International Journal of Intelligent Engineering and Systems

1

Authors: Baby Honnenahalli Thammannagowda, Sujatha Bangalore Ramchandra

In the present days, internet communication is adapting data hiding and cryptography that often needs billions of pseudo random numbers. The Pseudo Random Number Generator (PRNG) is used to fulfill privacy and security needs in most of the applications. In this paper, the PRNG using Lorenz Chaotic Circuit with Carry Select Adder (PRNG-LCC-CSLA) method used to generate the pseudo random numbers. The LCC implemented by using Verilog to generate the pseudo random numbers, and randomness of the LCC results also checked in National Institute of Standards Technology (NIST) testing scheme. Encryption process is implemented by employing binary value of both text and image. Decryption process is implemented by employing same LCC binary value of encrypted data. At last, the Matlab is used to observe the text and image data. The Field Programmable Gate Array (FPGA) results showed how Look Up Table (LUT), slice, flip-flop and frequency improved and Application Specific Integrated Circuit (ASIC) results showed area, power, delay, Area Power Product (APP) and Area Delay Product (ADP) improvements in the PRNG-LCC-CSLA method compared to the existing methods.

22

Authors: Parag Rastogi

Software testing is a major technique for designing the software without fault and use the resources efficiently for the development of any software. The software bugs can be detected by the process of executing an application or a program. Test Case Generation (TCG) is a method to identify the test data and satisfy the software testing criteria. An automatic TC technique automatically determines where the TC or test data generates by utilizing search based optimization method. In this paper, Gref-Wolf and Firefly algorithm (GWFF) method is used for optimization of TC and generation of path coverage within the minimal execution time. In general, the challenging task is constraint handling in which the fitness function is directly affected by updating the positions of the searching agents. In proposed method, there is no direct relation between the fitness function and search agents, because the agents update their positions with respect to the locations of alpha, beta and delta. Experimental results showed that the GWFF method helped to generate path convergence within minimal execution time. Comparing to the existing methods such as Bee Colony Algorithm (BCA), Particle Swarm Optimization (PSO) and Cuckoo Search (CS), the proposed GWFF method provided better performance in terms of fitness values.

42

Authors: Gulivindala Suresh, Chanamallu Srinivasa Rao

In this digital age, the extensive usage of digital devices and availability of open source image editing software leads to the easy manipulation of digital images. Copy-Move Forgery (CMF) is a guileless and widespread approach to hide or duplicate a certain portion of the image without leaving visual clues. Thus, it is difficult to detect the copy-move forgeries and there is a need for forensic experts to rely on an effective approach for CMF detection for forensic analysis. Hence, an efficient passive block based approach to detect and localize CMF is proposed. In the proposed method, texture features are extracted from Differential Excitation Component on the overlapping blocks of the image. Similarity measure is performed for block matching and mapping is done to identify the duplicated regions. Evaluation is performed qualitatively and quantitatively on CoMoFoD dataset; true detection rate of 0.99 and false detection rate of 0.08 has been achieved. Evaluation validates that proposed method out performs the other existing methods with regard to detection accuracy.

62

Authors: Nilar Thein, Teguh Bharata Adji, Kazuhiko Hamamoto, Hanung Adi Nugroho

CT imaging is widely used for a variety of diagnostic and therapeutic purposes. It is a best effective and efficient tool in kidney stones diagnostic and treatment strategies whenever contactless measurements of quality are necessary to examine different types of stones. Accurate segmentation is a key process in CT diagnostic imaging. But, there are some challenges to distinguish among stones, stents or nephrostomy tubes, calcification and bone fragments in stone diagnostic CT imaging. This study proposed and experimentally investigated a scheme to reduce false positive for stone detection and to extract meaningful structural information for diagnosis of kidney stones. Firstly, the interested object (stone) is segmented by Otsu’s thresholding method and morphological operation. Secondly, the parameters based on 3D morphological features are measured for stone detection. In false positive reduction, the unwanted disturbances (bone fragments, image noise, calcification and stents) are eliminated using novel volume-ratio based thresholding and SAV ratio based thresholding. Finally, it provided an output that identifies meaningful structural information (Surface-area-to-volume ratio, volume, location and density of kidney stone) for kidney stone diagnosis. Digitized transverse abdomen CT images from 35 patients with kidney stone cases were statistically analyzed and validated. The estimation of coordinate points in the stone region was measured independently by the expert radiologists to get the validated data for the analysis. The proposed algorithm was significantly reduced the false positives with 70% of overall accuracy. Moreover, it can present the meaningful information of the kidney stone. This may lead to more accurate stone management.

85

Authors: Riyanarto Sarno, Kelly Rossa Sungkono, Reynaldo Johanes, Dwi Sunaryono

An event log records the business processes of a company. Modeling event logs aim to help users in analyzing business processes. One of the problems in modeling event logs automatically is the addition of invisible tasks. Invisible tasks are dummy activities, other than activities of an event log, that are added to a process model to describe a correct process model. This research proposes a graph-based algorithm to mine the data from an event log. From the data, the graph-based algorithm establishes an additional-invisible-task process model by converting all of the processes in the event log into a link list and adding invisible tasks and operators for parallel relations, such as XOR Split or XOR Join. The experimental analysis explains that the fitness of the discovered process models by the graph-based algorithm was as high as that of compared algorithms, such as Alpha# models, Alpha$ models, CHMM-NCIT models, and CHMM-IT models. Furthermore, the graph-based algorithm is more efficient than existing algorithms. This was proven by the time complexity of the graph-based, which is O(n2) while both of Alpha# and Alpha$ algorithm have a time complexity of O(n4) and both of CHMM-IT and CHMM-NCIT algorithm have O(n3).

104

Authors: Nijaguna Gollara Siddappa, Thippeswamy Kampalappa

Classification over numerous real-world datasets has a peculiar drawback called unstructured class problem. A dataset is said to be unstructured when the majority of the class has more samples insignificantly than the minor class. Such drawbacks result in an ineffective performance of data classification techniques. Classification is a supervised learning method which acquires a training dataset to form its model for classifying unseen examples. However, in unstructured data classification, the class boundary learned by standard machine learning algorithms can be severely skewed toward the target class. As a result, the false-negative rate can be excessively high. The researches focus on the unstructured data classification using uncertain Nearest Neighbor (NN) decision rule and also found the major issues face by k-Nearest Neighbor (k-NN). In any case, given a dataset, prediction of accuracy is a monotonous task to improve the execution of kNN by tuning

125

Authors: Bandani Anil Kumar, Makam Venkata Subramanyam, Kodati Satya Prasad

Mobile Ad-hoc Network (MANET) is a collection of wireless mobile hosts that create a temporary network without the help of centralized administration or particular support services. Energy consumption is the most critical issue in the MANETs, since the superior part of the mobile hosts work on inhibited battery resources. Reduction in the energy consumption results in the increase in the lifetime and throughput of the network. The performance of existing strategies is less in terms of energy conservation issues. To overcome these limitations, this paper uses an energy consumption method incorporate with a proactive MANET routing protocol. The routing protocol depends on the energy level and the movement of the nodes. Estimation of energy cost is performed based on the calculation of the energy consumption level of the node, using K-Means clustering along with AODV (Ad-hoc On Demand Distance Vector) optimized with Ant Colony Optimization (ACO). The performance of the proposed “K-Means-AODV-ACO” model is compared with the existing Energy Efficient Hybrid Routing Protocol (EE-HRP) and Mobile Agents based Reliable and energy efficient routing protocol (MAREERP) MANETs. From the comparison result, it is clearly seen that the proposed model accomplished better execution in terms of number of Packet delivery ratio (PDR), end-to-end delay, delay, drop and energy consumption.

145

Authors: Satrughan Kumar, Jigyendra Sen Yadav, Kumar Manoj, Subramaniam Rajasekaran

Background subtraction is one of the most reliable approach to localize the moving object under static camera arrangement. As seen, the moving object detection is a preliminary task in many vision applications such as video analysis, object tracking and activity analysis. However, the quasi-stationary pixels, aperture effect, ghost trail and varying illumination are still an annoying factors in the extraction procedures of the actual moving object in video. To alleviate the above problems, a video segmentation method is proposed that utilizes background subtraction and the 3-class fuzzy c-means clustering algorithm for extracting the relevant moving pixels. The proposed algorithm modifies learning parameters of adaptive filters to adapt the changes in the background. Afterwards, it incorporates the Markovian framework in which the initial motion field provides a prior information to regularize the segmentation process. The method achieves better visual and quantitative performance than other well-known background subtraction methods reported in this paper.

170

Authors: Abdelghany Salah Abdelghany, Nagy Ramadan Darwish, Hesham Ahmed Hefni

Ontology is defined as an explicit specification of a shared conceptualization. Currently, ontology is used in semantic web, information retrieval, artificial intelligence, information systems, knowledge management, etc. The development of ontology involves a structural and logical complexity that is comparable to the development of software artifacts. Therefore, ontology building requires a methodology to ensure its reliability. In this context, there are several methodologies proposed for building ontologies. However, most of the existing methodologies failed to provide sufficient details for the activities and techniques employed in them with a defined ontology lifecycle. To build ontologies that are reliable, long lived and continually adapted, the ontology engineering (OE) should be supported by the software engineering (OE). But, SE was not initially meant to support the development of software artifacts such as ontologies. There is a significant gap between them in terms of popularity and maturity level. The aim of this paper is to bridge this gap by proposing an Agile Methodology for Ontology Development (AMOD). AMOD adopts the agile principles and practices in the ontology development. The final framework of AMOD fits the various ontology activities into the phases of the Scrum agile methodology. It has three phases: pre-game, development and post-game. AMOD was applied to develop ontology for software project time management. Additionally, a compliance analysis of different ontology methodologies with respect to the IEEE Standard was made. Results showed that AMOD resulted in 56% satisfaction for IEEE standard processes. This resembles 22% enhancement in the satisfaction against the other methodologies.

