International Journal of Intelligent Engineering and Systems

13

Authors: Jagadeesh Basavaiah, Chandrashekar Mohan Patil

Action recognition schemes enable several intelligent machines to recognize human action by using daily life videos. In the last decades, recognizing human actions in the video sequences have been a challenging problem because of its real-world applications. Several action representation techniques have been done to improve action recognition performance. Approaches based on local features and classification are used for action representation, but it failed to capture temporal relationship between actions. In this work, Video Action Recognition (VAR) is assessed by using Weizmann, KTH and own datasets. Initially frames are extracted from input videos and the frames are resized. After the pre-processing, the object detection is done by Blob detection algorithm and tracking the object frame by frame is done using Kalman filter. The features are extracted from the moving object using feature extraction algorithms such as Bi-dimensional Empirical Mode Decomposition (BEMD), Scale Invariant Feature Transform (SIFT) and Discrete Wavelet Transform (DWT). These feature extraction techniques were applied on pre-processed frames to extract the efficient features from multi-scale images. Finally, the features are given to Convolution Neural Network (CNN) classifier for action recognition prediction. The proposed method is called as Hybrid Feature Extraction (HFE) and CNN used for VAR (HFE-CNN-VAR) method. The experimental results showed that the HFE-CNN-VAR method improved the accuracy in action classification. The classification accuracy is 99.01% for KTH dataset, 99.33% for Weizmann dataset and 90% for own dataset.

35

Authors: Charles Brijilal Ruban, Balasubramanian Paramasivan

Vehicular Ad-hoc Network (VANET) has wide application in developed/developing nations, for the effective management of vehicular traffic, and inter vehicular communication. In vehicular communication, the data includes traffic alert and safety warning. In most of the previous research achieved better quality in VANET routing. In some research, the researchers also proposed the security model for the secure transmission of these data. Instead of giving importance to the security of these data, trust ability of data and data transmitter is essential in VANET communication. In this paper a new protocol called Virtual Trust-ability Data transmission (VTD) is proposed for the effective handling of VANET. In the proposed protocol, three stages are used in the first stage a virtual topology is crated using a simple clustering technique. Then in the second stage the trust ability of every node or vehicle is analysed. In the last stage, the data are transmitted via the trusted vehicle and remove the false data from the communication. Then the performance of the proposed VTD protocol is analysed based on packet failure rate, bandwidth utilization and network scalability. The proposed VTD provided 94% of packet delivery ratio with 6% of the average packet failure. Similarly, the other performance also improved as compared to the conventional technique. So, it is clearly proving that the proposed VTD protocol can become a better alternate for the trust aware routing in VANET.

57

Authors: Pitchaya Boontra, Tanairat Mata, Pisit Boonsrimuang

One of the problems of using Orthogonal Frequency Division Multiplexing (OFDM) technique is the higher Peak-to-Average Power Ratio (PAPR) which leads to the fatal degradation of the system performance in the non-linear channel. To solve this problem, the partial transmit sequence (PTS) was proposed for Space-Time Block Code (STBC) with Multi-Input and Multi-Output (MIMO)-OFDM as STBC MIMO-OFDM system which can reduce the PAPR effectively. However, its complexity in the PAPR reduction process becomes higher in proportion to the increase of its considered cluster number. To solve this problem, this paper proposes a low-complexity PAPR reduction by using the modified-PTS with artificial bee colony (ABC) and concurrent algorithm as ABC-Concurrent algorithm for STBC MIMO-OFDM system which can improve better PAPR with low-complexity and can perform better Bit-Error-Rate (BER) in the non-linear channel. The silent features of the proposed modified-PTS with ABC-Concurrent algorithm are to reduce the PAPR with decreasing side information (SI) to half by modifying PTS with the concurrent algorithm and to reduce the complexity in PAPR reduction process by applying ABC algorithm for STBC MIMO-OFDM system. From the various results by using computer simulation, it can be confirmed that the proposed modified-PTS with ABC-Concurrent algorithm can reduce PAPR by approximately 3.4 and 0.4 dB at CCDF 10-3 as comparing with the original STBC MIMO-OFDM and conventional PTS methods respectively, and can use the complexity approximately only 19% in PAPR reduction process which is decreased a lot from using the conventional PTS method. Moreover, the proposed method can perform the better BER in the non-linear channel than the original STBC MIMO-OFDM and the conventional. PTS methods which is close to that of operating in non-linear channel.

77

Authors: Rashid H. Al-Rubayi, Mohammed Kdair Abd, Firas M. F. Flaih

Economic dispatch issues in power system aim to try getting an optimal plan for the power generators to minimize the fuel cost (FC) in parallel with satisfying system constraints. This paper proposes a new enhancement based on particle swarm optimization (PSO) algorithm called multiple inertia weight PSO (MIW-PSO) to solve the combined economic and emission load dispatch (CEELD) issues in the modern electrical power systems. Two electrical test systems are investigated in this study to validate the competence of the proposed algorithm. The obtained results for CEELD case using MIW-PSO compared with MOCPSO indicate a promising performance in terms of minimizing FC and pollutant emission (PE) are reduced 84.96 $/h and 12.01 kg/h for the first test system. As well as, for the second test system, compared with NSGA-RL are reduced 0.241 $/h and 3.15 kg/h. Moreover, the proposed algorithm has more accuracy, better convergence time, and higher quality solutions for the minimum CEELD compared with other methods.

98

Authors: Eman Karam Elsayed, Dina Refaet Salem, Mohammed Aly

Noise can affect images while acquired, transmitted, stored or compressed. One of the best methods for noise removal is the sparse representation algorithm (SR). The Quantum Particle Swarm Optimization (QPSO) is one of the meta-heuristic algorithms. This paper shows excellent results in noise reduction in the quick version of QPSO, which uses benefit of the SRs and meta-heuristic algorithms. This approach is known as FQPSO-MP, depending on the matching pursuit algorithm (MP). A proposed Dynamic-Multi-Swarm (DMS) and a pre-learned dictionary (FQPSO-MP) method saves the time of calculating the learning dictionary. These modifications contribute to important benefits of computing efficiency (productivity improvements of approximately 90% are achieved) without sacred image quality in comparison with the initial QPSO-MP technique (the bigger reduction relative to the PSNR indexes is lower than 0.58 dB and 0.019). The proposed FQPSO-MP method compared to the original QPSO-MP method after modification. The scientific results show that the FQPSO-MP algorithm is more effective and quicker without sacrificing image quality than the FQPSO-MP algorithm. The experimental results show, in comparison to state-of-the-art denoising algorithms, that both quantitative and image quality results are achieved with the suggested FQPSO-MP method.

124

Authors: Arif Muntasa, Muhammad Yusuf

One of the most dangerous cancer is acute leukemia lymphoblastic (ALL). This type of leukemia has been attacking the most of children. This research aims to detect the ALL diseases through the principal features of the object. Furthermore, this paper contributes by providing a new approach to detect the ALL disease based on the Color Orthonormal Basis Entropy (COBE) and the Distribution of the Pixel Intensity (DoPI). There are some novelties in this research. Firstly, we combined three channels features to characterize an object for increasing the performance accuracy. Secondly, we have developed a model to improve the image through a contrast enhancement filter, the noises from the image can be recognized and removed automatically, and better image result has been delivered through enhancement of the green channel selection. In this research, we proposed a novel approach to detect the ALL based on the COBE and the DoPI. This new approach is divided into four main stages, i.e., image enhancement, segmentation, feature extraction, and performance measurements. Firstly, an image was enhanced through a channel separation, enhancement of the green channel, and fusion of the green channel enhancement with the original channel, i.e., the red and blue channels. An image enhancement process is conducted by a gamma correction of the contrast stretching. The hue channel is an output of the first stage. Secondly, the Otsu algorithm was employed to obtain the best threshold. Thirdly, the COBE and the DoPI were included to discover the main features. Lastly, the Manhattan method was employed to measure performance. In this research, the second Acute Lymphoblastic Leukemia-Image Database (The 2nd ALL-IDB, usually it is called as ALL-IDB2). It is carried out to evaluate the proposed approach. The results show this proposed method has produced high accuracy at 91.67% and small standard deviation at 0.022.

146

Authors: Boddepalli Kiran Kumar, Edara Srinivasan Reddy

Information Centric Network (ICN) is one of the growing network environments that provides the majority of internet activities, which are related to information access and delivery. In ICN, users can obtain information contents of the caches in ICN routers. However, how to use the caches effectively is one of the main challenges faced by the researchers. Thus, several research studies are developed for reducing the cache overlaps among routers, since that limits the interests towards the caches. In order to address this problem, a new shortest path estimation algorithm (Modified Floyd Warshall Algorithm (MFWA)) was developed to identify the shortest path in each subnet for retrieving the cache memory data quickly. In MFWA, a new iterative matrix was used to eliminate the invalid paths that helps to retrieve the cache memory data quickly. In this research work, an ICN network was generated based on the concept of GEANT topology and tested with dissimilar cache management policies such as Hash Routing Symmetric (HR-SYMM), Hash Routing Asymmetric (HR-ASYMM), Hash Routing Hybrid Asymmetric Multicast (HR-HYBRID-AM), Hash Routing Hybrid Symmetric Multicast (HR-HYBRID-SM), and Hash Routing Multicast (HR-MULTICAST). In experimental section, the proposed shortest path estimation method showed better performance in light of latency, link load and cache hit ratio as compared to the existing approaches: Floyd Warshall Algorithm (FWA), Dijkstra’s Algorithm (DA), and Enhanced Dijkstra’s Algorithm (EDA). The experimental outcome showed that the proposed approach improved cache hit ratio up to 4-30% as related to the existing methods.

