madiajagan Muthaiyan

The IT systems sent to satisfy their business functionalities and to provide good Quality of Service (QoS) parameters such as availability, scalability and performance. The failures are normal rather than exceptional in case of cloud computing. To improve system availability enhancing the system performance is an important factor. For that replicating the popular data to multiple sites is a best choice as the users can access the data from a nearby site. In this paper, we proposed a dynamic replication and placement algorithm to enhance the performance of the software system. We calculate popularity degree and replica factor to identify the suitable file to replicate. From this the most accessed file and the file need to be replicated is found and the replicate of the file is created. Then the proposed algorithm is used to place the replicas in the suitable location. We compare the performance of the technique with the existing technique.

10

Authors: Srinivasan Gopalan, Murugappan Shanmugam S

Interference is the most important issue in Wireless Sensor Network (WSN) as it causes severe degradation during communication. Also, as sensor nodes are power constraint, hence the system is expected to operate for longer duration without any delay during communication. In order to overcome these issue a delay sensitive and interference aware MAC protocol for WSN is proposed. In this paper, first MAC protocol is proposed which divides network into layer and assign node to each layer to avoid data loss in the network. Once network is divided to the multi-layer, timeslot is assigned to each node. To utilize the channel efficiently, timeslot steal mechanism is implemented. To avoid interference, a hybrid prioritization scheme is implemented to prioritize the packet queue based on weight of the packet queue and based on the idle time of the prioritized packet.

10

Authors: Kuruva Lakshmanna, Neelu Khare

DNA Sequence mining helps in discovering the patterns which can occur frequently, structures of DNA in DNA data sets. Frequent pattern mining is a central strategy for affiliation guideline discovery, but existing calculations experience the ill effects of low effectiveness or poor error rate on the grounds that natural groupings vary from general successions with more attributes. In our last work, we proposed Prefix Span with Group Search Optimization (PSGSO) to optimize the mined results from the Prefix Span method. We propose a new method called Frequent DNA Sequence Mining using Optimization (FDSMO) which combines Frequent Biological Sequence based on Bitmap (FBSB) and Hybrid of Firefly and Group Search Optimization (HFGSO) in this paper. The FDSMO process includes three stages: (i) applying the Frequent Biological Sequence based on Bitmap (ii) calculate length, width and regular expression (iii) optimization using HFGSO. The exploratory results demonstrate that FDSMO performs better than the existing methods, both in terms of running time and scalability.

10

Authors: Madhwaraj Kango Gopal

Software metrics play an important role in the planning and control of software development. The quality of the software is the degree to which a finished software product proves to be efficient to its specification. Recently, Object-Oriented (OO) methodology has emerged as a major approach for the software development in both scientific and business applications. As the necessity for productive software is increasing, the OO design technique for constructing software is proving to be a powerful method for developing efficient software systems. Generally, the software used by the organizations and the individuals is the one that is owned by the organization which develops it. But open source software is the one which is available free for the user and can be altered based on their needs. In such kind of the software, the maintainability and reliability play a major role regarding the quality of the software developed. In this paper, we have empirically analyzed the various design metrics of different versions of software using JDepend tool and their effect on maintainability is tested. Further, the reliability of the software is measured using Rayleigh’s model. The relationships between maintainability and reliability is found out by checking how these metrics influence the quality of software.

10

Authors: G Nagaraja, Pradeep Reddy Ch

Instant collaborative group communication can be achieved by deploying Mobile Ad Hoc Networks (or MANETs) without any pre-plan and pre-existing infrastructure setup. However, the curbs thrown by these networks, motivate the necessity of a group key management framework to secure data traffic. In this context, significant research work done in the last decade and proved that the trust based frameworks deliver better performance than others. In MANETs, trust quantified as the belief held by a node about another node. It is evaluated based on the assessing node direct experiences with assessed node itself, if they are neighbors and recommendations from remote nodes in the network with which assessed node has a direct experience, if they are remote. However, the framework based on recommendations by other nodes in the network might ultimately lead to system failure due to the menace of unfair recommendations by selfish and malicious nodes. Lying and on-off attacks on the system are such type of attacks. This motivates us to work on a framework which is immune to aforesaid attacks and should be lightweight due to node’s has limited computing resources. We propose a framework which scrutinizes dishonest recommendations based on the trust threshold value of remote nodes, distance from mean of trust recommendations and by maintaining a history of recommendations. Our simulation outcomes prove that proposed framework can deliver better network performance and resistance to lying and on-off attacks.