192

Authors: Youssef Mouzouna, Ahmed Mouhsen, Hanane Nasraoui, Azzeddine Sardi, Mohamed Taouzari, Jamal El Aoufi, Mohamed Rafi

This paper presents a new miniaturized microstrip planar tri-band antenna operating at 2.45 GHz and 5.8 GHz for RFID applications, and 3.5 GHz for WiMAX applications. The proposed antenna is designed by using the slot inserted in its patch, and Complementary Triangular Split Ring Resonator (CTSRR) and Defected Ground Structures (DGS) techniques inserted in its ground plan. The proposed techniques for this antenna is studied, optimised and designed on FR4-epoxy substrate with thickness h=1.6mm, dielectric constant 4.4, and loss tangent 0.025, and simulated by using CST microwave studio software. The simulated results offer a good performances in terms of matching and radiation pattern. The proposed antenna is fabricated on the FR-4 epoxy substrate and tested. A good agreement is found between measured and simulated results. The advantages of this antenna are planar, smaller in size, simple structure, three bands, low in cost and easy to be fabricated.

211

Authors: Dalia Sameh, Ghada Khoriba, Mohamed Haggag

Nowadays, Social Media has become an important communication tool. Behavior could be measured using machine learning or deep learning techniques. This study represents a proposed model that shows how classification techniques can be used to recognize the personality according to the individuals' tweets using voting technique. The individual behavior is classified according to Eysenck's Three Factor personality model. A comparison was conducted between various machine learning and deep learning approaches as an input to a voting algorithm; which gave us more accurate results in the classifying task. The test results were 84.208% accurate. The study's main target is to make a new hybrid model for Twitter behavior analysis to enhance the accuracy of every approach individually on the dataset. Introducing the deep learning algorithm was to overcome the complexity and the time consuming obstacles. So this study can be used to predict future customers' behavior in order to increase satisfaction.

236

Authors: Mandeep Kaur, Balwinder Singh Sohi

Wireless Sensor Networks (WSNs) provides a great support and solution in the field of wireless technologies used in the industrial application. WSN faces serious reliability and load management issue when the power management terminology is enabled to prevent the network from excessive collision. This might result into a very low Packet Delivery Ratio (PDR), reduced network lifetime and high latency. In real time environment, a sensor node may suffer failure due to dust, corrosive agents and due hard weather etc. In that case, if the load increases on sensor node, it requires a reliable sensor node which can share the load. Reliable communication and load balancing among sensor nodes saves undesired consumption in collisions, data resending and data checking. This paper focuses on developing a novel framework enhancing the features of WSN to solve the reliability issue and load balancing problem with broadcast mechanism never used before in this scenario. The proposed framework utilizes a unique combination of Genetic Algorithm-Particle Swarm Optimization and Artificial Neural Network for load balancing architecture. The proposed architecture also utilizes some features of Ad hoc network also but with new terminology so that the packet reaches to the sink in timely fashion and maximum battery and network life along with minimized delay is attained. The proposed framework is evaluated using delay, energy consumption and throughput. One of the recent research works in this area is presented by Liu, Anfeng, et al and the proposed framework is also compared with the results of Liu, Anfeng, et al. By comparing the proposed results with existing, energy consumption of proposed work is reduced with better network lifetime by using the concept of load balancing technique.

261

Authors: Hosahalli Mahalingappa Premalatha, Chimanahalli Venkateshavittalachar Srikrishna

In today’s scenario, frequent requirement changes in software development are a notable issue in the software field. Because of the frequent changes, fulfilling the user’s requirement is very difficult. As a solution to such issues, Agile Software Development (ASD) has efficiently replaced the traditional methods of software development in industries. Due to various aspects of ASD, it is extremely hard to follow, maintain and estimate the general item. Hence, in order to tackle the Effort Estimation Problem (EEP) in ASD, various types of EEP have been identified in existing methods. The Evolutionary Cost-Sensitive Deep Belief Network (ECS-DBN) model implemented in this paper for effort prediction in any agile technique. The ECS-DBN method has no impact on agility because it uses simple and small inputs. The proposed method used in planning stage of software development to support the project managers in further development of agile software. The project managers characterize the structure of the ECS-DBN, while the parameter estimation consequently gained from a dataset. This paper used different statistics like accuracy, prediction at m level to evaluate the accuracy of the model. The ECS-DBN method achieved nearly 99% accuracy compared to the existing methods.

278

Authors: Jagatheeshkumar Gopal, Selva Brunda

Text clustering is one of the most researchable topics, owing to the massive usage of textual documents. The aim of text clustering algorithms is to group similar textual documents together, which can improve the data organization and makes the process of data analysis simpler. Understanding the benefits, this article intends to present a text clustering algorithm that is based on Fuzzy C Means (FCM) and Whale Optimization Algorithm (WOA). The optimization algorithm (WOA) selects the cluster centre, which helps the FCM to arrive at better textual clusters. The performance of the proposed text clustering algorithm is tested in terms of precision, recall, f-measure, purity and entropy over three different benchmark datasets. The performance of the proposed algorithm is observed to be reasonable when compared to the existing algorithms with an average F-measure of 97.6%.

297

Authors: Tallapureddy Subba Reddy, Thimmasandra Venkataswamy Sreerama Reddy

Among various types of heat exchanger, Shell and Tube Heat Exchanger (STHX) is highly used in the industrial application. The thermal efficiency of the heat exchanger is considered as a major factor for many applications. A considerable amount of heat is present in the low-temperature geothermal heat source, which is not sufficient to generate the electricity due to its low temperature. Many researches have been made to increase the thermal efficiency of the exchanger by the using of Organic Rankine Cycles (ORC). In this research, the Kinetic Gas Molecule Optimization (KMGO) is used to find the optimized parameter settings for the power plant. The geothermal ORC system is developed in the simulation and other various factors are considered to optimize the performance. The TEMA NFN STHX with a double-segmented baffle is analyzed for single-phase flow model. The different aspects such as baffle cut, baffle leakage, etc., are given as input to the heat exchanger. The proposed technique is evaluated with two different fluids such as R245fa and R134a. The KMGO technique is based on the swarm behavior of gas molecules, which is used to provide the optimized performance. The simulated results are measured and compared it with the existing methods for validation of the performance. The outlet temperature is achieved as 64.52°C and the enhancement factor is achieved as 1.88 for the R245fa fluid in STHX. This is achieved due to the KMGO technique identifies the global minima effectively due to the kinetic gas molecules theory, which shows the high efficiency compared to the existing method.

1

Authors: Penubaka Kiran Kumar Reddy

This paper introduces a new age estimation method based on the fusion of local features extracted using histogram-based local texture descriptors. In this study, the age estimation performances of well-known powerful texture descriptor with improved modification Local Binary Patterns (mLBP), and Local Phase Quantization (LPQ), which have not been analysed in depth for age estimation, investigated. In addition, multi-scale and spatial texture analysis performed for all descriptors. In the spatial texture analysis, a new approach using Overlapped Blocks obtained by discrete blocks proposed to reduce the overlapped age groups between the neighbour age group. Then feature fusion performed to investigate the age estimation accuracies of different combinations of local texture descriptors. Multiple Linear Regression (MLR) used to estimate the specific age at the end of this proposed system. The results show that the age estimation accuracy of the proposed method is better when compared to previous methods on FG-NET, MORPH databases.

22

Authors: Mohammed Y. Kamil, Ali Mohammed Salih

Breast Cancer is one of the common and dangerous among women at the age of forty, so it is better for woman to have mammography testing as a significant step for the early detection of breast cancer and is diagnosis for treatment; There is an important need to an algorithm is used to determine the boundaries of the tumor in a finite accuracy. In this work, two algorithms were built depending on clustering approach as segmentation method. In the first algorithm has employed (K-mean) method, whilst in the second algorithm has employed fuzzy c-mean method (FCM). In both, the lazy snapping algorithm was used as an additional step to improve the segmentation performance of the detection of abnormal area. The proposed methods have been tested using mini-MIAS database, after assessment the results obtained. it indicates the accuracy of segmentation first algorithm, are 91.18% and accuracy of second algorithm is 94.12%. from results, it concludes that the proposed second algorithm is capable of estimate breast abnormal region boundary at high accuracy because it used fuzzy logic technique.

40

Authors: Kaliraj Shanmugaiah, Bharathi Ayyaswamy

Nowadays, Software development process has incorporated many techniques, scientific approaches and advanced tools to develop the highest quality featured software. The size of the software systems also become large and make the system more complex. So that, it is necessary to ensure the quality of software system before its delivery. Software reliability is one of the important measurement to predict the quality of software which will be used to reduce the time, cost and other resources required for the testing process. Here we have proposed one novel framework to predict the software reliability based on software architecture and path testing by incorporating in-degree and out-degree concept used in graph theory. This proposed framework will reduce the number of test path in the system which will be used to reduce the testing time and resources. The applicability of our proposed framework is validated by using two real time case studies and these results are compared with the prevailing standard reliability models to assess the prediction accuracy of our proposed framework. The result shows that our proposed framework is simple to apply in the real time system and prediction accuracy is also acceptable compared to the other models.

62

Authors: Thorepadi Kannappan Thivakaran, Sathya Veera Venkata Naga Chanukya Padira, Ayyanar Sanjeevi Kumar, Sureddy Sivar Reddy

Biometrics is the art of distinguishing or analysing the identity of a person in view of physiological or behavioural attributes. Biometric recognizable strategies have turned out to be more productive, natural and simple for human identification compared to conventional techniques. Biometric validation frameworks basically design recognition frameworks, the physiological qualities like finger impression, face, and hand geometry, DNA and iris acknowledgment. The primary aim of the method is to build up a biometric validation framework by utilizing the ear and fingerprint images to distinguish a genuine user. In pre-processing step, all images are enhanced by thinning and binarization method for extracting the features. Then, a feature extraction system incorporates Minutiae and Singular point procedure for fingerprint images. Ear features are extracted by utilizing Speed Up Robust Features (SURF) and Binary Robust Invariant Scalable Keypoints (BRISK) systems. The features are fused by concatenating the fingerprint and ear features to obtain accurate information. Finally, matching is completed by registration and similarity score process and after that by utilizing the threshold esteems, the users are distinguished as genuine or an imposter. The experimental results demonstrated that the proposed multi-model biometric accomplished 95.66% precision with the error rate of 0.0434.

