International Journal of Intelligent Engineering and Systems

1

Authors: Lahcen Mountassir, Touriya Bassidi, Salma Aziam, Hassan Nounah

This article intends to present a time-frequency representations adapted to the characterization of steels damaged by the environment. These techniques are applied for analyzing experimental acoustic signal backscattered by corroded steel immersed in water. These signals are processed using a spectrogram analysis, Pseudo Wigner-Ville Distribution PWV and continuous wavelet, whose interest is to provide a signal representation in time-frequency. The goal here is whether each pattern of representation is due to corrosion or default in the plate. The use of time-frequency representations for characterizing steels damaged by the environment proves interesting because of their ability to localize the energy of the signals in a two-dimensional plane. However, bearing in mind that these representations should check drastic properties about the studied signal, choose a representation adapted case by case. The results obtained highlight the advantage of using time-frequency representations for the study of the acoustic wave’s propagation in corroded steels.

22

Authors: Ramanjaneya Reddy Nalavala, Chenna Reddy Pakanati, Padmavathamma Mokkala

Multiple senders share bandwidth and send their data in high speed over the wired network. This may lead to congestion in the network. Congestion control plays a prominent role in network resource management. Effective mechanisms are needed to control the congestion in the network. In Transport layer, TCP Friendly Rate Control (TFRC) has been widely used for advanced multimedia streaming application over wired networks to give best smoothing sending rate. However, there exists packet loss due to high traffic in the network. To alleviate the barriers of congestion and reduce packet loss in the network, one needs to implement best queue management mechanisms to help the TFRC. In this paper, we have applied enhanced queue management scheme to TFRC. The simulation results have shown that this strategy reduces packet loss and increases throughput.

38

Authors: Balamurali Ananthanarayanan, Siva Balan, Anu Meera Balamurali, Karthika Balamurali

In the field of weather forecasting, especially in rainfall prediction many researchers employed different data mining techniques. There is numerous method of organizing agricultural engineering substance and it remains an open research issue particularly when taking to distinctive arrangements of clients - farmers, agricultural engineers, agri-organizations -both from proficiency point of view. Keeping these factors Indian farmers in mind, we have chosen to do research on efficient dissemination of rainfall forecasting to safeguard farmers from crop failure using optimized neural network (NN) model. Here, at first, we generate the feature matrix based on five feature indicator. Once the feature matrix is formed, the prediction is done based on the hybrid classifier. In hybrid classifier, particle swarm optimization algorithm is combined with Grey Wolf optimization for training the RBF NN. The performance of the algorithm is analyzed with the help of real datasets gathered from pechiparai and perunchani regions.

56

Authors: Patel Harshita, Thakur Ghanshyam Singh

Classification of imbalanced datasets is one of the widely explored challenges of the decade. The imbalance occurs in many real world datasets due to uneven distribution of data into classes, i.e. one class has more instances while others have a few that results in the biased performances of traditional classifiers towards the majority class with large number of instances and ignorance of other classes with less data. Many solutions have been proposed to deal with this issue in various crisp and fuzzy methods. This paper proposes a new hybrid fuzzy weighted nearest neighbor approach to find better overall classification performance for both minority and majority classes of imbalanced data. Benefits of neighbor weighted K nearest neighbor approach i.e. assignment of large weights to small classes and small weights to large classes are merged with fuzzy logic. Fuzzy classification helps in classifying objects more adequately as it determines that how much an object belongs to a class. Experimental results exhibit the improvements in classification of imbalanced data of different imbalance ratios in comparison with other methods.

75

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

An effective task scheduling is one of the vital aspects for effectually hitching the potential of cloud computing. The most important aspect of task scheduling focuses on balancing the load of tasks among virtual machines, which is independent in nature. Energy conservation is one of the major key issues in cloud environment which in turn reduces operation costs in cloud datacenter. Meanwhile, Energy-aware load balancing optimisation technique is a promising way to attain the goal. To ensure fast processing time and optimum utilization of the cloud resources, we propose an energy-aware Fruit fly optimisation algorithm (EFOA-LB) for balancing the load among virtual machines in the cloud system. The energy-aware EFOA-LB is a modern swarm intelligence algorithm inspired by the foraging behavior of fruit flies, aims to attain well-balanced load on virtual machines and reduces energy consumption accordingly. Based on results obtained from our simulations, the proposed algorithms minimizes makespan and reduces the energy consumption of the datacenter, while meeting the task performance. The experiment results indicate that the energy-aware EFOA-LB algorithm is more efficient than the existing load balancing algorithms.