12

Authors: M. Parimala, Daphne Lopez

Spatial data mining is a rapidly growing field for analysing the data related to space and time. Nowadays most of the applications are based on these factors, so numerous data mining algorithms are developed for spatial characterization and to analyse the spatial trends. The spatial trend analysis determines the change in pattern of some non-spatial attributes on neighbourhood objects. In this paper, we identify spatio-temporal mobility pattern on the dynamics of Epidemic disease (H1N1) that plays a significant role in analysing the outbreak of an infectious disease. Modelling the transmission among the human population with respect to time and space leads to improved understanding of transmission mechanisms. A compartmental model is designed to characterize the disease dynamics of a random variable extracted from binomial and multinomial distribution. ArcGIS tool is used to visualize the mobility distribution of the infected host spatially and yields an output of frequent mobility locations with respect to different time slices. The results thus obtained would help the district administrative authorities to take strategic decisions and prevent the spread of the disease.

7

Authors: Prashanth Kumar Bolisetty, Prasanth Yalla

The Software architecture is generated by using the interfaces and structural components of the software systems in an organization. Software architecture, along with the structure and behavior, also concerned with functionality, performance, reuse, economic and technological constraints etc. In software its components are related to one another in large variety of ways. The main intension of our research is to build the component based software architecture with adaptive configurations using Particle Swarm Optimization and clustering techniques. Building architecture is an inspiring progression. In this paper, we will propose Hybrid Particle Swarm Optimization (PSO) – Cuckoo Search (CS) algorithm for developing an adaptive software architecture based on the Process control model. Initially, the components are selected based on testcases generated. After that, adaptive architecture will be built by using PSO - CS on the basis of clustering results. The architecture is built along with its functional requirements, responsibility and evaluation. The functional requirements are given as graphs of functional responsibilities where modifiability, efficiency and traceability are considered as the quality attributes. The proposed method produced solution with increased quality and better metric values.

7

Authors: Farzan Rashidi

To improve wide angle, bottom-interacting situation with complex boundaries in ocean acoustic wave propagation, in this paper the finite element method is used to discretize the continuous depth system of equations. Coupled linear Galerkin finite element method in association with parabolic equation model is applied for depth discretization of underwater wave equation. This discretization leads to equivalent system of ordinary differential equations (ODEs) that are applied over complex shapes covered by variable nature of surface and bottom boundaries. These ODEs are estimated by Crank-Nicolson approach. Simulation results show the capability of the proposed method in comparison with the physically behavior of underwater sound propagation and some numerical results of standard software. The obtained results demonstrate good agreement and efficiency of the proposed method.

9

Authors: Younes Khandouch, El Houcein Aassif, Said Agounad, Gérard Maze

A number of researches in different areas has been conducted on using the Principal Component Analysis PCA combined with Artificial Neural Networks ANN. In this current research, we use these two techniques to estimate the radius ratio b/a (a: outer radius, b: inner radius) of an elastic tube based on its form function. The PCA technique is used to estimate the component loadings corresponding to form function. Then, the component loadings it’s used in the ANN technique to estimate the radius ratio of the tube. To get the optimal network, several configurations are implemented and tested. The optimal configuration selected is a network with 16 inputs, 2 hidden layers composed of 4 and 1 neurons respectively, and trained by the back-propagation algorithm. This configuration is able to estimate the radius ratio

10

Authors: Anju Malik, Vinod Kumar Jain

Cloud providers offer several storage services for their users in efficient manner. Cloud users are allowed to store their data in cloud. The data leakage, lack of proper security control policy, and weakness in the data security are the main worries of the companies. We intended to propose a secure cloud storage system RSA-ABC algorithm for encryption and decryption process. Initially the user will login with the aid of private key and public key. Then send a request to the data base administrator for upload a file in a cloud. In this phase we will give one time password and keys. Then the DBA will verify that data and accept that file and encrypt that file and stored in to the cloud server. If the user want that file means they send a request to DBA continue the same process after they will decrypt the file and send to the user.

10

Authors: Vinu Sundararaj

Electrocardiographic (ECG) signal is significant to diagnose cardiac arrhythmia among various biological signals. The accurate analysis of noisy Electrocardiographic (ECG) signal is very motivating challenge. Prior to automated analysis, the noises present in ECG signal need to be eliminated for accurate diagnosis. Many researchers have been reported different methods for denoising the ECG signal in recent years. In this paper, an optimized threshold mechanism is proposed for wavelet based medical signal noise reduction. This scheme is based on a variable step size firefly algorithm (VSSFA) in dual tree complex wavelet scheme, in which the VSSFA is utilized for threshold optimization. This approach is evaluated on several normal and abnormal ECG signals of MIT/BIH arrhythmia database, by artificially adding white Gaussian noises with variation of 5dB and 10dB. Simulation result illustrate that the proposed system is well performance in various noise level, and obtains better visual quality compare with other methods.