167

Authors: Yuni Yamasari, Supeno M. S. Nugroho, Kayo Yoshimoto, Hideya Takahashi, Mauridhi H. Purnomo

The rapid growth of information and communications technology-based educational tools generates a large volume of student data with many features (characteristics). However, the mining process in the clustering task of student data is not often done optimally, so the performance of the system decreases. To overcome this problem, we propose a discretization method on logistic regression to determine the most optimal number of clusters. Additionally, we introduce a technique that combines the features selection using a filter- and wrapper-based procedures (HFS) to identify the dominant features of the students' cognitive domains. Furthermore, we evaluate the identification result by three clustering methods, namely: K-means, EM, and Farthest first. Finally, we propose the clustering-based classification so the results can be measured by using the classification metrics. Here, we apply two evaluation techniques, namely: cross-validation and percentage split. The experimental results indicate that our approach describes predominance, in terms of classification metrics over conventional methods. Our approach is around 10,847-11,134 percent higher in terms of accuracy average than the original features on both the assessment techniques. Also, this approach significantly reduces the time taken to create a prototype between 0.0167-0.027 seconds. This gives the impact on a significant reduction in the model created to the number of unsuitable students on classes based on the cognitive domain, namely: 3-12 students.

191

Authors: Maganti Syamala, Nattanmai Jeganathan Nalini

E-Commerce product features and reviews are considered to be the essential factors in real-time e-commerce sites for product recommendation systems. Due to inaccuracy decision patterns, in most cases e-commerce user fails to predict the products based on the user ratings and review comments. Traditional sentiment classification models are independent of data filtering, transformation and sentiment score computing techniques which require high computing memory, time and mostly leading to false-positive rate. To overcome these issues, a novel sentiment score-based product recommendation model is proposed on the real-time product data. In this model, a new product ranking score, filtering, and hybrid decision tree classifiers are implemented. Initially, real-time amazon product review data is captured using Document Object Model (DOM) parser. The features from the review comments are extracted using lexicon Feature Dictionary (FD) and AFINN, Normalized Product Review Score (NPRS) are generated to compute the class label for product review sentiment prediction. Ranked Principal Component Analysis (RPCA) is used as a feature selection measure to overcome the problem of data sparsity. Random Tree, Hoeffding Tree, Adaboost + Random Tree, the three variants of decision tree classifiers are used for product sentiment classification. The proposed filter-based improved decision tree sentiment classification model for real-time amazon product review data recommends the product based on the user query by prediction using a new novel normalized product review sentiment score and ranked feature selection measure. The proposed product recommendation, the decision-making system maximizes sentiment classification accuracy. Experimental results are compared against the traditional decision-making classification models in terms of correctly classified instances, error rate, and PRC, F-measure, kappa statistics. The proposed model experimental results show high efficiency.

214

Authors: Imad Sassi, Samir Anter, Abdelkrim Bekkhoucha

Hidden Markov Models are widely used for time continuous problems modelling and prediction. This paper presents two new improved algorithms for Gaussian continuous and mixture of Gaussian continuous Hidden Markov Models cases for solving learning problem for large scale multidimensional data. The design of our parallel distributed algorithms is based on Spark, the Big Data framework, thereby we can distribute data over several nodes through Resilient Distributed Datasets which allow to apply, in parallel, a set of operations. The proposed algorithms have two main advantages: a high computational time efficiency and a high scalability as well as an easy integration in Big Data frameworks. The complexity comparison results show great improvements in computational complexity (by a factor of (states number)2) and execution time. Moreover, the new algorithms might be more effective by reducing the communication costs between the elements of the system involved in the learning task.

237

Authors: Yuning Widiarti, Wirawan Wirawan, Suwadi Suwadi

Intersymbol interference (ISI) is one of the main issues in underwater acoustic communication (UAC) link development. Meanwhile, Time Reversal (TR) precoding is a promising method to mitigate ISI by means of its inherent property of spatial-temporal focusing. For the ISI problem solution and a reliable UAC link achievement, the authors propose the combination of TR precoding and maximum ratio combining (TR-MRC) techniques applied to a single-input-multi-output-OFDM (SIMO-OFDM) scheme. The system performance is observed based on Monte Carlo simulations and the analytical expressions in terms of bit-error-rate (BER) are presented to support the simulation results. There are three schemes compared, i.e.: proposed method (TR-MRC), TR-equally gain combining (TR-EGC) and simple TR. The three schemes are compared in three conditions of water depth. The analytical performance and simulation results show that the proposed method has the best performance between the other two schemes in three water depths of the towing tank. By using 2 hydrophones as a receiver array, the proposed method is proven to be superior to the simple TR. The superior performance produced up to 6 dB compared to the TR-EGC scheme and 8 dB over simple TR when the BER value is 0.1.

259

Authors: Nattapon Jaisumroum, Pholchai Chotiprayanakul, Wipoo Sriseubsai

In this article, the manipulation to handle the object on a plate using neural networks, design of 1-DOF robot arm under cooperative control, will be explained. The robot’s system specifies the object position and velocity, an assignment oriented components for cooperative control. The novelty of this experiment is that under the decentralized control, the robot estimates the position and speed of the object for control end-effector of robot arm using the camera to track position and speed of the object according to training and assigned collaborative tasks which differ from other experiments that use sensors. The experiment includes three robot manipulators which were capable balancing the objects on flat plate with dataset to training and control servo motor assigned to the corresponding position and the end-effector in decentralized to control robotics. Overall, neural network method can be a training scheme using a cooperative robotics in a decentralized control.

282

Authors: Younes El hachimi, Abdelmajid Farchi

A miniaturized tag antenna is presented in this paper, for operating in the RFID UHF American band, in free space or in a metallic environment. The proposed antenna is printed on a single layer FR4 substrate, and designed first to work in a free space. The shape of the π-match allowed us to achieve a good matching between the antenna impedance and that of the chip. The correspondence between simulation results in CST and HFSS pushed us to manufacture a prototype of our tag. After that, we simulated the proposed antenna in a metallic area, by adding a square metal plate separated with the antenna by a foam-layer, in order to test the performance of the tag in this environment. An optimization procedure enabled us to achieve a good performance when operating in metallic surroundings. Finally, we tested the reading range of the manufactured tag. We obtained a good range of about 6.5m. The final design of our proposed tag has a simple structure, a dimension of 51×26,63×0.8

301

Authors: Balasubramanian Gayathri Devi, Venkatasubbu Pattabiraman

Recommender systems are in recent past becoming more notable in the age of swift evolution of the information technology ensuring e-commerce users’ suitable items. In recent years, various thread-based and trust based approaches have been proposed, which improve the precision and recall of recommendation to certain extent. However, these approaches are less accurate than expected when users’ preferences evolve over time and when several aspects of the users are involved. To solve these problems, we propose a novel recommendation method called, Soft Cosine Gradient and Gaussian Joint Probability (SCG-GJP) for Online Social Networks (OSN). Specifically, the Local Soft Cosine Clique Clustering (LSCCC) is first used to identify good candidate items to be recommended. Second, using adjacency matrix factorization, probable past preference of individual user and overall preference of whole community is formed to reduce the problem of mean absolute error. Again, the obtained matrix is trained through the stochastic function gradient to rank the most suitable candidates. Finally, Gaussian Mixture Joint Probability Post-filtering model is proposed to provide useful recommendations. The experiments are executed with real-world dataset (DS) and compared with the conventional recommendation methods. Experimental results demonstrate that the SCG-GJP obtains higher accuracy, and lesser error as well as complexity.

324

Authors: Amira M. Idrees, Essam M. Shaaban

This research aims at focusing on building a knowledge base shell which task is to automatically build the knowledge base component from text sources. The proposed research aims at considering the Arabic text sources as one of the main challenges in due to the difficulty in processing Arabic text that suffers from the high inflection, shortcuts, and the varieties in letters’ representation. The research restored the confidence in the term frequency method and the research adopts the semantic relations network representation. The research applied the experiment for building the weeds’ identification facts’ part in the knowledge-based system. The experiment applied the proposed framework on two sources, the precision, recall, and f-score have been measured which resulted in an average of 91.2%, 91.1%, and 91% respectively. The evaluation metrics provide a promising perspective in the proposed approach with more focus on the bottlenecks for enhancements.

344

Authors: Panem Penchala Prasad, Fidele Sagayaraj Francis, Soudagar Zahoor Ul Huq

Nowadays, several researchers are facing challenges on the prediction of diseases from the huge volume of medical databases. So, researchers are using data mining techniques like association rules, classification and clustering to address challenges. The physicians make the right decisions for successful diagnosis of various diseases by using the prediction. The existing work classifies the data to predict the certain diseases, but still it faces the difficulties due to overfitting in the training data. The main aim of this research work is to classify the Medical Data (MD) by developing the feature selection based approach in MD Classification as MDC. The irrelevant features are eliminated from the MD by using Recursive Feature Elimination (RFE) method, then ranked the features to reduce the computation cost of the proposed method. The ranked features from the RFE are given as input to the Fuzzy Neural Network (FNN) with Reinforcement Learning (RL), which is used for classification. The proposed RFE-FNN method has the accuracy of the 98.57%, 98.15% sensitivity, 98.64% specificity and 95.47% F-Measure in Heart disease dataset.