95

Authors: Yarabham Gangadhar, Venkata Sesha Giridhar Akula, Chenna Reddy Pakanati

Watermarking is one of the efficient approaches for digital authentication. An adaptive feature point extraction model is proposed in this paper for robust watermarking. The host image is treated with number of geometric attacks for extracting the consistent feature points form the image. The proposed watermarking scheme follows the adaptive feature point extraction method to retrieve the feature points from the host image. The response value of the feature points is calculated for improving the selection of feature points. The watermark insertion procedure is employed by inserting the watermarking bits with in the place of feature points. The feature point portion is extracted from the image and replaces the portion with watermarking bits. The watermark extraction procedure is used to restore the original image from the watermarked image. The watermarking bits in the image are replaced by the feature point portion for restoring the original image. The simulation experiment is carried with the MATLAB simulator. The proposed algorithm is tested with geometric attacks such as scaling, rotation, noise pollution and JPEG compression. The proposed method proved its efficiency when compared to other remaining algorithms.

114

Authors: Ramanathan L, Angelina Geetha, Khalid M, Swarnalatha P

Early prediction of student performance helps to take action for better achievements of students. To achieve the better education standard, several attempts have been made to predict the performance of the student, but the prediction accuracy is not acceptable. To accomplish the enhanced prediction, neural network (NN) based method is proposed. In this paper, an approach to predict student academic performance in college education based on Lion-Wolf artificial neural network is proposed. Lion algorithm and Grey Wolf optimizer is integrated to develop a Lion-wolf training algorithm to find the optimal weight for every neuron in NN. The proposed prediction model is validated based on Mean Squared Error (MSE) & Root Mean Square Error (RMSE) with the existing NN based prediction model. The experimental results show that performance of proposed prediction model is improved compared to existing prediction model with MSE of 5.25 and RMSE of 2.3.

134

Authors: Shakkeera L, Latha Tamilselvan

With the rapid and tremendous growth of real-time resource discovery process ubiquitously, Internet of Things (IoT) and mobile cloud technologies have become an essential part of the future internet. The adaptive and energy-efficient mobile application execution is still in its infancy stage due to the diversity of IoT applications, manipulating workflow of an application and handling a large volume of streaming data in the heterogeneous resource-rich environment. To overcome these constraints, this paper introduces a GAming IoT workflow Fuzzifiers and Supporting MAximum Resource utilization and optimal Task execution on mobile cloud (GAF-SMART) paradigm offering an intelligent way of computing the mobile applications on the remote server. The GAF-SMART employs Game-theory and Fuzzy logic assisted adaptive task scheduler to attain optimal execution for maintaining a workflow of an application. Thus, the proposed mobile cloud IoT paradigm offers high Quality of Service (QoS), increases resource utilization and minimizes the mobile device energy.

154

Authors: Deepa Mani, Anand Mahendran

Cloud management organisation is an imperative part of cloud computing platform and serving as the resource manager for cloud platforms. The multifaceted nature of cloud-management base makes its high availability (HA), a standout amongst the most necessities. Different innovations have been produced to build a consistent quality and availability of cloud administration infrastructure. In any case, little work concentrated on quantitative examination of its accessibility. While this ability accomplishes a developed availability with small fault rates, corporate requests conveyed over the autonomous zones may encounter unique Quality of Service (QoS) because of various physical frameworks. The key target of this paper is to show how the Markov-based model can fulfil the client request. For this reason, a few scenarios of the failure rate of virtual machine's practices, for example, single system failure, multiple system failures, power outage are considered by applying the Markov model. The improved repair strategies of accessibility in various circumstances are also investigated. The Queuing models like Markovian and non-Markovian models are examined using phase type expansion and renewal theory keeping in mind the end goal to sufficiently speak to and to assess. The considered element unwavering quality perspectives if there should be an occurrence of the most part dispersed lifetimes and times to repair.

176

Authors: Ramani Selvanambi, Jaisankar Natarajan

The main aim of the proposed method is to remove cyclic repeated patterns in sequential pattern mining. Initially the input dataset is fed to the clustering process, in which fuzzy c means clustering algorithm is used to cluster the available data based on the similar sequential pattern. This approach is able to mine the patterns with the help of association rule mining, here two major tasks are present one is frequent item set generation and rule generation. In frequent item set generation, support and confidence value is evaluated for each pattern. Based on that, the rules are generated in rule generation. After the mining process, the threshold is fixed and based on that repeated cyclic patterns are removed and stored to the database. The performance of the suggested method is evaluated by means of execution time, memory and database difference ratio. The implementation is done with the help of JAVA platform.