10

Authors: Thiruppathy Kesavan Venkatasamy, Radhakrishnan Shanmugasundaram

In this paper, Cluster Based Dynamic Keying Technique for Authentication in Wireless Sensor Networks (WSN) is proposed. The CH enabled by Global Positioning System (GPS) determines a Combined Cost Value (CCV) for each sensor node based on nodes location, node degree and virtual battery power. Since the key to the encryption scheme changes dynamically when there is change in CCV, separate dynamic keys are generated in each cluster by each CH and using RC5 encryption mechanism the source CH forwarding the data along different clusters along the path to the sink are able to verify the authenticity. When the source CH needs to transmit the aggregated data packet to the sink, it splits the packet into shares by threshold secret sharing algorithm and forwards them to the sink using multi-path dispersal routing technique.

11

Authors: Kuruva Lakshmanna, Neelu Khare

In this paper, we propose a 3-step DNA sequence mining algorithm, called 3s-DNASM, incorporating prefix span, length and width constraints and group search optimization. The complete mining process is comprised into following vital steps: 1) applying prefix span algorithm, 2) length and width constraints, 3) Optimal mining via group search optimization (GSO). We first present the concept of prefix span, which detects the frequent DNA sequence. Based on this prefix tree, length and width constraints are applied to handle restrictions. Finally, we adopt the group search optimization (GSO) algorithm to completeness of the mining result. The experimentation is carried out using DNA sequence dataset, and the evaluation with DNA sequence dataset showed that the 3s-DNASM system is good for sequence mining. The simulation results illustrated that when min_support=4, the number of DNA sequence mined only 29 patterns by 3s-DNASM system, and in this case prefix span mined about 2168 patterns.

15

Authors: Vedururu Sireesha, Sandhya Rani Kasi Reddy

In the document, we are intending to present an innovative technique for the multimodal biometric authentication. Initially the input image is preprocessed then offered to feature extraction, where the modified local binary pattern is effectively utilized. Thereafter, the extracted features are furnished to the feature level and score level fusions. In feature level fusion, extracted features are offered to the GSO where the optimal features are shortlisted, and are furnished to the optimized neural network which effectively detects the iris and fingerprint image. In score level fusion, extracted features from the iris image are offered to the PSO and naive bayes classifier here one score value is achieved. After that, extracted features from the fingerprint image are applied to the AGFS and then one score value is attained. Finally, both the score values are combined. The evaluation tools utilized precision, FAR and FRR. The proposed method implemented in MATLAB platform.

6

Authors: Pawan Whig, Syed Naseem Ahmad

: The Spice Model for photo catalytic sensor (PCS) proposed by Whig and Ahmad overcome several drawbacks like complex designing, non-linearity and long computation time generally found in Flow Injection Analysis (FIA) technique by Yoon-Chang Kim et al for the determination of Chemical Oxygen Demand (COD). FIA technique involves the complete analysis including sampling and washing. Flow Injection Analysis is analytical method for the measurement of chemical oxygen demand using photochemical column. This method uses a bulky setup and takes 10-15 minutes to get the output result which is a tedious and time consuming job. If conventional method is continuously used for a long time than it is stable only for 15 days. The purpose of this paper is to propose a new Quasi Floating Gate Photo Catalytic Sensor (QFGPCS) approach which has low power consumption and compared the results with the conventional models. The proposed model operates under sub threshold conditions which are appreciated in large integrated system design. The results of simulation were found to be fairly in agreement with the theoretical predictions. The results are exhibited near linear variations of parameters of interest with appreciably reduced response time.

7

Authors: Chintan Bhatt, C. K. Bhensdadia

Data Mining techniques are broadly utilized crosswise over numerous orders to recognize hidden patents, rules or relationships among gigantic volumes of information. Induction Tree, for example, C4.5 is the most favored technique since it functions well under any dataset set being utilized. Exponential ascent in the utilization of the internet because of informal organizations began to get enormous volume of information crosswise over various areas in brief timeframe. These attributes by which the colossal measures of informal organization information are produced make them to order as Big Data. When adapting to huge information (Big Data), the greater part of the current discretization methodologies won't be very productive with respect to implementation. The most effective method to separate significant data from huge information has been a famous open issue. In this paper, we are proposing new algorithm of decision tree in big data. At last, we have shown some result using weka.

13

Authors: G. Balaji, R. Balamurugan, L. Lakshminarasimman

Generator Maintenance Scheduling (GMS) problem in a vertically integrated electric utility system is considered in this paper. The cost and reliability objectives are included in the problem formulation. The GMS is formulated as a non linear, heavily constrained multi-objective optimization problem and the non dominated solutions are obtained using multi objective hybrid integer coded differential evolution (MOHDE) algorithm. To attain the best compromise non dominated solution between cost and reliability, fuzzy based mechanism has been used in MOHDE. In this paper, fuzzy clustered multi objective hybrid differential evolution (FC-MOHDE) is proposed for solving thermal generator maintenance scheduling. To validate the performance of the proposed algorithm, two systems having 4 and 22 generating units are considered. The test results confirm the superiority of the algorithm in determining the best compromise solution for the multi objective GMS problem.