364

Authors: Asepta Surya Wardhana, Mochamad Ashari, Heri Suryoatmojo

To concentrate the rays to the focal point and stabilize the temperature on the receiver were one of the complex works of the two-stage solar concentrator. This paper offers a new method of Fuzzy cascade controller system based on tuning up the optimization at Genetic Algorithm-simple additive weighting (GA-SAW) on the dual parabolic dish concentrator with Compound Parabolic Concentrator. The model and computation of three degrees of freedom robotic arm movement with the ray tracing method were used as the concentrator position predictor. In this research, the acquired fuzzy controller result had an average settling time 0.497 sec, and average rise time 0.277 sec faster than a conventional PID controller. This research was able to overcome the disturbance which was diffused rays, so there was a better output with the power and heat flux increased up to 62.49%. The monte carlo ray tracing method from Tonatiuh Software was used to investigate and movement validation of the dual parabola concentrator by showed the flux distribution on the receiver’s absorber. In the last stage, an experiment with a prototype had been conducted as a response verification of the controller system which produce a receiver temperature output as high as 121ºC. The final result showed that the controller system managed to optimize the temperature on the receiver’s absorber and generate a stable thermoelectric output with power of 1.01 Watt.

14

Authors: Suttipong Boontaklang, Chow Chompoo-Inwai

Driving ultrasonic piezoelectric transducer (UPZT) at maximum performance in real-time operations is a very challenging task due to the deterioration problem of ultrasound output causing by human-skin impedance fluctuations. This paper proposed ideas and solutions on how to maximize the UPZT driving circuit performance in or deter to solve the deterioration problem. A newly-designed feedback control block with a phase-locked-loop (PLL) technique controlled by microcontroller has been added to the conventional UPZT driving circuit to ensure the automatic resonance frequency tuning and tracking capabilities. After intensive experiments, it is clear that the proposed technique works flawlessly and much more efficient compared to the conventional one. The experimental results also confirm that the percentage of error in tuning to the desired frequency is less than ±0.2% and the speed of tuning convergence time is between 9-12 milliseconds on average.

37

Authors: Apichata Thongrak, Suchada Sitjongsataporn, Sanya Khunkhao, Phichet Moungnoul1

As described of electrical circuit have been considering that there are three fundamental passive two-terminal circuit elements; resistor, capacitor, and inductor respectively. In 1971, Prof. L. Chua proposed and described memristor which defines the relationship between magnetic flux and charge. This paper are purposed a memristor emulator circuit based on operational transconductance amplifiers (OTAs). The proposed circuits can be realized using commercially integrated circuits OTA, op amp as AD844, TL084, inductor, capacitor and resistors. The characteristic of memristor emulators can be examined in a practical experiment with the active and passive components. As described, the results could be demonstrated with the memristor circuit application. Furthermore, OTA is used to realize electronically tunable current conveyors emulator circuit. As a result of simulation, the decremental and incremental memristor emulator circuit is suitable for connecting in a series circuit. It was found that the frequency-dependent pinched hysteresis loop of proposed memristor emulator circuits can be updated by changing the value of capacitance and resistance, besides it increases the current gain of current conveyor. In addition to the expected results of memristor (RM), Memcapacitor (CM) and meminductor (LM) are proposed to determine in time-domain characteristics of R-L-C mode circuits. The results of experimental are discussed of phenomena studies by the memory characteristics.

59

Authors: Ammar Fahim Majeed, Ibrahim A. Murdas

Patients who are in healthcare places need for continuous healthcare monitoring system for long periods to surveillance human body vital signs. This system should be secure, reliable, and guarantee not to interfere with available radio frequencies or sensitive electronic devises such MRI (magnetic resonance imaging) since it deals with human life. In this paper, we designed, implement and evaluate a prototype model for Li-Fi healthcare monitoring system over different propagation links as an integrated optical system. The key idea of the created system is to remove the optical background noise power and to ensure patient's mobility at distance up to 4 m and 2.5 m in Los (line of sight) and diffuser links respectively. The system is created to receive data from a fibre Bragg gratings sensor (FBG) that will be sent to the end receiver through different propagation links having 100 % accuracy. Also, it is particularized with low power consumption, low-cost components, and could be adapted to different other systems of sending or processing operation. The proposed system provides, much more accurate received data compared to previous related literature.

81

Authors: Shoffi Izza Sabilla, Riyanarto Sarno, Kuwat Triyana

This research studied the use of sensors in an electronic-nose (e-nose) system to select the most efficient features to detect three classes of data signals. The data signals were collected from sensors that were assembled specifically for recognition of meat types. The development of e-noses has been the subject of many researches. However, researchers tend to install multiple sensors without optimizing their utilization, leading to higher costs. This research proposes three significant contributions to optimize the number of sensors to be used in e-noses, i.e. (i) tuning the threshold of the correlation coefficient; (ii) obtaining significant features by determining the optimal threshold of independent features; (iii) optimizing efficiency by considering the significant features and the accuracy of the results. The proposed efficiency method suggested that the number of sensors in the system could be reduced from 10 to 5. The optimal threshold suggested that 12 out of 20 features were correlated to each other. The highest accuracy obtained in this research was 92% for three classes of meats.

100

Authors: Hebatullah R. Younes, Ahmed Sharaf Eldin Ahmed, Eman K. Elsayed

System of Systems (SoS) is a huge system consisting of small stand-alone systems to offer unique abilities. In such complex construction, managing achievement of SoS objectives along with those of its Component Systems (CS) is a challenge. Whereas the integration of CS may cause an extended impact issue called conflict. In this research, the principles of basic belief function were used to measure conflict in the SoS. This is intended to detect conflicts between CS in an attempt to resolve it. Moreover, we have based on the ontology in the SoS integration. In addition to the identification of relationships between the CS semantically. We have proposed a method to resolve conflicts called Ontology Belief Function System of Systems (OBFSoS). As a homogenous SoS, we apply the OBFSoS on Real-lime Strategy (RTS) game. We depend on the use of a learning agent. It reduces conflicts and saves time due to attempts. For execution, we got the result of 10 attempts with about 30 systems (players) in the SoS. It giving a good decision in a short time with an accuracy of approximately 87%.

123

Authors: Kokila Subramanian Surampatti, Brindha Devi Cinniyampalayam Lakshmanan

MANET is severely energy constrained, due to the capricious mobility rate and the deployed hostile environment. The most energy consuming activity of mobile sensors is the process of routing. Hence, the energy efficient routing is an evergreen research topic for MANET. Additionally, the energy efficient routing algorithm must be as lightweight as possible to conserve the energy. Considering these points, this work aims to present an energy efficient, trustworthy route selection algorithm based on trust index. The trust index is formed by considering the trust metrics such as energy backup, packet forwarding rate and node loyalty. The trust index is considered as the fitness value for the route selection algorithm. The route selection algorithm is based on the metaheuristic algorithms such as Artificial Bee Colony (ABC) and Whale Optimization (WO) algorithms. The performance of the proposed work is analysed in terms of packet delivery rate, average loyalty, energy efficiency and network lifetime. The proposed work shows better performance than the comparative approaches with increased energy efficiency and thereby network lifetime.

147

Authors: Bouchiba Guelta, Redouane Tlemsani, Samira Chouraqui, Mohamed Benouis

This paper presents multi-modal biometric authentication approach using gait and electrocardiogram (ECG) signals, which can diminish the drawback of unimodal biometric approach as well as to improve authentication system performance. In acquisition phase, data sets are collected from three different databases, ECG-ID, MIT-BIH Arrhythmia database and UCI Machine Learning Repository (Gait). In Feature extraction phase of both signals (ECG and Gait) is performed by using 1D-local binary pattern. Features are obtained by merging two modalities as one feature. In classification approach, three classifiers are developed to classify subjects. K-nearest neighbour (KNN), relying on Euclidean distance, PNN (Probabilistic Neural Network), RBF (Radial Basis Function) and Support Vector Machine (SVM), relying on One-against-all (OAA). The proposed multimodal system has been tested over 18 subjects, and its identification accuracy was about 100%. Our result demonstrate that our approach outperforms rather than unimodal biometric system in terms of Correct Recognition Rate, Equal Error Rate, False Acceptance Rate and False Reject Rate.

167

Authors: Shidiq Nur Hidayat, Aldin Rusman, Trisna Julian, Kuwat Triyana, Ana C.A. Veloso, António M. Peres

An electronic nose (E-nose), comprising eight metal oxide semiconductor (MOS) gas sensors and a moisture-temperature sensor, was used for classifying three quality grades of superior java cocoa beans, namely fine cocoa dark bean < 20%, fine cocoa dark bean > 60%, and bulk cocoa bean that is a harder task compared to the discrimination of high versus low-quality cocoa beans. The E-nose signals were pre-processed using the maximum value method. The capability for discriminating the quality grade of the cocoa beans was checked by applying multivariate statistical tools, namely, linear discriminant analysis (LDA), support vector machine (SVM) and artificial neural networks (ANN). For this, the experimental dataset was split into two subsets, one for training (i.e., establishing the classification models) and the other for external-validation purposes. Furthermore, hyperparameter optimization and K-fold cross-validation variant were implemented during the model training procedure to select the best classification models and to avoid over-fitting issues. The best predictive classification performance was obtained with the E-nose-MLP-ANN procedure, which allowed 99% of correct classifications (overall accuracy) for the training dataset and 95% of correct classifications (overall accuracy) for the external-validation dataset. The satisfactory results clearly demonstrated that the E-nose could be applied as a quality control tool in the cocoa industry, requiring minimum and simple sample preparation.

199

Authors: Guruh Fajar Shidik, Edi Noersasongko, Adhitya Nugraha, Oki Setiono

Dynamics Virtual Machine management (VM) is a solution for reducing the huge resource and energy usage problems in the cloud data center. By managing the VM usage, it can decrease the operating cost and mitigate the spread of carbon dioxide (CO2) into the environment. This research presents the used of Fuzzy C-Means clustering to select VM in an overloaded host and distributes it into another available host. The proposed method was implemented in workloads dataset which provided by CloudSim PlanetLab. From evaluation process, the proposed method can produce better performance in reducing energy consumption, especially using four clusters in Heterogen workload comparing with other VM selection techniques such as Random Choice and Minimum Time Migration with reduction percentages are 9.34% and 3.89% respectively. This evaluation result denoted that our proposed method has better performance in reducing EC in the cloud data center.