84

Authors: Krishna Kumar Palavalli Radharamana, Chandra Mouli, Udaya Kumar

Internet of Things (IoT) provides information services based on daily usage that depend on the sensors of devices and network platform. An increase in the number of connected devices through internet, increases the demand for the number of low-latency services. In this research, Open Cloud Computing Interface (OCCI) technique is used in IoT architecture that helps to encompass application level interface. The OCCI-based architecture is proposed to manage and store the data, by introducing the resource analyzer, edge devices and monitoring manager that helps to transfer the data effectively. The monitoring manager, edge devices and broker schedule the data to minimize the traffic in the IoT. Representational State Transfer (REST) methods using Hyper Text Transfer Protocol (HTTP) for communication are presented in this technique. Simulation result showed the effectiveness of the proposed OCCI architecture. Arduino and Raspberry Pi 3 are the two major hardware used in this technique. The result of the OCCI-based architecture uploaded to the ThingSpeak server, which is the external server. Several parameters such as Temperature, Round Trip Time (RTT), latency, Clock difference and frequency are evaluated in this work. Round Trip Time reduced to 0.96 seconds by reducing the delay in the system.

104

Authors: Anju Shukla, Shishir Kumar, Harikesh Singh

Grid is collection of heterogeneous resources which are highly dynamic and unpredictable in nature. Various resource allocation techniques are used for effective use of resources. Tasks are submitted with constraints (cost, deadline, etc.) to improve the performance of distributed environment. Based on resource availability, several tasks select the same resource to achieve minimum execution cost. Therefore, an effective scheduling is required to execute tasks on the same resource to reduce average waiting time of tasks. In this paper, an extensive survey of resource allocation techniques is done based on several parameters. A resource allocation and task scheduling mechanism is proposed to optimize cost and average waiting time of heterogeneous tasks. The uniqueness of the proposed model is that it makes task execution schedule and sends calculated data to resource manager before actual task execution for reducing average waiting time of tasks. Further, we validate the effectiveness of the proposed model using GridSim by considering random values of task and resources. Experimental results show that the proposed model provides up to 37% reduction in average waiting time than existing approaches under various environment conditions.

123

Authors: Kalaivani Dhanasegaran, Rajkumar Soundrapandiyan

These days, road safety applications have become significant due to the assurance of Quality of Service (QoS) requirements such as scalability, latency, and security facilitates multi-hop wireless broadcast capabilities in VANETs. Factors such as limited bandwidth, minimum contention window size and mobility of vehicular nodes make multi-hop broadcast a challenging endeavor. This paper proposes a suitable data diffusion model with emergency message transfer and clustering of neighborhood nodes based on zones. Based on the location of emergency-message senders, zones such as Red Orange and Yellow (ROY) are created to partition the vehicle in terms of directional, bi-directional and multi-directional broadcast in diverse directions. The adjustments in urgency factor on the basis of the emergency seamlessly handle the traffic target areas. The analysis and simulation result shows that the proposed diffusion model improves the performance aspects such as delay, message overhead and reception rate compared to the existing models.

142

Authors: Nainavarapu Radha, Tummala Ranga Babu

Multi-focus image fusion objective is to add relevant information from multiple images of the same scene but with different focuses into a sharper image that is more suitable for visual sensor networks. Natural and artificially obtained multifocus color images are considered for fusion. The existing fusion methods like multi scale and multi-resolution transforms are proved to be good in Multi-focus Image Fusion. However, they suffer from computational complexity in kernels calculation. In this paper, Multi-focus color Image Fusion based on Walsh-Hadamard Transform and sum-modified-Laplacian focus measure is proposed. Walsh-Hadamard Transform is a non-sinusoidal, orthogonal transform with symmetry, seperability and orthogonal properties. These properties make it more apt for image fusion than other transforms. And the sum-modified-Laplacian focus measure helps to get sharper image. Proposed method performance is evaluated in terms of reference and non-reference measures. The experimental results indicate that proposed method not only produces sharp details in fused image but also reduces the computational complexity.

165

Authors: Padmaja Katta, Nagaratna Parameshwar Hegde

Twitter is a microblogging site where clients read and compose short messages on different points each day. Political investigation using social media is getting the consideration of numerous scientists to understand the general assessment and pattern, particularly at the time of elections. In this paper, a proficient approach with respect to fuzzy rules to dissect the general assessment and further anticipated poll results. The system contains mainly three modules in particular Data extraction, Pre-processing and characterization of sentiment. The initial two stages have undertakings, namely social media information extraction and data pre-processing in order to, expel all Uniform Resource Locator (URL) frame separated tweet. The last stage is a grouping stage where a strategy called Adaptive Neuro-Fuzzy Inference System (ANFIS) where the fuzzy based ontology is designed by actualizing Non-Linear Support Vector Machine (SVM) classifier analysis to improve the fuzzy principles. The result showed that the ANFIS- NonLinearSVM based Sentiment analysis of social network data is less complex and gives the high rate of accuracy compared to the existing methodologies.

189

Authors: Sowmya Gali, Venkatram Nidumolu

Security is a more important aspect in the IoT based communications due to the vast heterogeneity of devices used in the network. Considering the challenges in the provision of security in the IoT network, this paper proposes a new trust ensuring mechanism with multi contextual aspects including the interactions between the IoT nodes and their energy levels. Furthermore, a minimum hop count mechanism is also proposed to select a path with less processing delay. Combining all these multi facets, a composite routing metric is derived in this paper to define the trustworthiness of IoT node before choosing it as a next hop communicating node. An extensive simulations are carried out over the proposed approach by varying the network parameters and the performance is measured through the performance metrics namely, packet deliver ratio, malicious detection rate, network life time etc. The obtained Malicious detection rate and network life time of proposed approach outperforms the conventional approaches.

211

Authors: Ahmed Ghaloua, Jamal Zbitou, Larbi El Abdellaoui, Mohamed Latrach, Abdelali Tajmouati, Ahmed Errkik

In this paper, a highly miniaturized microstrip antenna array is proposed for the small wireless device. The antenna array is designed to cover the Industrial Scientific Medical band (ISM) at 2.45GHz. The antenna array is miniaturized using a new geometry of defected ground structure (DGS) etched in the ground plane of each radiating element of the antenna array. The array size is 140.74mm by 57.15mm, which means 78.63% miniaturization ratio is achieved at 2.45GHz as compared with a conventional microstrip antenna array with an ordinary ground plane of 367.26x102.46mm2 resonating at the same frequency (2.45GHz). The structure is implemented on an FR-4 substrate of a relative permittivity equal to 4.4, 1.6mm of thickness and 0.025 of loss tangent. This compact antenna array is designed and optimized by using the electromagnetic solver based on finite integration technique, and the results agree with those obtained by the high-frequency electromagnetic field simulation solution (HFSS). The simulated 10-dB return-loss bandwidth of the printed array is about 157.5MHz, whereas the value of gain is about 7.026dB, which is relatively satisfied. Finally, the final circuit of miniaturized antenna array is fabricated and measured.

232

Authors: Gusti Pangestu, Fitri Utaminingrum, Fitra A. Bachtiar

People with multi-disability order need more handling than normal people. Smart wheelchair also developed with conventional mechanism that has more improvement in any aspect such as automatic system until control system. We develop a way for controlling smart wheelchair with movement of eyelid. In other ways, our system have capability to recognize state of the eye whether the eyes in opened or closed state. Opened and closed state in the eye can made a benchmark to control a wheelchair. In our research, we develop a system to recognize whether eye in the opened or closed state by using Multiple Method consisting of Haar cascade to detect eye location, Vertical Image Projection (VIP) to extract the feature and Learning Vector Quantization (LVQ) Neural Networks to recognize state of the eye. Our method produce an accuracy until 90.32% to recognize eye state. This approach can be used as an alternate way to control movement such go forward or stop by using their eyelid movement in smart wheelchair for disabled people.

253

Authors: Ilyass Ouazzani Taybi, Mohcine Bouksim, Rachid Alaoui, Taoufiq Gadi

The similarity measuring between 3D objects can be reduced to a distance between their descriptors. Therefore, the principal challenge is the mapping of objects into reduced and compact representations referred to as descriptors. Generally speaking, descriptors based on global features cannot be effective to describe incomplete 3D objects. This paper presents a novel partial 3D object retrieval approach called adaptive slices clustering (ASC). To ensure that similar objects will be decomposed similarly, normalization is applied to the objects as a preprocessing step. Then, we characterize each object by a set of 2D slices corresponding to its three principal axes, transforming the shape-matching problem between objects into similarity measuring between their slices. Next, we choose among the extracted 2D slices the most representative to reduce searching time without diminishing the performance of our approach. The results show that our approach provides better performance than the descriptors that characterize the whole object.

277

Authors: Pimporn Moeynoi, Yuttana Kitjaidure

Automatic sleep stage classification using various signals is an important tool to assess sleep disorders and sleep quality. However, the numerous variables from multi sources lead to the classification problems. To improve the problem, this paper proposes a new hybrid dimensionality reduction by combining the feature selection based on the nature-inspired algorithms (NIAs) and multi-sets transformation technique by Canonical Correlation Analysis (CCA)/ Discriminant Canonical Correlation analysis (DCCA). The NIAs is first adopted to generate the updated population positions in each dataset, and then the CCA/DCCA is used to fuse the selected subsets. The proposed algorithm performance is demonstrated on sleep-wake detection and multi-class sleep stage classification. Furthermore, the proposed method called Dissimilarity Binary Grey Wolf Optimization – Discriminant Canonical Correlation Analysis (DisBGWO-DCCA), modified by using the differential evolution (DE) technique based on the similarity of Jaccard coefficient, provides the best classification accuracies with 97.75% of the sleep-wake detection and 95.85% of the multi-class sleep stage classification when comparing with other single dimensionality reduction approaches[1-2]. Moreover, the proposed algorithm achieves the computational cost better than the conventional NIAs as shown later in experiment. Our experiment is also operated on both healthy subjects and sleep disorder patients with efficient sleep stage classification.