215

Authors: Shashidhara Doddamane Nagendrappa, Chandrappa Dasanapura Nanjundaiah, Puttamadappa Chaluve Gowda

Prefetching the data is a popular technique that improves data accessibility in wired or wireless networks. In mobile ad hoc networks(MANET), the possibility to improve the access latency and cache hit ratio may reduce due to the mobility of nodes and only limited cache space available in mobile hosts (MHs). To eliminate this problem in wireless networks, Global Cooperative Caching(GCC) with Image Prefetching (IP) algorithm is proposed that pre-fetches the image based on similarity measurement between data items and integrated with Gateway Selection Algorithm (GSA) for increasing throughput and reducing delay in the networks. Design implementation and analysis are done using Network Simulator 2(NS2) and MATLAB tool used to calculate the performance parameters of the proposed algorithm and also the parameters are compared with GCCIP to prove that proposed GCCIP-GSA algorithm have greatly improved the system performance in terms of packet loss, delay, throughput and resource utilization compared to existing schemes. Throughput analysis shows that GSA GCCIP results 13.63% higher throughput and 31.25% reduction in latency than conventional GCCIP. The Mean Square Error(MSE) and Peak Signal to Noise Ratio (PSNR) has been evaluated for different kind of images such as cameraman, lena, and baboon. In Proposed GCCIP-GSA, 28.8%, 23.75%, and 18.66% of MSE has been reduced compared to GCCIP for three different images respectively. At that same time, 60%, 55.6%, and 50% of PNSR has been improved in proposed method than conventional method.

236

Authors: Cahyaningtyas Sekar Wahyuni, Kelly Rossa Sungkono, Riyanarto Sarno

Existing methods, such as Graph Edit Distance (GED) and Cosine measure, still have drawback in obtaining similarity of parallel relationships by neglecting the control-flow patterns, i.e. AND, OR, and XOR. Since AND > OR > XOR, the similarity value of AND versus OR is greater than XOR versus OR and AND versus XOR. This paper proposes two new similarity methods, Tree Declarative Pattern Edit Distance (TPED) and Cosine-Tree Declarative Pattern (Cosine-TDP). They provides value to the control-flow pattern so the value of similarity can be seen more differently. The new methods utilize tree model of the declarative pattern. The results show that the proposed methods are better at differentiating parallel relationships than the existing methods, GED and Cosine measure. In obtaining AND versus OR, XOR versus OR, and AND versus XOR, TPED obtained 0.821, 0.811, and 0.78 while Consine-TDP obtained 0.834, 0.826, and 0.693. Meanwhile, GED obtained 1 for all parallel relationships whereas Cosine measure obtained 0.02, 0.08, and 0.04.

259

Authors: Aziz Fajar, Riyanarto Sarno

Port Container Terminal (PCT) involves multi organizations which are Customer, PCT, Custom, and Quarantine. The business processes of those organizations interact asynchronously from container arrivals to the release of the containers. An organization can be handled by multi agents, which represent the actors involved to perform a sequence of activities (tasks) in the organization. Asynchronous Waiting Time (AWT) occurs when an agent assigns a task to another agent which is still working on another task. The previous study discovered that AWT contributes a significant amount of time. Therefore, this research proposes a method to reduce the AWT by parallelizing the agents of an organization and simulating the parallelized agents using agent based simulation. The simulation results time and cost are then optimized using three methods namely Stochastic Multi-Criteria Adaptability Analysis-2 (SMAA-2), Multi Objective Optimization on the Basis of Ratio Analysis (MOORA), and Complex Proportional Assessment (COPRAS). The three methods achieve the same reduced AWT to 9.4%. From those methods, MOORA achieves the highest accuracy of 80% and the sensitivity coefficient of 7. However, COPRAS results in 78% accuracy with lower sensitivity coefficient of 6. SMAA-2 results in the lowest accuracy of 40% and the highest sensitivity coefficient of 13.

12

Authors: Nagesh Byragondlu Komarappa, Bangalore Rangappa LakshmiKantha

Energy Storage System (ESS) is an essential component in the Electric Vehicles (EVs) because of its high power and high energy density. In Hybrid Electric Vehicles (HEVs), the Hybrid ESS (HESS) is introduced to resolve the problem occurs due to single energy storage system. This HESS is utilized for achieving high power and energy simultaneously. In this paper, the HESS is used in HEVs to improve energy management over the vehicles. The HESS is the combination of battery and Ultra-Capacitor (UC). Here the Nickel-Metal Hydride (NiMH) battery is used as a primary energy source and UC is used as a secondary energy source. The energy management over the HEVs made by an effective Artificial Neural Network (ANN) based controller which is utilized for handling the charging and discharging operation of the ultra-capacitor. The operating mode of UC using ANN is mainly based on the State of Charge (SOC) from the battery at each iteration. The UC is on, when the SOC of the battery is less than 60%, it shows that the battery does not have the capacity to achieve the load requirement of Permanent Magnet Synchronous Motor (PMSM). The performance of the proposed system analysed in terms of speed, torque and current. Then it evaluated by using the system with ANN and without ANN. The motor speed and vehicle speed of the HEV with ANN at 70 Sec improved 8% and 9.38%, respectively compared to the HEV without ANN.

33

Authors: Gutta Srinivas Rao, Atluri Srikrishna, Nagesh Vadaparthi

In the field of digital image processing, Image Contrast Enhancement (ICE) is an emerging research area, which helps to attain the sub-standard quality images. The existing ICE techniques have a few limitations like brightness preservation and contrast limitation. Also, the existing ICE techniques affects from over-enhancement that leads to unrealistic results and halo-like artifacts in the high contrast region. To address these issues, a new contrast enhancement method (Second Order Derivative with Weber’s Law (SODWL)) is developed in this research study. The proposed SODWL method includes two major phases (i) estimate the relationship between different image exposures and identify the appropriate Camera Response Model (CRM), (ii) illumination estimation method is applied to calculate the exposure ratio map. The SODWL method improved the performance of ICE effectively, because the developed approach considers even the small edge pixel values for contrast enhancement. After that, Weber’s law is applied for adjusting the lighting conditions of the input image. The experimental analysis was conducted on the reputed datasets: VV, LIME-data, DICM, and MEF. Finally, the performance of proposed method is evaluated by applying the evaluation metrics such as, Lightness Order Error (LOE), and Visual Information Fidelity (VIF). The experimental result showed that the proposed system improved the VIF value upto 0.4-0.1 and also reduced the LOE value up to 2447-2 as related to the existing systems.

53

Authors: Sudheesh Kannur Vasudeva Rao, Basavaraj Lingappa

Indefinite and uncontrollable growth of cells leads to tumours in the human brain. A proper treatment and early diagnosis of brain tumours are essential to prevent permanent damage to the brain. In medical image diagnosis, the tumour segmentation and classification schemes are used for identifying the tumour and non-tumour cells in the brain. The automatic classification is a challenging task which utilizes the traditional methods due to its more execution time and ineffective decision making. To overcome this problem, this research proposes an automatic tumour classification method named as Hybrid Kernel based Fuzzy C-Means clustering - Convolutional Neural Network (Hybrid KFCM-CNN) method. The performance of Hybrid KFCM- CNN method is validated using T1-Weighted Contrast Enhanced Magnitude Resonance Imaging (T1 –W CEMRI) database. A tumour portion is segmented from the MRI brain image by using Hybrid KFCM method. After the segmentation process, statistical features and super pixels based SIFT, Discrete Cosine Transform (DCT) methods were performed on the segmented image to enhance feature subsets. The best feature values were given to the CNN classifier as an input; it is classified into different types of brain tumours such as Meningioma, Glioma and Pituitary. The experimental outcome shows that the Hybrid KFCMCNN method improved the accuracy of brain tumour classification up to 14.06 % than existing classifiers: SVM and CNN.

74

Authors: Hajar El Hanhach, Mohammed Benkhalifa

Sentiment polarity classification (either explicit or hidden) is the process by which information can be extracted to be analysed as positive or negative opinion. Much work on supervised machine learning based sentiment classification has been done considering balanced datasets. However, due to the imbalanced nature of data distribution, sentiment classification becomes a complex task that requires investigating more efficient approach, especially for hidden sentiment. Multinomial Naïve Bayes (MNB) algorithm is one of the most widely used methods in this field, due to its computational efficiency and relatively good predictive performance. However, this classifier performs poorly on imbalanced datasets. In this paper, we propose transformation in MNB prior and conditional probabilities designed to handle explicit and hidden sentiment classification for imbalanced data. To support MNB, we introduce an original frequency model based on using synonym and antonym semantic relations. Our approach is empirically evaluated according to: (1) its impact on MNB training data quality for minority classes, (2) its comparison to similar works involving MNB for imbalanced datasets and (3) its comparison to commonly used classifiers for sentiment classification. The experimental results show that our model improves training data quality and therefore helps MNB boost its performance to achieve the best results for explicit and hidden imbalanced datasets.

95

Authors: Chaidir Chalaf Islamy, Tohari Ahmad

The transmitted information on the internet are prone to security concerns. To overcome this issue, some security methods have been developed in the last few decades. One of the reliable techniques to secure data is called data hiding. This method can be used in various multimedia data, like image, video, text, and audio. Histogram shifting (HS) is a reliable data hiding method used in digital images. It can preserve the similarity of the stego and original image; although, the amount of the protected data depends on the frequency of the most amount pixel in the original image. To solve this problem, we combine HS with prediction error (PE). By using the histogram generated by PE, the size of data that can be embedded increases. In the embedding phase, we do not shift the histogram, different from the existing methods. It is to maintain its original form as high as possible. Moreover, each PE value can contain more data. Based on the experimental result, the average increase in PSNR is 24.8075 dB for general images and 21.5625dB for medical images; while the average improvement of SSIM is 0.026823 and 0.022863 for general and medical images, respectively.

116

Authors: Ricardus Anggi Pramunendar, Sunu Wibirama, Paulus Insap Santosa, Pulung Nurtantio Andono, Moch Arief Soeleman

From literature reviews, the marine environment influences the quality of underwater images and makes the identification of fish species more complex and challenging. The images of the marine environment have low image quality that causes the generated features to be reduced; therefore, this decreases the performance of the classification method. To the best knowledge of the authors, we found out that many researchers have focussed only on determining identification methods without considering the quality of the original data. Therefore, the impact of image enhancement toward the accuracy is yet to be known because this has not been studied comprehensively. To deal with this research gap we propose a new workflow of fish species identification. The workflow for our proposed approach is by using the gray-level co-occurrence matrix (GLCM) feature extraction fed into the back-propagation neural network (BPNN) with contrast-adaptive color correction technique (NCACC) as image enhancements. The experiments demonstrated an improvement in accuracy and kappa measurements for fish species identification from 4.68% to 93.73% and improve from 0.05 to 0.92 respectively. Therefore, our proposed method has the potential to support automatic fish identification systems based on computer vision technology.

139

Authors: Gayathri Selvaraj, Punithavalli Muthuswamy, Chaitanya Vasanth Kumar

An Alcohol Use Disorder (AUD) can be easily predicted by using General Linear Model (GLM) which evaluates the interaction of alcohol expectancy and impulsivity associated with the reduced Gray Matter Volume (GMV) of the right posterior insula in women and the left thalamus in both women and men. However, this analysis does not support the detection of non-linear distribution of GMV at the human brain. Therefore, in this paper, detection of GMV reduction of a brain also known as Alcoholism Detection (AD) is improved by using a Fuzzy CRegression (FCR) model that supports the non-linearity of gray matter in the brain. In this method, spatial information is also incorporated into the FCR model to remove the noise in the brain images. Moreover, the Alcohol Expectancy Test (AET) score is estimated based on the deviations of GMV that mediates the correlation between GMV and alcoholism. This AET score is mostly related to the reduced GMV of the left thalamus in women and men combined and in men alone. Thus, the GMV reduction in the right posterior insula and left thalamus indicates the relationship between AET score and alcoholism. Finally, the experimental outcomes demonstrate that the proposed FCRAD method achieves an accuracy of 86.06% which is 8.77% higher than the existing GLMAD method.

161

Authors: Rubin Thottupurathu Jose, Sojan Lal Poulose

Presently, the distributed Resource Description Framework (RDF) partition the data across several computer nodes. In that, many existing RDF systems results in expensive query evaluation and high start-up cost. To address these issues, a new optimization algorithm: modified Grey Wolf Optimization (GWO) has been developed in this research paper. In conventional GWO algorithm, after finding the best values of

179

Authors: Chaisit Wannoi, Narumon Wannoi, Churirut Boonkhun, Chivalai Temiyasathit, Chai Chompoo-inwai

This paper presents a novel transmission line’s Availability Transfer Capability (ATC) assessment method by considering the Generation Dispatch Limit (GDL) and renewable generation capacity. This method has calculated the power remaining from the Total Transfer Capability (TTC). The TTC in this study is the total generation available declaration of all maximum Generation Shift Factor (GSF) generations to transfer power on transmission line depended on the Power Purchase Agreement of each power generation and of each renewable generation integrated in the system. The maximum GSF generations are defined by using the power sensitivity analysis. Results of this novel method are compared with the traditional transmission line’s thermal limit method. Main findings showed that newly proposed method for ATC calculation can be used for the ATC assessment of transmission line and can be a good predictive index to protect power system’s voltage stability issue. Moreover, it can be applied to the congestion management scheme with the optimal primary respond generation to control the stability of power transfer on the transmission line.

202

Authors: Maryamah Maryamah, Agus Zainal Arifin, Riyanarto Sarno, Yasuhiko Morimoto

Document searching using queries that can understand the context can affect the intent and purpose of the user's desire when searching documents. Many studies have been conducted on understanding the context of the query, but differences in terms of language can lead to different methods of context understanding; therefore, methods implemented in the previous studies need to be improved. In this paper, we proposed a query expansion method based on BabelNet search and Word Embedding (BabelNet Embedding). Query expansion method focuses on developing queries based on semantic relationships on queries to understand the context of the query. Candidate queries were developed by finding synonyms, measuring similarity using WordNet, Word Embedding on all articles of Wikipedia, and BableNet Embedding on articles Wikipedia Online. We compared our proposed method with the existing semantic query expansion. Our result provided better result in retrieving relevant document with accuracy of 89% in searching Arabic documents.

224

Authors: Nam-Soo Kim

We propose an overlay cooperative cognitive radio (CR) non-orthogonal multiple access (NOMA) network which has channel estimation errors and imperfect successive interference cancellation (SIC). To further performance enhancement, spatial diversity gains can be offered by relay selection which is chosen based on channel state information (CSI) among the candidate secondary transmitters. We derive an analytical expression of the outage probability of the primary receiver (PR) and the secondary receiver (SR) as a function of the channel errors and imperfect SIC, respectively. And the analytical outage probability expressions are validated through simulations. Numerical results show that when the number of relays (N) is fixed, the performance of the PR degrades as the increase of the channel estimation errors. In the case of N=2, to satisfy the required outage probability of 1×10-4, 8.2 dB of the transmit SNR degradation is noticed with correlation coefficient ρ=0.98 in comparison with ρ=1. Moreover, the SNR degradation caused by the less correlation coefficient decreases as the number of relays increases. On the other hand, as further imperfect SIC, the error floor that the performance enhancement does not expect with the increase of the transmit power is noticed. The optimum power allocation which maintains the minimum outage probability of SR is realigned by the error floor.

241

Authors: Soma Sekhar Rao Kolusu, Ajay Babu Modala, Boddapati Taraka Phani Madhav, Badugu Prudhvi Nadh

The article presents the design of the concentric circular ring antenna is designed for wireless application like WLAN, Bluetooth, ISM and WiMAX applications. The antenna provides the triple band characteristics frequencies ranging from2.2-3.6 GHz with impedance bandwidth of 48% at the first band,4.3-4.6 GHz with impedance bandwidth of 9% at second operating frequency and third band ranging from 5.2-5.7 GHz with impedance bandwidth of 10% is observed. The antenna provides the bidirectional and omni directional radiation patterns. The antenna provides the peak gain of 2.05 at 2.9 GHz,3.9dBi at 4.3 GHz and 3.5 dBi at 5.4 GHz at three resonating frequency. The radiation efficiency at the three resonating bands is 96% at 2.05 GHz,93% at 4.3GHz and 86% at 5.4 GHz. The frequency reconfigurable properties are observed.

257

Authors: Viyyapu Lokeshwari Vinya, Gurrala Venkateswara Rao

Cognitive Radio (CR) is an emerging technology to solve the issue of scarce spectrum resource utilization. The routing plays a significant role in CR Ad-hoc network and it has two major problems, those are spectrum load balance and energy efficiency. Hence, maintaining the spectrum load balance and energy efficiency becomes a difficult task in CR Ad-hoc Network. In this research, a new Multicast Ad-hoc On-Demand Distance Vector (MAODV) with Particle Swarm Optimization (PSO) algorithm was introduced for selecting best route of less energy consumption in the CR networks. The routing algorithm is implemented with an efficient particle encoding scheme and multi-objective fitness function. Moreover, the data packets are secured by using the RSA algorithm. The optimized lifetime attained from the desired CR ad-hoc networks by using this RSA technique. The multiple scenarios were simulated to compute and analyse the performances in terms of energy consumption, Packet Delivery Ratio (PDR), end to end delay, and throughput. The proposed method named as MAODV-PSO-RSA, which implemented in Network Simulator-2 (NS2). The main objective of the MAODV-PSO-RSA method is to improve the energy consumption during the routing process. The simulation results showed that MAODV-PSO-RSA method had improved 1- 2.2 % of network performance compared to the existing methods such as Bio-Inspired Routing Protocol (BIRP) and Improved Frog Leap Inspired Protocol (IFLIP).

278

Authors: Prasanth Kumar, Gera Pradeepini, Pille Kamakshi

In recent years, a number of researchers have concentrated on medical data analytics because machine intelligence in medical diagnosis is a new trend for enormous medical applications. Generally, medical datasets are massive in size, so traditional classifiers suffered from overfitting and under-fitting problem of training set. In this paper, Gradient Descent Logistic Regression (GDLR) classification method is proposed for medical data classification. The Pearson Correlation Coefficient (PCC) is used to calculate the correlation between the features. After that, Random Forest (RF) algorithm ranks the features and selects the most relevant features to improve performance of the medical data classification. The regression technique processes the features effective and analyse the feature importance based on the weight values. The Random Forest (RF) assigns the features importance in the tree structure. The random forest is used to select the features and features are applied for the GDLR to classify effectively. The GDLR method further analysis the features for effectively analysis the feature importance based on the weight values and more relevant features are identified than the RF. The experimental analysis demonstrated that the performance of GDLR algorithm achieved better than traditional methods Neural Network for Threshold Selection (NNTS) and Mean Selection (MS). The accuracy of the proposed GDLR method achieved as 97.5% in the Hepatitis dataset, while existing mean selection method has the accuracy of 82.58%.