300

Authors: Dharmendra Ganage, Yerram Ravinder

Direction of Arrival (DOA) estimation is the important stage in a smart antenna that estimates the direction of impinging signals on an antenna array, and the most high-resolution DOA estimation subspace algorithm is Multiple Signal Classification (MUSIC). MUSIC algorithm is best suited for estimating the non-coherent signals due to its accuracy and a high degree of resolution. However, its performance degrades in presence of multipath propagation that has coherent and closely spaced signals and at low Signal to Noise Ratio. Many researchers proposed modifications to the MUSIC algorithm to improve its performance even under multipath, coherent and closely spaced signal environment. This paper focuses on investigating a proposed preprocessing technique, referred to as CW-ICA MUSIC that improves the performance of the MUSIC algorithm under above scenario even at low SNR. Simulation results confirm that the proposed preprocessing technique based on Dual Tree Complex Wavelet Transform (DTCWT) and Independent Component Analysis (ICA) outperforms the reported modified MUSIC algorithms such as Wavelet-MUSIC, and SS-MUSIC.

1

Authors: Shilpi Singh, Raj Shree

Software regression testing is one of the most critical phases of software development life cycle, used by developers with the intent of detecting new faults to validate modified software prior to delivery to the customer. To validate updated features, new test cases are generated by the testers which increment the test suite size automatically. The resulting test suite may contain obsolete, redundant, and ambiguous test cases. Therefore, there is a strong requirement of an intelligent testing approach to reduce the test suite size by removing those unessential test cases economically. This paper proposed an interesting approach, which involves the combination of regression testing techniques: minimization, and prioritization both. The main focus is on multiple regression activities with multiple criteria rather than using only single activity to produce an optimal solution. In this paper clustering approach is also considered, which could simplify and enhance the minimization and prioritization task. To evaluate the effectiveness of the strategy, we performed an experimental investigation together with an eminent heuristic Harrold Gupta and Soffa (HGS), considering the testing measures of the minimized test suite size and fault coverage. The results show that, similarity-based greedy approach with multiple coverage criteria can be quite effective in terms of fault detection loss of reduced test suite without much affecting the percentage of suite size reduction.

21

Authors: Vedavathi Gauribidanur Rangappa, Sahani Venkata Appala Varaprasad Prasad, Alok Agarwal

The catastrophic heart diseases such as Myocardial Infarction (MI), Heart Failure (HF) and Ischemic Heart Disease (IHD) is a chain process leads to Coronary Artery Disease (CAD). The analysis of CAD from Electrocardiogram (ECG) signals by manual techniques are quite difficult. Therefore, there is need of techniques without human interaction for classifying the CAD should be improved. This work presented the recognition of five types of ECG beats by using a three-step system. In the first step, Pan-Tompkins algorithm (PTA) is used for detecting the peaks in ECG signals. The second step includes extraction of three interval features combined with ECG higher order statistics. In the third step for classifying ECG beats K-Nearest Neighbour (KNN) technique is employed. This approach analysed the heartbeats as normal or abnormal as accurately, and the experiments were conducted on MIT/BIH arrhythmia database for classifying the ECG signals. The results stated that the accuracy of the proposed approach is up to 98.40% for segregating the signals.

44

Authors: Ahmed Elhamraoui, Elhassane Abdelmounim, Jamal Zbitou, Hamid Bennis, Mohamed Latrach

A new dual-band monopole microstrip antenna with modified ground structure designed for Radio Frequency Identification (RFID) readers is proposed. The rectangular slots inserted in the ground plan permit to achieve dual frequencies and also provide a broadband operation at high frequency. The entire area of the proposed antenna is 33×35 mm2 and mounted on an FR4 substrate with dielectric permittivity constant 4.4, thickness of 1.6 mm and loss tangent of 0.025 and fed by a 50 Ohm microstrip line. This antenna has been analyzed and simulated to cover the RFID at 2.45/5.8 GHz. The simulation has been carried out by using two electromagnetic solvers ADS and CST Microwave Studio. By fabricating and measuring the performances of the prototype of the proposed optimal antenna, experimental results show that the dual operating bands excited with -10dB return-loss bandwidths of about 70 MHz centered at 2.45 GHz and of about 490 MHz centered at 5.8 GHz were obtained.

65

Authors: Fajar Astuti Hermawati, Handayani Tjandrasa, Nanik Suciati

The research proposed a method that combined non-deep learning detector that called Aggregated Channel Features (ACF) detector and Convolutional Neural Network (CNN) that named Faster R-CNN detector to extract a cross-sectional area of the fetal limb in an ultrasound image. This combination is appropriate to solve the problem of object detection where the object has no clear characteristic, it has shape variation, blurred, and no clear boundaries, which is difficult to solve using the common thresholding or the edge detection method. This method also deals with the ultrasound image analysis which the training set is small. The pre-trained CNN can establish the classification model from the small annotated training data. ACF detector provides the region proposals of the non-cross-sectional area as an input of pre-trained CNN. The proposed method could improve the average precision of detection result when it was compared with Faster R-CNN and ACF detector alone. Also, the combination method could reduce the elapsed time of the Faster R-CNN training phase significantly.

85

Authors: Saravanan Ramalingam, Pothula Sujatha

Stock Price Prediction (SPP) is a potential fiscally rewarding arena as it involves several dependent factors and complex structure. As the stock price fluctuates during every business day, it is nontrivial to forecast the stock prices by the buyers and sellers. Numerous methods have been developed for SPP based on Machine Learning (ML) techniques. In this work, an Ant Colony Optimization (ACO) based SPP model is introduced to forecast the stock prices accurately. The efficiency of the proposed ACO-SPP model is analyzed by comparing the performance with a set of traditional ML based classifiers such as Naive Bayes (NB), Support Vector Machine (SVM), Multilayer Perceptron (MLP), Radial Basis Function (RBF), J48, RF, Classification and Regression Tree (CART) and Olex-GA. To emphasize the benefits of ACO-SPP model, it is applied to a benchmarked dataset named Dow Jones Index (DJI) dataset and three datasets from Yahoo finance on daily, weekly and monthly basis. The experimental results infer that the ACO-SPP accurately predicts stock prices than the compared methods in terms of accuracy, F-score, AUC, Discriminant Power and Youden's Index.

108

Authors: Yadaiah Balagoni, Ramisetty Rajeswara Rao

The Apache Hadoop with cloud had become an emerging and popular service. Irrespective of its huge dominance in large scale data processing, it has challenges yet to be addressed. The primary challenges in yarn scheduler are the abilities to automate and control the resource allocation to different workloads in order to meet the deadline-based Service Level Agreement (SLA) in the cloud environment with optimal energy consumption. Our study with the Hadoop YARN addresses this problem in a controlled homogeneous environment. In cloud data-centers, heterogeneity had become a normal phenomenon. Hence, this paper proposes the problem of energy-aware heterogeneous Hadoop Yarn cloud with deadline based SLA. We proposed a SLA-Aware Green Scheduling (SAGS), a Dynamic Voltage/Frequency Scaling (DVFS) based approach along with SLA-Aware scheduling algorithm in the heterogeneous environment. We evaluated SAGS by using benchmark datasets and, compared its performance with previously proposed solutions. Our observation with experimental results shows that, the proposed approach outperforms existing approaches.

128

Authors: Hemalatha Kivudujogappa Lingappa, Hosahally Narayangowda Suresh, Sunil Kumar Manvi

In image segmentation field, the Fuzzy C-Means (FCM) algorithm is a well-known algorithm for its simplicity and membership function that can control the overlapped clusters effectively with a predefined number of clusters. Despite the fact, the standard FCM algorithm is noise sensitive. To solve the issue, we proposed a new method of clustering named Kernel Fuzzy C-means (KFCM) clustering. KFCM performed well in terms of clustering however, for pattern recognition KFCM has issues. The first one is grouping the similar objects in a single partition due to non-awareness of patterns and the second one is misclassification of data due to the standard structure of the membership subspace plane. Non-awareness of patterns of KFCM is solved by an Extreme Learning Machine (ELM) algorithm and Artificial Bee colony (ABC) algorithm utilized for optimizing the structure of the membership subspace plane. Experimental results showed that the proposed KFCM algorithm performed better segmentation for pattern recognition. At last effectiveness of the proposed algorithm has been evaluated based on comparing the K Means, FCM, spatial FCM, and KFCM algorithms in terms of centroids, segmentation accuracy, and pixel error. The proposed methodology improved the segmentation accuracy up to 0.8-5.5% compared to the existing methods.

148

Authors: Nam-Soo Kim

In this paper, we derive the effect of channel errors on the performance of user relaying cooperative NOMA system in closed-form. The max-min selection strategy is adapted to the selection of a relay among near users (NUs) and selection combining (SC) is applied to a far user (FU) for the performance improvements. As the previous studies found, the max-min relay selection without channel errors is effective to improve the outage performance. However, we demonstrate that the outage performance is highly sensitive to the channel error. It is also found that the outage probability of the relay user is more susceptible to the channel errors. As the channel correlation coefficient approaches to 0, the performance gains are banishing and the outage probability with the max-min relay selection becomes identical to that with the random relay selection. NOMA system has an optimal power allocation coefficient to minimize the outage performance, but as the correlation coefficient decrease, the effect of power allocation to the performance becomes lessen. The provided numerical results confirm that the derived expressions match well with Monte Carlo simulations.

167

Authors: Sivakumar Kaliappan

The Privacy Preserving Data Mining (PPDM) is known to be the most critical perspective among analysts. As privacy preserving data mining grants, sharing and exchanging of privacy susceptible data for analysis, it has exploited increasingly popular. Since one of the critical aspects of data mining is safeguarding privacy. The diverse technique is embraced for preserving privacy while maintaining the real characteristic of data under consideration. In the proposed work, the high-dimensional data are isolated into various parts by utilizing the k-mean clustering technique and each partition is considered as a cluster. By then the mean estimate of each cluster is processed, after that, the contrast between each cluster member and the mean of the cluster esteem is processed. In the succeeding stage, the clustered information is enhanced by utilizing the Ant Colony Optimization (ACO) algorithm. After that, the optimized clustered particles are perturbed by utilizing the Random Rotation Perturbation (RRP) algorithm which makes the values hard to be recognized. These perturbed values are then stored in the public cloud and the key parameters for randomizing and the clustering is stored in the private cloud. Our approach would contribute in the diminishment of a lot of storage in a private cloud, in case we essentially store the entire sensitive information on private clouds. The experimental results demonstrate that the RRP algorithm has better privacy preserving contrasted with the other existing technique.

190

Authors: Sumanth Venugopal, Guruprasad Nagraj

In day to day life, the process of projecting the correct information to the authorized person is more difficult, which makes complexity to the decision making process. Web Page Recommendation Systems (WPRS) used in various fields to identify the customer needs and to help the users to take appropriate decisions over the service or product according to his/her preference. The group of users with similar preference will be identified by using Possibilistic Fuzzy C-Means (PFCM) algorithm with an S3I Similarity Measure (SM). The proposed method will determine the gain and loss of the web users based on the web directories which can be modified by using Relevance Feedback Bayesian Network (RFBN) technique. The experimental results are conducted on the MNSBC dataset and the outcomes are compared with the existing methods like Singular Value Decomposition (SVD) methods. The method predicts the accuracy up to 85% when compared with the existing methods and the outcome results proved the effectiveness of the PFCM - RFBN method.

209

Authors: Mariyam Ouaissa, Abdallah Rhattoy

Machine to Machine (M2M) communication is a technology for machines communicating with each other without human intervention. Long Term Evolution (LTE) and Long-Term Evolution-Advanced (LTE-A) cellular networks technologies provide the resources required for M2M communication. However, LTE is required to meet M2M communication requirements, such as power management and support for a large number of devices while maintaining the Quality of Service (QoS) requirements for these devices. Resource allocation or scheduling is one of the key challenges facing M2M communications over Long-Term Evolution networks. M2M traffic has unique characteristics; it usually consists of a large number of small data packets, with specific delays, generated by a potentially large number of devices competing on scarce radio resources. In this article, we will present the scheduling techniques on the Packet Downlink Control Channel (PDCCH) in a M2M scenario based on the priority and differentiation between heterogeneous traffic (real time and non-real time) to achieve high spectral efficiency in cellular systems and provide quality of service guarantees for system users. The results obtained show that the FLS downlink scheduler is the optimal solution based on our proposed method which offers more priority to the real time service (video and voip) than the non real time service (best effort).

234

Authors: Bhagirathi Bai Vittal Rao, Hanume Gowda Chandrakanth

Today, Internet is the best place to communicate and share information among the people throughout the world and gives an endless support of knowledge and entertainment. The main objective of Internet technology is to increase efficiency and decrease human effort. With the introduction of Internet of Things (IoT) in the last decade, we have been pushing for ubiquitous computing in all spheres of life. Physically challenged people are also using the Internet with the help of Speech commands (SC). The main objective of this paper is to minimize the effort and increase efficiency of the Voice recognition and IoT based secured automation, which is named as Raspberry pi and IoT based Speech automation (RASP-IoT-SA) technique. This RASP-IoT-SA system has only worked for authenticated users which improves the security in home and industrial automations. This RASP-IoT-SA system consists of four different Process such as secured speaker prediction, voice data transmission, web update and voice data Receiver processing. In this research work, Mel frequency Cepstral Coefficient (MFCC) technique is used for Feature Extraction (FE). Artificial neural network (ANN) technique is used for two different process such as, speaker prediction and speech data recognition (SR). The experimental result shows that the MFCC and ANN based technique provides better results in terms of accuracy, precision, false measure and recall. The RASP-IoT-SA technique delivers 99% of speech recognition accuracy, which was higher than the existing Speech automation technique.

252

Authors: Sayed Ahmed, Romani Farid Ibrahim, Hesham Ahmed Hefny

In emergencies cases, the main problem is how to arrive to an incident location in a minimum time. In such cases, using the classical distance based routing techniques does not always provide the optimal path in the time needed. In this paper, we present an enhanced routing technique which is based on Dijkstra's algorithm and Analytical Hierarchical Processing (AHP). Eight impendence factors that cause latency in emergency situations are analyzed using the AHP technique to give an overall weight for every road in Greater Cairo Region (GCR) road network. Then, the classical Dijkstra’s algorithm applied on the weighted road network to give the best path between any two nodes in the network. The eight impedance factors considered are: Road Length, Travel Time, Road Type, Road Width, Velocity Limits, Traffic Volume, Mass Density, and Junction Delay. The suggested routing approach is implemented as a mobile application, and a comparison with the classical Dijkstra’s algorithm is performed. The results show that the enhanced routing technique is more reliable than the classical one and more suitable for emergencies situations.

271

Authors: Tipirneni Venugopal, Vusthikayala Siva Kumar Reddy

In this research study, a multi-purpose image water-marking approach was developed for delivering self-recovery, ownership verification and tamper localization of the cover image. For robust watermarking, color and grayscale images were used to deliver wide applicability and generalization to proposed work. Initially, the cover image was converted into wavelet domain by using Integer Wavelet Transform (IWT) and then the logo image was encrypted by using chaotic logistic mapping. This procedure makes the digital image free from false positive error and also helps to provide decent capacity. After encryption and transformation, Least Significant Bit (LSB) was used to provide self-recovery feature as well as to locate the tempered region in the digital image. Then, asymmetric encryption algorithm: Rivest–Shamir–Adleman (RSA) was utilized to improve the hidden data security with high computational efficiency, and good embedding capacity. Finally, the experimental outcome shows that the proposed approach delivers a high secure network with low computational complexity by means of entropy value and Peak Signal-to-Noise Ratio (PSNR). The proposed approach improved PSNR value upto 19 dB and entropy value upto 0.07-0.7 compared to the existing methods: Discrete Cosine Transform (DCT) + Arnold transform and differential evolution algorithm.

292

Authors: Sumalatha Sunkara, Gollamandala Udaykiran Bhargava

A good orthogonality is provided by the wavelet -based orthogonality frequency division multiplexing (W-OFDM) and well reduction of bit error rate is gained with W-OFDM over conventional OFDM systems. Most importantly, it provides enhanced spectral efficiency since it does not require cyclic prefix (CP) which leads to the achievement of ultra-high data rates. Hence, this can be used in 5th generation (5G) wireless communication networks. It is proposed to use the discrete wavelet transform (DWT) based OFDM in place of conventional discrete fourier transform (DFT) based OFDM in the applications of long term evaluation (LTE). However, discrete wavelet transforms (DWT) based OFDM system suffers from the down sampling, which mitigates the actual size of input bit streams and there by degradation in the system efficiency. To resolve this issue, and for enhancing further mitigation in bit error rate (BER) performance, a novel W-OFDM is proposed. Our proposed W-OFDM system utilizes translation invariant wavelet transform (TIWT) in place of DWT for enhanced spectral efficiency. In addition, fading channel environment is considered to disclose the effectiveness of proposed TIWT-OFDM system. Extensive experimental results show that the BER performance of proposed TIWT-OFDM is superior to the existing OFDM systems like DFT-OFDM and DWT-OFDM systems.

1

Authors: Chundru Raja Ramesh, Komaragiri Raghava Rao, Gunamani Jena

Clustering is a procedure of finding similar data items (patterns, documents etc.) and then group the similar data together. Items belongs to different clusters are dissimilar data items, generally cluster values are considered as 1 or 0. The clustering process is not appropriate for all the cases sometimes these values are less than one. In practical situations clusters are not crisp, then it is represented as fuzzy. In order to enhance the clustering rate, two appropriate clustering approaches: K-means clustering and Fuzzy C Means (FCM) are considered. These approaches are modified by minimizing the objective function known as squared error function. The experimental research was performed on the publicly available database (i.e. yeast dataset) to validate its clustering performance in terms of accuracy, specificity, sensitivity and execution time. Experimental outcome shows that the proposed technique improves the accuracy in clustering rate up to 1.5-35% compared to the existing methodologies FCM and k-means approaches.

21

Authors: Arjin Numsomran, Thanit Trisuwannawat, Vittaya Tipsuwanporn

This paper presents an adaptive fractional order PIλDμ control for multi-configuration tank process (MCTP) toolbox which aims at demonstrating the problem of reference tracking and cross coupling rejection in multi-input-multi-output system. Moreover, we investigate the cases where the system is in the mode of minimum phase and non-minimum phase configuration. Besides providing theoretical control system analysis and design, we develop the multi-configuration tank process software toolbox for providing the non-linear functions of dynamic models of multi-configuration tank process which is the advantage tool for investigating the performances of the controllers. The software toolbox is developed from discrete state P-file S-function which is operated within the MATLAB environment for providing many non-linear functions of dynamic models of multi-configuration tank process such as multi-input multi-output quadruple tank full-interacting process, multi-input multi-output quadruple tank process, multi-input single-output triple tank interacting process, multi-input multi-output coupled tank interacting process. All actual process attributes are encapsulated in S-function as the input parameters, thus the multi-tank function block can be simply adjusted by specifying the physical properties of the tank system. The study explains about the mathematical model of multi-configuration tank process, nonlinear dynamic characteristic, minimum phase and non-minimum phase configuration, software toolbox features and also describes the design of the adaptive fractional order PIλDμ controller including the performance validation. The results have been illustrated that the proposed controller design scheme can provide the sufficient effectiveness in the performance, stability and robustness. Furthermore, these tests reinforce the usefulness of MCTP toolbox as a complete simulation tool for users to perform an engineering research of multi-configuration tank control system analysis and design, moreover, it contains very useful for validating the control algorithm of multi-input-multi-output system.