299

Authors: Eman Karam Elsayed, Mohammed Aly

One of the main risks to food security is the plant diseases, but because of the absence of needed infrastructure and actual noise, scientists are faced with a difficult in detection plant diseases in the real image without de-noisy process. The proposed solution in this paper is based on Ontology to support semantic segmentation. Where, the semantic segmentation divides images into non-overlapped regions, with specified semantic labels allocated. The QPSO (quantum particle swarm optimization) algorithm has been used in segmentation of an original noisy image. The proposed method outperforms state-of-the-art algorithms in terms of global accuracy. The proposed method achieves 88% which is superior to other approaches, likely because of the consistency of the semantic implementation, weighting of features and the implementation of data and knowledge required. Our results show that a classification based on the proposed method is better than the state-of-the-art algorithms. The proposed method is evaluated, with 49,563 images from healthy and diseased plant leaves, 12 plant species were identified and 22 diseases. The classification accuracy of the proposed method is 86.2%, showing that the strategy is appropriate. We enhanced PDO (Plant Disease Ontology) to be EPDO (Enhance Plant Disease Ontology). The segmented noisy image elements are paired with EPDO with derived features that come from QPSO. The proposed method also saves time of de-noisy process and effort for removing the noise at a noise level from the input image σ =70.

10

Authors: Shailaja Acholli, Krishnamurthy Gorappa Ningappa

The security and confidentiality of the information has become a significant factor in the communication field. Several techniques for encryption and decryption are proposed to promote the security of the communication systems. Most powerful and significant part of the encryption is a key generation. Presently, hackers are able to break the key with help of the modern high computing machines. In this research, hybrid cryptographic algorithm - Extended Tiny Encryption Algorithm (XTEA) combined with International Data Encryption Algorithm (IDEA) was implemented for improving the security in real-time applications. A large key size ensures the randomness, but proportionally maximizes the network load with high complexity. To overcome this problem, the random numbers were generated using a Linear Feedback Shift Register (LFSR) scheme for key function. This algorithm was appropriate for encryption and decryption of online streaming data. The proposed method is named as International Data - Extended Tiny based Encryption Algorithm –LFSR (ID-XT-EA-LFSR) method. The ID-XT-EA-LFSR method improves the FPGA performances up to 50.43% compared to the QTL, DROM-CSLA-QTL and XTEA algorithms.

31

Authors: Kelly Rossa Sungkono, Ulva Erida Nur Rochmah, Riyanarto Sarno

Process discovery obtains a process model of activity records. There are two representations of process model, i.e. a probabilistic model and a deterministic model. A deterministic model takes all of activity records to depict a process model, however, the probabilistic model chooses several activity records that satisfy a threshold. Determination of the right threshold leads the emergence of many discovery algorithms of probabilistic models, such as Heuristic Miner, Fodina, Modified Heuristic Miner, and Modified Time-Based Heuristic Miners. Those algorithms determine a threshold based on users or an average of probabilities of activity records, so the quality of the model depends on user proficiency or frequent activities. This paper proposes a new algorithm of probabilistic model discovery, i.e. Heuristic Linear Temporal Logic (HLTL), which determines the threshold based on four quality aspects, i.e. Fitness, Precision, Generalization, and Simplicity. HLTL utilizes Linear Temporal Logic to create a formal representation of process model and store the weight of relationships used for the threshold formation. The result shows that the process model constructing by HLTL has better quality aspects than the process model constructing by Modified Heuristic Miners and Modified Time-Based Heuristic Miners. The generalization value of HLTL is 0.8422 and the generalization value of Modified Heuristic Miner and Modified Time-Based Heuristic Miners are 0.8421.

50

Authors: Sabah Sayed, Mohammad Nassef, Amr Badr, Ibrahim Farag

The challenge of processing the Microarray datasets with its high dimensionality opened multiple research directions. Different feature selection techniques have been employed to reduce the dimensionality of such Microarray datasets before being attempted by classification algorithms. This study presents an ensemble feature selection approach based on t-test and Genetic Algorithm with five different classification algorithms as its fitness function: Support Vector Machine, Random Forest, Nearest Centroid, K Nearest Neighbour, and Maximum Likelihood with 5-fold cross validation. The proposed approach has been applied on two different datasets for Lung cancer; Microarray Gene Expression and DNA methylation datasets aiming to find the Lung cancer biomarker genes. The experimental results showed that the three genes (DLX5, KRT5, and SELENBP1) resulted from processing both datasets have higher classification accuracy (92.31%) compared to separately processing the Gene Expression and the DNA methylation datasets with accuracies 90.38% and 86.54% respectively. Moreover, the classification accuracy achieved using the three aforementioned genes could not be achieved by other research studies unless by using more genes.

71

Authors: Deepti Sehrawat, Nasib Singh Gill

Lightweight security algorithms are tailored for resource-constrained environment. To improve the efficiency of an algorithm, usually, a tradeoff is involved in lightweight cryptography in terms of its memory requirements and speed. By adopting several performance enhancement techniques, a security framework for IoT enabled applications is presented in this paper. Proposed BRIGHT family of ciphers is comparably better than existing lightweight ciphers and support a range of block and key sizes for constraint environment. It enables users to match their security needs with application requirements by supporting a range of cryptographic solutions. The BRIGHT family of ciphers is a software-oriented design. The performance of BRIGHT family of lightweight ciphers is evaluated on different parameters. All versions of BRIGHT family ciphers fulfill Strict Avalanche Criteria, key sensitivity test, and randomness test. BRIGHT family ciphers show better performance in terms of memory requirements, cost and speed as compared to existing lightweight ciphers.

90

Authors: Bommalingaiah Nanjappa Mohan Kumar, Rangaraju Hulivangala Gangappa

Finite Impulse Response (FIR) filter is a major building block in Digital Signal Processing (DSP) system. The conventional FIR filters are implemented by utilizing a normal adder, which consumes more area. Minimizing the hardware utilization of the adder is a significant challenge in FIR filter design. Generally, complexity of the FIR filter is dominated by the adder. So, an efficient FIR filter was designed with the help of Carry Select Adder (CSLA) to reduce area and hardware complexity of the accumulation block. The area has been reduced by using CSLA that is very significant idea while designing a less area filter. The proposed method named as Low Area- CSLA-FIR (LA-CSLA-FIR) filter. The LA-CSLA-FIR filter was implemented in Xilinx Field Programmable Gate Array (FPGA) on different devices such as Virtex-4, Virtex-5 and Virtex-6 by using Verilog code. The experimental results of the LA-CSLA-FIR filter design reduced average the FPGA device utilization: 15.38 % of LUTs, 8% of flip flops and 8% of slices on Virtex-5 compared to existing filter designs.

111

Authors: Sara Diouani, Hicham Medromi

Given the development of the Cloud Computing recently, clients and customers using the Cloud for both individual and business needs have expanded to an uncommon scale. This has normally prompted the expanded deployments of Cloud data centers over the globe. As a result, Cloud data centers are seen to be monstrous energy consumers and natural polluters. They require an extraordinary measure of regular energy which has made an effect on the energy supply and natural conditions of the environment. This is the reason why the vulnerability of persistent energy supply, later on, is being referred to. In this way, there is a need of an energy-aware cloud-based system which automatically and efficiently manages and optimize cloud computing data center resources by considering energy consumption as an essential Quality of Service (QoS) parameter. This paper, focus on the energy utilization of the data centers and how this can be limited so as to make the cloud computing greener. Thus, a new autonomic resource optimization manager has been proposed to avail the most optimum level of resources with reduced server energy consumption. The proposed framework has been verified theoretically and tested experimentally. The experimental analysis has demonstrated that the effectiveness of the proposed solution is greater than the state-of-the-art methods in terms of the achieved results related to reducing energy consumption and response time in cloud computing data centers.

142

Authors: Reza Amalia Priyantina, Riyanarto Sarno

Product reviews are usually determined by sentiment of customers; however sentiment analysis based on aspects still need further research. A hotel commonly has five aspects, which are location, meal, service, comfort and cleanliness. This research proposes methods to determine review sentiment according to the hotel aspects. A hotel reviews are preprocessed into a term list. Firstly, Latent Dirichelet Allocation (LDA) determines the hidden topics of a term list; then Semantic Similarity categorizes the term list based on the topic resulted by Latent Dirichelet Allocation (LDA) into the five aspects of a hotel. Then in calculating similarity measurement, the term list is expanded by using the Term Frequency-Inverse Cluster Frequency (TF-ICF) method. Finally, a classification of customer sentiment (satisfied or dissatisfied) is conducted by using the combination of Word Embedding and Long-short Term Memory (LSTM). The results show that the proposed method can classify the reviews into the five hotel aspects. The highest aspect categorization performance is obtained by using LDA + TF-ICF 100% + Semantic Similarity which reaches 85%; the performance sentiment classification for the highest aspect-based sentiment analysis is obtained by using Word Embedding + LSTM which reaches 93%; and the comfort aspect receives more negative sentiments compared to the sentiments of other aspects. Also the results show that a sentiment is influenced by an aspect.

169

Authors: Youness Abouqora, Omar Herouane, Lahcen Moumoun, Taoufiq Gadi

The interest point detection is a technique for reducing the amount of data that needs to be processed in certain applications, this technique is widely used in 2D and 3D image analysis. In this paper, we introduce a new method for detecting the mesh interest points based on geometric measures. The key point of this new approach is to extract a satisfactory 3D interest points with a high repeatability and relevant distribution of the whole object. It consists of two steps: an interest point detection and a refinement post-processing. First, we compute the Harris operator function, that is robust to several transformations, around each vertex to selects the maximal responses as candidates of 3D interest points. Second, we use the shape index-based optimization method to select the final interest points with high curvedness. The performance of the proposed method is evaluated and it proves encouraging results.