48

Authors: Yarragudi Madhu Sudhana Reddy, Ramaswami Sachidanandan Ernest Ravindran, Kakarla Hari Kishore

Retinal fundus images play an important role in the diagnosis of Diabetic Retinopathy. The detailed information of retinal images like Retinal Vessels, Exudates and microaneurysms may be in low contrast and retinal image enhancement helps in the accurate diagnosis of retinal images related diseases. This paper proposes a novel contrast enhancement method to make the low-contrast retinal images more qualitative such that the detailed information will be clearer. For this purpose, the proposed mechanism considers the spatial mutual relationships between the gray-levels of image and makes the gray-levels in the output image not only linear to the gray-levels of input image but also related to the neighboring gray-levels. The proposed approach enhances the image both in global and local fashion. Spatial mutual entropy based contrast enhancement is accomplished for global contrast enhancement and Greedy contrast enhancement is accomplished for local contrast enhancement. Extensive simulations carried out over various low-contrast retinal images of High Resolution Fundus (HRF) image database shows the outstanding performance of proposed approach. And, different from some other fundus image enhancement methods, the proposed method can directly enhance color images.

74

Authors: Halebeedu Subbaraya Suresha, Srirangapatna Sampathkumaran Parthasarathy

Detection of Alzheimer disease using Magnetic Resonance Imaging (MRI) is the most challenging aspect in the field of medical image processing and analysis. In this paper, the proposed methodology has three major steps: image acquisition, image pre-processing and segmentation. Initially, the brain images were acquired from the dataset: Open Access Series of Imaging Studies (OASIS). After image acquisition, image pre-processing was carried out using median filter, it utilized for cutting down the noise and to improve the quality of acquired brain images. Then, segmentation was carried-out using Fast-Independent Component Analysis (Fast-ICA) along with Otsu multi-level thresholding. It was a flexible high level machine learning technique to localize the object in complex template. In experimental analysis, the proposed approach distinguishes the brain MRI tissues: White Matter (WM), Cerebro-Spinal Fluid (CSF), and Grey Matter (GM) by means of Tanimoto index, similarity index, precision, and recall. The proposed methodology improved the Alzheimer tissue detection up to 15-30% compared to the existing methods: Band Expansion Process (BEP), ICA and BEP-ICA in terms of precision and recall.

96

Authors: Mohamed Ribate, Jamal Zbitou, Rachid Mandy, Ahmed Erkik, Mohamed Latrach

This paper proposes a Broadband Power Amplifier (BPA) for L and S bands applications based on ATF13786 transistor, using GaAs process. To improve the broadband performance, an approximate transformation of previously designed lumped elements into transmission lines, and a multi-section quarter wave impedance transformer are used. With neatly design of the broadband networks and bias circuit, a maximum gain of 14.89 dB is achieved across 1.1 GHz  3 GHz. At 2 GHz, the simulated large signal results demonstrate that the designed BPA achieves a saturated output power of 17 dBm, with 1-dB compression point at 4 dBm input power level, and a PAE of 20 %. For the whole bandwidth, the input return loss and output return loss are below than -10 dB. The maximum value achieved by the reverse transmission is – 20 dB over the operation frequency ranges. Considering the broad frequency coverage, the performance of the proposed design compares favorably with the state-of-art.

114

Authors: Chiranjeevi Pandi, Teja Santosh Dandibhotla, Vishnu Vardhan Bulusu

Opinion Target Extraction (OTE) identifies opinionated aspects in review sentences with the goal of achieving Aspect Based Opinion Mining (ABOM). The available review corpuses and datasets for this task contain a large number of opinionated sentences from wider class of products from different e-Commerce applications. This raised the need for constructing the dataset from these corpora for OTE in a new way called distant supervision. This is by selecting the review sentences based on the learned opinionated relations among the terms in the sentence. The obtained review sentences are then labelled and thus dataset is constructed. The attention model based review sentence selection that uses distant supervision has ignored the review sentences that do not contain opinionated aspect terms and treated them as noise. This has reduced the precision in terms of number of extracted aspects. In order to improve the precision of extracted aspect terms, the attention based review sentence selection is replaced with the learned corpus linguistic rules based review sentence selection. This takes into account the non opinionated aspects available sentences and these aspects are treated as neutral opinion targets. Finally, the extracted aspect terms are analyzed for opinion orientations. It is shown that the approach of corpus linguistics rules based review sentence selection outperforms the attention model based review sentence selection with distant supervision.

140

Authors: Aslam Shaik, Ankireddy Srinivasulu Reddy

Power quality disturbances (PQD) are becoming more and more complicated due to the vast growth of power quality oriented applications particularly the power electronic equipment’s. Since the power with less power quality consequences to a huge loss in the electronic devices, there is a need to detect the power quality disturbances more effectively. This paper proposes a new method for PQD detection and classification based on the entropy characteristics of PQD signals and a novel support vector machine algorithm (SVM). The proposed approach develops a flexible entropy based feature selection (FEFS) mechanism to extract the unique characteristics of all types of PQDs such that the detection system can detect more effectively with in less time. Further to increase the accuracy of individual PQDs, this paper proposed Multi Class SVM (MC-SVM). Various experiments are conducted over the proposed work and at every experiment; the performance is measured through Classification accuracy, False Alarm Rate and computation time. Further a comparative analysis reveals the outstanding performance of proposed approach when compared with conventional approaches.

162

Authors: Mona Abdelbaset Sadek Ali, Mohamed Abd-Elfattah

The aim of this research is to develop a hybrid Model (SVM-LOA) for epileptic seizure detection with support vector machine (SVM) and lion optimization algorithm (LOA) to locate the optimum parameters of support vector machines (SVMs) for classification of Electroencephalogram (EEG) signals . The proposed approach attempts to find the best integration of all available features that offers typical epilepsy detection and gives a better classification rate. Furthermore, the discrete wavelet transform (DWT) has been implemented to divide EEG signals into five combinations. Nonlinear parameters have been calculated and used to be the features to intern the SVM classifier that affects the classification accuracy. We used lion optimization algorithm based approach to optimize the SVM parameters. The overall experimental results shown the LOA-SVM classifiers have 96.78% for accuracy, but SVM that have obtained 80.05% for accuracy. Therefore; the SVM-LOA is an efficient model for neuroscientists to detect epileptic seizure in EEG.

183

Authors: Madugundu Neelakantappa, Amjan Shaik

Mobile Ad hoc Network (MANET) is a ubiquitous, infrastructure less and self-containing wireless network. In this networks nodes that are in mobility forms a temporary network without any infrastructure. In MANETs every node acts as host as well as router and therefore communication can be established on need basis without use of any fixed network equipment. But in presence of malicious nodes, this network leads to major security concerns as these nodes may disrupt the process of routing. In this scenario, detecting or preventing malevolent nodes, launching collaborative gray-hole or black-hole attacks is a major challenge. In this paper, a new secured routing protocol referred to as “Malicious Node Detection by Collaborative Bait (MNDCB)”. This protocol can be regarded as hybrid protocol as it poses the advantage of both reactive and proactive defending architectures. This MNDCB protocol is defined with a reverse tracking method to detect the malicious nodes and there by defending their collaborative attacks. Simultaneous results proves that in the presence of malevolent node attacks, the MNDCB gives better performance over their best counter parts base DSR, 2-ACK and best effort fault tolerant (BFTR) routing protocols in-terms of performance metrics: packet-delivery-ratio, throughput and routing overhead.

206

Authors: Bhavika Gambhava, Chandu Bhensdadia

Deployment of wireless links (terrestrial and satellite) along with wired links has made extension of the Internet even in remote places feasible. TCP/IP protocol suite is an integral part of the Internet. Congestion control of TCP plays a vital role in the performance of the Internet. TCP’s unconditional flow control in case of a packet loss has always been a concern for researchers. Further, halving congestion window in such conditions without taking in to account the current network state is also considered inappropriate. The problem is compounded in wireless networks where packet losses occur often due to channel errors rather than the shortfall in the available bandwidth. In this situation, TCP’s conservative behaviour underutilises the bandwidth. We therefore, propose a scheme to address the issue of underutilization of network resources. The proposed approach, Discrete TCP (DTCP), differentiates slow start and congestion avoidance phases while tuning data flow over a transport connection. DTCP evaluates ssthresh and cwnd before setting up parameters, based on the existing network condition to enhance the performance. The proposed scheme is compared and analyzed with various existing schemes with the help of extensive simulations using ns2. Results of simulation based experiments indicate significant performance improvement of DTCP on erroneous links and in heterogeneous networks and confirm its suitability.

226

Authors: Yassir Rochd, Imad Hafidi

With the blasting growth in data, uptake data mining techniques to mine association rules, and then find useful information hidden in large data has become ever more important. Several existing data mining techniques often through mining frequent itemsets draw association rules and get to relevant knowledge, but with the rapid arrival of the era of big data, traditional data mining algorithms have been impossible to meet large data's analysis needs. Lately, the PrePost algorithm has been suggested, a new algorithm for mining frequent itemsets based on the idea of N-lists. PrePost in most cases outperforms other present state-of-the-art algorithms. In mind of this, we present the HPrePostPlus algorithm. A better version of PrePost based on Hadoop, that utilization a HashMap to traverse effectively the PPC tree, and improve the process of creating the N-lists related with 1-itemsets. We combine also the characteristic of Hadoop with a view to process large data. Experience has demonstrated that HPrePostPlus algorithm is greater than the state-of-the-art methods in terms of performance and scalability.

246

Authors: Saptarshi Gupta, Rattan Lal Sharma

In this paper, a miniature improved gain Circular Patch Antenna (CPA) is reported. An array of Circular Split Ring Resonators (CSRRs) is loaded into a circular patch to transfigure the surface current distributions leading to miniaturization and significant gain enhancement. To obtain a bi-conical beam, the antenna is operated at Higher Order Mode (HOM). The HOM is excited by suitably selecting the feeding location. The unit cell of the CSRR is analyzed and the permittivity of the CSRRs loaded antenna structure is extracted using Scattering Parameter Inversion (SPI) method. The proposed antenna operates in 5.2 GHz band and built on a thin (0.254 mm) RT/Duroid 5880 substrate thus, flexible in nature. The bending analysis is performed showing a natural frequency shift. To validate the proposed design, the prototype is fabricated showing good agreement amongst simulated and measured results is found.

265

Authors: Geetha Megharaj, Mohan Kabadi

Load Balancing is an important aspect of cloud service centers for optimizing the resource utilization. Consumption of excess power in cloud centers can result in monetary wastage. It is critical that, the resources in the cloud center are utilized optimally; so that, both monetary savings and client satisfaction can be achieved. One of the most popular techniques to achieve load balancing is the Virtual Machine (VM) migration technique; wherein, some of the VMs from overloaded Physical Machines (PMs) are migrated to lightly loaded PMs; however, this technique requires excessive time and monetary cost. Recently, a load balancing technique which migrates VM tasks instead of the actual VM was proposed in the literature. This technique was able to overcome some of the limitations of VM migration technique. Here, the overloaded VM does not accept any new task; however, the new tasks are migrated to lightly loaded VMs. Even though this technique migrates extra tasks to achieve VM load balancing, the already overloaded VMs are not relieved from their existing task burden. If some of the existing and suitable tasks in overloaded VMs are migrated, it could improve efficiency of load balancing. In this work, a new run time VM task migration technique is proposed, which migrates tasks from overloaded VMs. The suitable tasks for migration are selected through a discriminant function, which identifies heavy resource consuming and limited execution progressed tasks for migration. Since, it has been shown in the literature that, optimal task-resource mapping is NP-hard, Particle Swarm Optimization (PSO) based solution search technique is proposed. This proposed technique substantially reduces computing load, and achieves good power/energy conservation in the overloaded VMs, when compared with the contemporary VM task migration and VM migration techniques.

290

Authors: Otmane Zebraoui, Mostafa Bouzi

In this paper, we present a modeling and simulation study of a hybrid energy system for isolated sites, which combines photovoltaic system with wind turbine based on permanent magnet synchronous generator (PMSG). With the aim of improving the performances and extracting the maximum energy produced by the two Photovoltaic (PV) and Wind sources, we developed a maximum power point tracking (MPPT) controller with robust algorithms based on the nonlinear sliding mode control (SMC). Thus, other conventional MPPT methods have been simulated and evaluated to compare them with the proposed algorithms in order to deduce the most efficient in terms of the maximum power reach and the system behavior during optimal power point tracking. And for a better analysis study, the simulations are carried out under variable weather conditions. The simulation results show that the proposed robust sliding mode control techniques (RSMC) are more efficient and have better performance than other MPPT algorithms.

314

Authors: P. Vo Ngoc, T. Vo Thi Ngoc

Sentiment classification has been used in many different fields because it has many significant contributions in everyday life, such as in political activities, commodity production, and commercial activities. We have proposed a new model for big data sentiment classification by using a combination of an unsupervised learning algorithm of a machine learning with a Ruzicka Coefficient (RC) in this work. A Self-Organizing Map Algorithm (SOM) of the machine learning is used in clustering the documents of the testing data set (TES) comprising 7,500,000 documents, which are the 3,750,000 positive and the 3,750,000 negative in English, into either the positive group or the negative group of our training data set (TRA) which is 3,000,000 sentences including the 1,500,000 positive sentences and the 1,500,000 negative sentences in English. In this study, we do not use a vector space modeling (VSM). We do not use any multi-dimensional vectors according to both the VSM and many sentiment lexicons. We use many sentiment lexicons of our basis English sentiment dictionary (bESD). We use many one-dimensional vectors based on the sentiment lexicons. We use a similarity coefficient in this study. We do not use any one-dimensional vectors based on the VSM. We have achieved 88.64% accuracy of the TES. The execution time of the proposed model in a distributed network environment - DNE is less than that in a sequential system - SS. Many commercial applications and surveys of the sentiment classification can widely use the results of the proposed model.

1

Authors: Preeti Ivy Barrow, Swathi Jamjala Naryanan

While the genome sequences of different crops have been published and annotated, relatively little is known about the transcriptional networks that regulate gene expression. There are different ways of finding significant motifs in genome sequences, evolutionary motif finding algorithms being one of them. The work presented in this paper uses Differential Evolution (DE) Algorithm to discover new motifs in rice genome sequences and aims to provide an intuitive wizard to analyse the micro-array analysis of expression patterns (motifs) for the queried genes. The used DE algorithm involves crossover operation with fine tuning. Further, to search for already known motifs parallel version of adaptive hash based pattern matching algorithm has been implemented. The study is conducted on rice genome sequences for twelve chromosomes each consisting of approximately 70000 genome sequences. In this paper, experimental results provide a comparison of serial and parallel version of the algorithm applied for the same dataset which shows that parallel implementation improves the performance by almost 25-30% to search the patterns in genome sequences and provide a list of newly discovered significant motifs.

20

Authors: Sribindu Sattu, Moparthy Gurunadha Babu

In cognitive radio systems; primary clients can use auxiliary clients as agreeable transfers to expand their transmission rates while auxiliary clients will consequently acquire more range to get circumstances, prompting coordinated cognitive radio systems. Earlier research proposals in Coordinated Cognitive Radio Networks (CCRNs) predominantly concentrate on giving omnipresent access and high throughput primary client for clients yet have infrequently contemplated vitality proficiency. Moreover, most existing works expect that auxiliary clients are inactively chosen by primary clients, paying little respect to auxiliary client’s eagerness to help, which is clearly not feasible. To address energy issue, this paper identifies an Energy-Efficient Cooperative Strategy (EECS) by leveraging temporal and spatial diversity of the primary network. In particular, auxiliary clients with delay-tolerant packets can proactively settle on the agreeable choices by together thinking about primary channel accessibility, channel state data, primary client’s movement stack, and their own particular transmission necessities. Principle challenge for this cooperative relay determination issue is to choose a relay effectively and economically. Since the quantity of auxiliary clients is expanded step by step because of the quickly developing number of wireless communication gadgets, it is unfeasible to examine all the auxiliary clients and after that pick the best. Predominantly primary client transmitter successively watches auxiliary clients. An arrangement of auxiliary clients is chosen in view of transmission criteria. The hand-off choice plan selects the best hand-off from an arrangement of accessible transfers which are chosen in light of transmission criteria of coordinated connections. In the observation of a hand off primary client need to settle on choice in regards to whether to stop its observation and to pick that as best hand-off or to skip to see next transfer. This problem is addressed by optimal stopping theory and optimal stopping rule. To evaluate the performance of proposed scheme Optimal Selection Approach (OSA), it is compared with auxiliary client’s energy efficient and existing cooperative schemes. Extensive simulation study is conducted so that impacts of different system parameters are investigated and algorithm proposed can satisfy different system requirements.

42

Authors: Jeong-Su Kim, Jeong-Beom Kim

The challenge to solve in large capacity ternary content addressable memory (TCAM) is the large power consumption of search-line and match-line. In this paper, we propose a circuit that uses match-line selective-charging scheme and low-swing search-line scheme simultaneously to reduce power consumption of search-line and match-line which occupies a large ratio in power consumption of TCAM. The circuit was verified by Hspice simulation using a 1.8V supply voltage in a CMOS 0.18 um CMOS process. The proposed 128x128 bit TCAM shows 45.4% power savings in match-line and 74.7% power savings in search-line compared to the conventional precharge high scheme in search cycle. Also, the minimum search cycle time is 3 ns (333MHz).

61

Authors: Nandoori Srikanth, Muktyala Siva Ganga Prasad

Due to shortage of power source, energy is the significant concern part in Wireless Sensor Systems. Wireless Sensor Networks (WSNs) are currently used in numerous kinds of WSN simulators have been established to visually illustrate the coverage, energy consumption status and expected lifetime of WSN sensors. Most of the Researchers have proposed and develop routing algorithm to increase the network lifetime. Among them clustering is the best technique and it is well known for accomplishing energy effectiveness in WSNs. In this paper, Energy Efficient Multi-hop Routing Clustering Protocol (EEMRCP) based on Fuzzy K-Means and Centralized Mid-point Algorithm (FKM-CMA) is proposed for network lifetime improvement. An existing and frequently used algorithm is K- means algorithm in WSN. The main limitation of this K-means algorithm is random initial centroid selection. To improve K-means algorithm, a Midpoint method in initial centroid selection is used. This is named as FKM-CMA. There are two main considerations for cluster head selection in proposed work, one is it’s residual energy, and another one is Euclidean distance used in basic Fuzzy K-means algorithm. Finally, Multi-hop communication is performed for transmitting the packets from CHs to Base Station depending on the distance between them. Furthermore, stimulated outcomes showed that the total network performance of the proposed approach is improved than the other existing approaches.

88

Authors: Udaya Dayanandan, Vivekanandan Kalimuthu

Software quality assessment at the architectural level is efficient because it provides various advantages to find and fix design errors at the earliest stage of software development life cycle (SDLC).Numerous methods have been evolved to validate quality-related issues at the architectural level. The software quality comprises of a total of eight attributes where maintainability is one of the important attribute which affects the overall system quality. Though few methods are available for maintainability analysis, none of the state of art method employs fuzzy logic with Quality Model for Object Oriented Design (QMOOD) metric suite. In this paper, the fuzzy analytic hierarchy process (FAHP) method is proposed with Buckley method to evaluate the influence of maintainability and its sub-attributes in software quality at the architectural level. The proposed evaluation model is tested against several versions of the MFC and OWL windows application frameworks and the obtained results are compared with the state-of-the-art methods. This experimental analysis shows that the FAHP with Buckley method is superior to the other compared methods.

107

Authors: Hemanth Somasekar, Kavya Naveen

Text categorization means dividing a set of input documents into the two or more classes to which these documents belong. Because of increase in availability of data in digital form in large amount, it becomes necessary to organize it. Feature extraction is the crucial step in text classification. Most of the existing text classifiers are lacking in finding out the relations among the terms. In this research work proposed text clustering of Improved Markov Clustering Model (IMCM) and text classification in which the nonlinear relations among the terms are also considered. This approach is scalable to huge dataset also and its classification power is not affected if relations among terms are large. The proposed IMCM method improved the text document clustering significantly and helped to avoid the human interaction in text clustering. Experimental results have shown that proposed system is 91% accurate for 8 categories and decreases, as increase the classes from 8 to 12, from 96% to 94%, respectively. Also, compared proposed model with existing Naive Bayes and k-Nearest Neighbor classifiers. Experimental results show that the proposed model is more accurate than these two classifiers. The better results demonstrated that presented system is developed effectively.