192

Authors: Md. Yusuf Mulge

Task Scheduling (TS) is used to improve the performance of Cloud Computing (CC) and optimizing the Quality of Service (QoS) of the cloud. The main objective of this research work is to optimize the TS by minimizing the MakeSpan (MS) and energy consumption. The scheduling issues are overcome by Modified Mean Grey Wolf Optimization algorithm (MGWO) and it enhances the overall performance of the system. In this MGWO, the leadership hierarchy is simulated by four grey wolves such as alpha, beta, delta and omega for the major steps like hunting, searching, encircling and attacking the prey. Here, hunting and encircling modified using the mean value to improve the effectiveness of GWO. CloudSim toolkit is used for evaluating the objective of the Modified-MGWO in standard workload (left-skewed & right-skewed dataset). The proposed Modified-MGWO algorithm reduced energy consumption nearly 10% when compared with the existing algorithms such as Particle Swarm Optimization (PSO) and standard GWO. Moreover, the MS of the Modified-MGWO method reduced nearly 13% when compared with existing methods for both datasets.

212

Authors: Mariya Ouaissa, Mariyam Ouaissa, Abdallah Rhattoy

The Internet of Things (IoT) is a new concept that is developed considerably in current wireless communications. The main idea of this paradigm is the exchange of information and data from real-world devices with the internet by using unique addressing systems can communicate and interact with each other to achieve common objectives. With the emergence of next generation battery-powered of smart mobile phone, open source application platforms, security and the terminal’s energy consumption have become big issues. Long Term Evolution (LTE) is one of the most popular 4th Generation (4G) technologies defined by 3rd Generation Partnership Projects (3GPP) also IP Multimedia Subsystem (IMS) is a prominent architectural framework for multimedia services delivery in 4G/5G networks. In this paper, we focus on IoT security in particular, we explore authentication that is a critical security mechanism which accords authorized users access to a network. It is observed that the end user requires two authentication steps to access multimedia services. The first is the LTE network layer authentication, and the second is the IMS service layer authentication. The authentication steps utilize energy and are carried out using the Authentication and Key Agreement (AKA) protocol. As defined by 3GPP, IMS- Authentication and Key Agreement protocol (IMS-AKA) is the official authentication procedure in IMS and the standard EPS-Authentication and Key Agreement protocol (EPS-AKA) is proposed to provide the authentication service between the device and the network in Evolved Packet System (EPS) system. However, the procedures are prone to different weaknesses both on security and performances aspects. This paper proposes a secure and efficient AKA protocol that authenticates the user on both the network layer and the service layer without double execution the AKA protocol, simplifies the authentication steps, protects user’s identities and reduces the authentication process complexity due to the use of Elliptic Curve Cryptography (ECC). The proposed solution was checked by the security protocol verification tool, Automated Validation of Internet Security Protocols and Applications (AVISPA), which indicated that it is a very secure level. Simulation results showed that our proposed compared to IMS-AKA and EPS-AKA offers better performances in terms of energy consumption and storage cost.

233

Authors: Mohammad Nasrul Aziz, Tohari Ahmad

Various machine learning technology approaches have been applied to intrusion detection system (IDS). To get optimal results, it needs to take several stages for processing the traffics. Among them is the feature selection method, where irrelevant and redundant features are removed. In the previous research, the system is developed based on feature grouping that used a clustering approach as the evaluation criteria. In this research, we propose a method for improving the performance of machine learning with the feature selection approach based on feature clustering. We propose cluster based feature selection derived from the value of mutual information and Pearson correlation. The cluster hierarchy is used in forming filters that are used to create selected and reduced clusters. In developing the cluster hierarchy, single, complete, and average linkage method are used to determine the formation of the best feature clusters. The classification method with Support Vector Machine (SVM), Naïve Bayes, and J48 decision tree are applied to observe the performance of the proposed feature selection. Based on the experimental results, we find that the highest accuracy (i.e., 99.842%) is obtained when a single linkage in the J48 classification is implemented in the Kyoto 2006 dataset.

255

Authors: Varinder Kaur Attri, Jatinder Singh Bal

Software Size Estimation is the most essential and crucial calculations of the Software Development Life Cycle (SDLC) process. If done wisely, it may accumulate a large amount and time and if done badly, it may cost a lot of amounts. In this modern era of development, traditional methods of estimation do not stand a chance to provide output precisely. Constructive Cost Model (COCOMO)-2 is one of finest calculation methods for size and cost. This paper presents an automated early size estimation technique using Artificial Intelligence (AI). The categorization of the size estimation is into three parts, that is, UML diagrams/code folder mapping via COCOMO-2, COCOMO-2 metrics training, and the classification process to have an appropriate size. This paper focused on regression based training of AI which makes the estimation model more precise. For the evaluation, Mean Square Error (MSE) and Size estimation have been considered according to project samples. It has been seen that the proposed mechanism has attained a minimum MSE of 0.0115 and the difference in sizes is also not high. The comparison has also been done to depict the efficacy of the proposed work with A. B. Nassif et al. and S. Lohmor with B.B. Sagar for True positive rate and Mean square error.

275

Authors: Dewi Ayu Khusnul Khotimah, Riyanarto Sarno

In the industrial era 5.0, product reviews are necessary for the sustainability of a company. Product reviews are a User Generated Content (UGC) feature which describes customer satisfaction. The researcher used five hotel aspects including location, meal, service, comfort, and cleanliness to measure customer satisfaction. Each product review was preprocessed into a term list document. In this context, we proposed the Probabilistic Latent Semantic Analysis (PLSA) method to produce a hidden topic. Semantic Similarity was used to classify topics into five hotel aspects. The Term Frequency-Inverse Corpus Frequency (TF-ICF) method was used for weighting each term list, which had been expanded from each cluster in the document. The researcher used Word embedding to obtain vector values in the deep learning method from Long Short-Term Memory (LSTM) for sentiment classification. The result showed that the combination of the PLSA + TF ICF 100% + Semantic Similarity method was superior are 0.840 in the fifth categorization of the hotel aspects; the Word Embedding + LSTM method outperformed the sentiment classification at value 0.946; the service aspect received positive sentiment value higher are 45.545 than the other aspects; the comfort aspect received negative sentiment value higher are 12.871 than the other aspects. Other results also showed that sentiment was affected by the aspects.

306

Authors: Ahmad Ilham, Romi Satria Wahono, Catur Supriyanto, Adi Wijaya

Determining the cluster number in k-means is problematic since it affects the quality of cluster for numerous applications in the data mining. The automatic clustering differential evolution (ACDE) is one of the most used clustering methods that are able to determine the cluster number automatically. However, ACDE still makes use of the manual strategy to determine a value k activation threshold thereby affecting its performance. In this study, the u-control chart (UCC) method use to tackle the ACDE method problem. UCC method used for the initial step to get a value of the variables sought before initialization of the variable vector on ACDE. The UCC is a method from statistical process control (SPC) field which has proved to be effective in solving the problem of management control attributes. The performance of the proposed method was tested using seven public datasets from the UCI repository and Clustering basic benchmark repository and evaluated with Davies Bouldin Index (DBI) and CS measure. The results show that, the proposed method yields excellent performance compared to prior researches for most datasets with optimal cluster number yet lowest DBI and CS measure. It can be concluded that the UCC method is able to determine k activation threshold in ACDE that caused effective determination of the cluster number for k-means clustering.

328

Authors: Chellan Edward Jaya Singh, Eppies Baburaj

In this modern world, most of people are using cloud computing due to its accessibility. Users are uploading plenty of images to the cloud every day. But still, data confidentiality is the major issue for the cloud storage. This paper proposes a new method of image encryption scheme with less storage in the cloud. The proposed method performs double encryption with de-duplication over the images for better security and also reduces the storage overhead problem using the scaling and cropping operation. In the first level, the modified Paillier encryption technique is used and an XOR encryption takes place in the second phase. After that, a secure block-level image de-duplication technique is performed to eliminate the identical images and protect the image confidentiality. The simulation result exhibits that the proposed encryption with secure de-duplication outperforms other state-of-art approaches with respect to the performance metrics like accuracy, sensitivity, specificity, encryption quality and runtime. In terms of accuracy, the performance of proposed method is better of 8.1%, 7.2%, and 6.1% for image 1, 10.2%, 7.1%, and 6.1% for image 2, 8.4%, 6.2%, and 5.8% for image 3, and 8.3%, 6.2%, and 5.6% for image 4 when compared with other three existing research works.

348

Authors: Shaik Khaleelahmed, Nandhanavanam Venkateswararao

Enhancing the efficiency of power allocation and the achievable sum rate of Multiple Input Multiple Output (MIMO) based multi-user systems are essential as conventional methods are very complex. The traditional methods are not only less efficient but also do not address user fairness. The existing conventional methods use simple user ordering schemes that remain unsuitable for MIMO Non-Orthogonal Multiple Access (NOMA) system. Accordingly, a new elegant propose method using Particle Swarm Optimization (PSO) algorithm based priority scheduling in power allocation (PPPA) prioritizes the users based on the maximum power and QoS constraints. The proposed method offers a capable platform that will also provide energy efficiency in power allocation. The priority-based scheduling based on PSO algorithm that prioritizes the users optimally based on objective function, and it is effective due to faster convergence and adaptive nature of the algorithm. The proposed method for power allocation compared with the existing methods using the number of admitted users in the system. We observed that the energy efficiency and achievable rate is found to be superior by 6.76% and 58.37 Mbits/sec. Thus our proposed method not only provides better energy efficiency but also has a profound improvement in data rate.