128

Authors: Mallaiah Vorsu, Vinaya Babu Arramraju, Sujatha Bhanothu

Object Oriented Program (OOP) provides the way to reuse the program by pre-implementing functionalities in the software. It is difficult to develop the object oriented software, which is important in the computer programing. For OOP, modularization mainly depends on the class. There are many methods present for the assigning responsibilities, but most of them are based on the human for decision making. In this research, back propagation neural network (BPNN) was used to provide the solution to object oriented design of the software. The Cinema Booking System (CBS) was taken as the input documentation and Formal Concept analysis (FCA) then found the relationship of the element in the lattice manner, after that the relationship was set with each other. The result showed that the proposed system outperformed the existing system and also the design made manually.

147

Authors: Mohammed Lahraichi, Khalid Housni, Samir Mbarki

Object tracking is a crucial step in video analysis, as well as detection and recognition. Particle filter, based on color histogram, is considered among multiple approaches that prove their effectiveness in this domain. It is apparent that if an object and its background, or more objects to be tracked, having nearly the same color histogram, the quality of tracking will be negatively affected. Therefore, combining color histogram and texture features for pixels in motion is considered a potential solution to this issue. In this paper, we propose a particle filter tracking method consisting of two stages: at the prediction stage, the state space model is represented by Box–Muller transform, and at the correction stage, the observation likelihood model is calculated in three steps: firstly, the background subtraction is used to extract the moving objects in order to reduce the impact of background pixels on color histogram. Secondly, the texture Gabor features are computed using moving pixels to distinguish between objects having the same color histogram. Finally, the histogram and texture features are combined to define a likelihood observation model. The proposed method is implemented using several video sequences. The visual and numerical results obtained are compared with those of traditional particle filter method, and show that our method is very efficient particularly in case of similar objects, and clearly improves the quality of tracking.

167

Authors: Pushpa Mohan, Kiran Kumari Patil

One of the most experimentally difficult problem in the world is weather forecasting, which is a basic mechanism in meteorology. Especially in data mining system, there are different information mining strategies are available, for example, K-Means, Artificial Neural Network (ANN) and Support Vector Machine (SVM), etc. These weather predicting strategies are financially high and also very inconsistent for large datasets. To overcome these issues, an effective dimensionality reducing strategy: Self Organizing Map (SOM) is proposed along with Latent Dirichlet Allocation (LDA). The SOM strategy is one of the proper dimensionality reducing strategy to highlight the self-arranging outline. After reducing the measurement, the dimensionality reduced information are used to forecast climate for a reasonable outcome. A reasonable season for an appropriate crop is arranged with the guide of Deep Neural Network (DNN) classification system. This research work depends on finding appropriate information model, which helps in accomplishing high precision and simplification for value forecast. Finally, the experimental outcome shows that the proposed approach improved accuracy in weather and crop prediction up to 7-23% compared to the existing methods.

188

Authors: Pranai Kungwalrut, Viriya Kongratana, Thanit Trisuwannawat, Vittaya Tipsuwanporn, Arjin Numsomran

Optimized water management strategies are among the most crucial concerns in a biomass power plant. Especially during the times of water scarcity, an improper water supply planning and operation result in a detrimental power generation or shut down processes. This paper aims to propose the cyber-physical cloud computing (CPCC), a mechanism of analysis and control physical process using cloud-based framework, in order to improve the efficiency of water management system in a biomass power plant. In this study, a 9.9 MW biomass power plant is implemented to investigate the performance of CPCC. The architecture of the proposed system consists of three physical tiers. The first tier is a tier of physical devices included with pressure, flow, pH sensors, water pumps and valves that are responsible for detecting the physical data and interacting with the water production process. The second tier is an edge computing device which functions as a controller, embedded server, data storage, gateway and switch to manage all operational tasks in an intranet area. The third tier is a cloud computing system which enables big data applications such as online monitoring and visualization of process operation, adaptive filtration fouling control, consuming water and total water cost analysis. The results validate the effectiveness of the proposed system as the ability of an adaptive fouling control system to adjust backwash scheduling and chemical dosing so that achieving the target of purified water quality even during the fluctuation of raw water qualities. Subsequently, the water treatment system can achieve the capability for optimal total water cost operations under a target permeate flow rate and the percent of the recovery. Significantly, the data analytics in the CPCC reflect current operational requirements of water under local climate conditions, likewise contribute to the practical solution for sustainable water resource planning in the energy production.

212

Authors: Mylam Chinnappan Babu, Sankaralingam Pushpa

Discrimination is one of the most important challenging tasks in web mining due to its many legal and ethical features in social media and enterprise based industries. There are an enormous amount of anti-discrimination measures available to prevent discrimination such as using some features like race, religion, gender, nationality, disability, marital status, and age besides situations like employment and training, access to public services, credit, insurance, etc. Practically, those systems are not possible to use in industries due to large datasets. Indirect discrimination contains a set of rules or techniques which are not explicitly specifying discriminatory features, deliberately or accidentally and could create unfair decisions. Existing systems have low classification accuracy and data loss with high discrimination data detection time. To overcome these limitations, an Efficient Discrimination Prevention and Rule Protection (EDPRP) approach has been proposed for removing the discrimination and protects the rule without damaging the data quality. The proposed system designing pre-processing discrimination prevention approach and specify the different features and represent to deal with direct or indirect discrimination. EDPRP is capable of preventing Indirect and direct discrimination, and it allows automatic and routine collection of large amounts of data from the public. In EDPRP, the discrimination prevention model is based on partial data sets as part of the automated decision making. Based on Experimental evaluations, proposed method improves 8% (percentages) of Support and 8 ms (milliseconds) of Execution Time compared than existing methods.

231

Authors: Doaa Elsayed, Eman Nasr, Alaa El Din Ghazali, Mervat Gheith

Web Service Composition (WSC) has attracted considerable attention and research to support Service Oriented Architecture (SOA). WSC constructs complex applications by combing atomic Web services together to achieve users’ requirements. The selection of the best Web service that fulfils the Functional Requirements (FRs) and optimizes the Quality of Service (QoS) requirements, such as response time, cost, reliability, etc., is a critical part of WSC, especially in a dynamic environment. Since requirements frequently change, this demands that WSC must be adapted to provide the most suitable composite services and fulfil the users’ requirements that emerge. This paper presents a new hybrid approach for dynamic optimization of WSC using Parallel Genetic Algorithm (PGA) based on Q-learning, which we integrate with K-means clustering. We call it PGAQK. Q-learning is utilized to generate an initial population to enhance the effectiveness of PGA. K-means clustering is used to cluster Web services based on a fitness function. It is applied in the mutation operator, and used for Web service substitution to prune the Web services in the search space to find the best Web services for the environment changes that might occur at runtime. To the best of our knowledge such a hybrid approach has not been used for WSC before. We implemented PGAQK over the .NET Framework using C# programming language. A series of comparable experiments were carried out showed that PGAQK outperforms traditional PGA and Q-learning approaches in terms of fitness values. PGAQK allows WSC to modify itself dynamically and achieve a better fitness value to fit the changing environment, where the characteristics of composite web services continue to change.

250

Authors: Deepak Nandal, Om Prakash Sangwan

This paper estimates the effort for software by optimizing the COnstructive COst MOdel (COCOMO) model parameters using hybrid BATGSA (Bat inspired Gravitational Search Algorithm) algorithm. The performance of the COCOMO model completely depends upon its parameters which can be optimized by using meta-heuristic algorithms. This paper uses hybrid BATGSA algorithm which hybrids the improved bat algorithm with the gravitation search algorithm (GSA) to optimize the COCOMO model. The bat algorithm demonstrates the hunting and routing behavior of the bat which is improved by using a random walk in the exploration phase. The exploration phase is further improved by using GSA as gravitation force affects the velocity of the bat. The algorithm has been analyzed on four NASA datasets downloaded from promise repository. The comparison of the algorithm has been made with existing three states of art techniques i.e. COCOMO model, BAT algorithm, Improved BAT(IBAT) algorithm by using normalized error as a parameter. The reduction in error ranges from 2% to 10% on different dataset as compared other state of art algorithms proves the significance of proposed algorithm.

275

Authors: Mukesh Bansal, Chaitanya Purushottam Agrawal

This paper develops a technique by using Jaya algorithm and feed-forward neural network to determine the quality of object-oriented software by using Chidamber & Kemerer (CK) along with Li & Henry metrics. The technique basically focuses on the maintainability factor of software quality which in turn depends upon the software complexity. The software complexity is directly proportional to the number of changes done per class which is determined by the technique. The analysis has been done on UIMS (User Interface Management System) and QUES (Quality Evaluation System) datasets by using the mean absolute error as the analysis parameter. The reduction in the mean absolute error as compared to the existing state of art techniques along with the individual component of proposed technique proves the significance of the technique.

293

Authors: Maddela Vijaya Lakshmi, Pardhasaradhi Pokkunuri, Boddapati Taraka Phani Madhav

A coplanar waveguide fed asymmetric ground frequency reconfigurable antenna is designed and analyzed in this paper. The proposed antenna is occupying the compact dimension of 50X45X1.6 mm on FR4 substrate with dielectric permittivity 4.4. The designed antenna is operating in the wideband from 2 to 5.8 GHz with coverage in applications like Bluetooth, LTE, Wi-Fi and WLAN. The reconfigurable nature of the antenna is analyzed with the switching of PIN diodes placed on the antenna structure. The on and off conditions of the diodes D1 and D2 providing switching between the LTE, Wi-Fi and WLAN bands with good impedance matching. The advantage of PIN diodes includes high power handling capability, very low driving voltage and extremely low cost. Here the switch placement and control/biasing strategies depend on the type of the switch and the topology of the antenna. The prototyped antenna with biasing is providing excellent correlation with High Frequency Structure Simulator (HFSS) and Computer Simulation Technology (CST) simulation results. The reconfigurability of the antenna and its parametric optimization is analyzed and presented in this work.