367

Authors: Yassir Rochd, Imad Hafidi

Frequent item exploration is a fundamental element in many data mining problems aimed at finding interesting models in the data. Recently, the PrePost algorithm, a new algorithm for extraction frequent element sets based on the idea of N-lists, which in most cases surpasses other current state-of-the-art algorithms, has been introduced. The PrePost algorithm's performance deteriorates when it comes to handling big data. Nevertheless, the current existing PrePost algorithms in place implemented with the MapReduce model are not sufficiently powerful for iterative computation. To reduce IO overhead and take advantage of cluster memory, this article offers an enhanced version of PrePost, the Distributed PrePost (DisPrePost), a parallel algorithm built on the Spark framework, which incorporates the concept of resilient distributed datasets and performs in-memory processing to optimize the execution time of operation, that also utilises a HashMap to further refine the N-list creation process. Experience has shown that the DisPrePost algorithm is more efficient and scalable than the two advanced state-of-the-art methods HPrePostPlus and the well-known algorithm HFIM.

11

Authors: Tossapol Kiatcharoenpol, Thanakarn Vichiraprasert

Design of Experiment (DOE) has prevailingly used in various industries from its effectiveness of a statistical technique for managing and improving processes. In this work, the three DOE approaches, Classical DOE or factorial design, Taguchi method and Shainin DOE are applied in a case of plastic injection molding process. A comparative analysis is carried out to express the way of implementation and the merit of each approach based on the accuracy of information and the ease of use. The results show that significant factors obtained from the three DOEs are aligned together. For all approaches, the most three factors influencing to the response, %volume shrinkage, are exactly the same including the optimal process parameters chosen. This is one of an evidence of effectiveness for both Taguchi method and Shainin DOE in practice. The adoption of DOE approaches outlined in this paper and their application to an industrial process will be valuable to practitioners to gain understanding of DOE techniques and to use them fit to their quality issue.

31

Authors: Lukman Hakim, Riyanarto Sarno, Kelly R. Sungkono

Process discovery is a concern in analyzing business processes. Process discovery methods handle many aspects, e.g. a hybrid of non-free choice and invisible task. A hybrid of non-free choice and an invisible task is a condition that needs additions of invisible tasks in non-free choice relationships. Process discovery methods, e.g. Alpha++, Alpha#, Graphical method, Time-based Alpha Miner, cannot discover the hybrid of non-free choice and invisible task, especially invisible tasks in the overlapping pattern. To modeling a hybrid of non-free choice and invisible tasks, this research proposes an algorithm of adding invisible tasks, including invisible tasks of overlapping pattern, automatically in the event log and processing the event log with the rules of Alpha++ algorithm. To test the proposed algorithm, namely Invisible Task Log with Alpha++, the experiment compares it with Alpha++, Alpha#, Graphical Method, Time-based Alpha Miner, Hidden Markov Model-Parallel, and Coupled Hidden Markov Model-Invisible Task. The experiment results proved that Invisible Task Log with Alpha++ algorithm obtains higher precision than the obtained precision by comparison methods.

50

Authors: Rokesh Kumar Yarava, Rajendra Prasad Singh

Cloud storage auditing technique helps to verify the data integrity in the cloud on behalf of the user. Earlier researcher proposed many techniques on the cloud auditing, but it still lags in computational overhead for the resource constraint users. In this research, the Diffie-Hellman method is used to exchange the key in the Third Party Auditor (TPA) to improve the performance of the auditing method. The key is generated in the third party auditor and this is shared with the user as an encrypted key. The Diffie-Hellman creates the shared encryption key between the two parties and can be communicate in the insecure channel. The Diffie-Hellman generates ephemeral keys and this is extremely fast in generating new key pairs. This method is suitable for the big server and have more efficiency over the state-of-art method, which doesn’t involve in ephemeral key generation. The proposed method is evaluated in the simulated environment and compared with the other state-of the art methods. The verification cost of the proposed auditing method has achieved 1.1 s, while the existing method has 1.19s.

68

Authors: Archana Kalidindi, Prasanthi Yavanamandha, Anusha Nagina Kunuku

The recommendation system plays a major role in e-commerce services and online applications such as social-networking, product recommendation, etc. The recommendation system uses the most popular techniques namely Collaborative Filtering (CF) and Deep Learning Neural Network (DLNN) approach. The Cold-Start (CS) problem and the recommendation efficiency are considered as crucial challenges and affected the efficiency of the most popular techniques. In this paper, a Discrete Deep Learning (DDL) based on hashing framework is implemented to achieve Hamming space for mapping Users and Items (UIs). Hamming distance used here to calculate the user’s preference for an item, and the efficiency of online recommendation significantly improved by this computation technique. The performance of DDL framework evaluated by conducting various experiments on the large Netflix rating dataset. The experimental outcomes showed that online recommendation efficiency and CS recommendation accuracy of DDL outperformed the existing methods.

88

Authors: Agus Urip Ari Wibowo, Selo Sulistyo, I Wayan Mustika

Probabilistic Routing Protocols using History of Encounters and Transitivity (PRoPHET) assume that nodes do not move randomly but encounter destinations that have been encountered before. In some original PRoPHET and original PRoPHETv2 testing, they are superior to other routing protocols. Testing in environments that use trains as relay nodes, original PRoPHET performance is worse than the epidemic. This happens because messages delivery to other relay nodes are assumed to encounter the destination (having a larger probability of encountering destination) while the other relay nodes are moving away. The original PRoPHET and the original PRoPHETv2 performance can be improved by delivering a message to relay nodes that will encounter the destination. This can be done by delivering a message to relay nodes with lower probability values encountering the destination. The result obtained is a modified PRoPHET and a modified PRoPHETv2 performance improvement compared to the original PRoPHET and the original PRoPHETv2.

108

Authors: Arna Fariza, Agus Zainal Arifin, Eha Renwi Astuti, Takio Kurita

Tooth component segmentation is a crucial task in computer-aided design for forensic odontology, especially to estimate chronological age. Tooth segmentation on radiographic data is a very challenging task due to noise, low contrast, and uneven illumination. The Fuzzy C-Means clustering is generally used for image segmentation that allow pixels to be classified into one or more clusters according to their membership value. However, this clustering method still has problems associated with the shifting of cluster centers and sensitivity to the overlapping intensity distributions between classes. This paper proposes a modified strategy of the conditional spatial Fuzzy C-Means (csFCM) that incorporates global and spatial information into a weighted membership function by replacing the Euclidean distance with the Gaussian kernel distance to increase insensitivity to noise and outliers. The aim of this paper is to divide the tooth into 3 components, i.e. enamel, dentine, and pulp. Therefore, this modified algorithm is preceded by dental X-ray image pre-processing and continued by combining each dental component clusters into one composite image. The tooth image is pre-processed using Contrast Limited Adequate Histogram Equalization (CLAHE) and gamma adjustment to enhance the dental X-ray images quality from the non-uniform lighting. The Gausian kernel-based conditional spatial Fuzzy C-Means (GK-csFCM) segments the dental image into four class clusters, namely enamel, dentine, pulp, and background. Through iterations, the resulting cluster centers are more convergent with real cluster centers, thus ensuring the proposed method improves the drawback of inherent FCM-based methods and further promoting image segmentation performance. The experimental results on the real dental X-ray images showed that GK-csFCM has better performance than the typical FCM and csFCM clustering algorithms in terms of both qualitative and quantitative metrics, i.e. accuracy, specificity, sensitivity, and precision.

129

Authors: Ahmad Saikhu, Agus Zainal Arifin, Chastine Fatichah

Data relating to everyday phenomena can be recorded in the form of time series. Recently, modeling of time-series has become a topic of interest in data mining. Conducting an analysis of multivariate time series effectively is an essential task for decision-making activities in fields such as meteorology, medicine, and finance. Features selection is a key problem in the analysis of multivariate time series. Rainfall prediction, biomedical classification, pattern recognition, sensor network analysis and so on all have different input features. The problem is that these features have interdependencies and time-delay relationships. Currently, research on the selection of input features of these data still depends on whether they are linear or non-linear. In this paper, we propose a new integration strategy between Pearson Correlation and Symmetrical Uncertainty for relevant feature selection based on linear and non-linear relationships for multivariate time-series classification. We evaluated the goodness of fit of feature subsets using merit value. The meteorological data set was used to test the proposed method. The result showed that the method was able to reduce the number of features by 77.9% features and increase their merit value 2.25 times compared to no input features selection.

148

Authors: Gundugonti Kishore Kumar, Narayanam Balaji, Kotha Srinivasa Reddy, Vemu Thanuja

This paper presents a modified 2D Discrete Wavelet Transform (DWT) architecture with a proposed 16-bit Radix8 Booth multiplier. Existing architecture makes use of Canonic Sign Digit (CSD) representation and when replaced the CSD multiplier with the proposed 16-bit Radix8 Booth multiplier it achieves better performance with small area and low power. In proposed Radix-8 Booth multiplier, the necessary product terms are generated and the remaining terms are truncated. In this method, the n order bit required by the specific coefficient is obtained and the remaining n bits are truncated so that 2n bit output truncated to n bit. The modified 2D DWT architecture is proposed to enhance that it occupies less number of clock cycles, so that it improves in the speed of operation By comparing synthesis results for existing CSD multiplier and the proposed Radix-8 Booth multiplier achieves an improvement of nearly 29.02% Area Delay Product (ADP) and 26.13% Power Delay Product (PDP).

166

Authors: Majeti Venkata Sireesha, Panchala Venkata Naganjaneyulu, Kaparapu Babulu

To depict the hyperspectral data, here a sturdy mixing model is implemented by employing various perfect spectral signatures mixture, which enhances the generally utilized linear mixture model (LMM) by inserting an extra term that describes the potential nonlinear effects (NEs), which are addressed as additive nonlinearities (NLs) those are disseminated without dense. Accompanying the traditional nonnegativity and sum-to-one restraints underlying to the spectral mixing, this proposed model heads to a novel pattern of sturdy nonnegative matrix factorization (S-NMF) with a term named group sparse outlier. The factorization is presented as an issue of optimization which is later dealt by an iterative block-coordinated descent algorithm (IB-CDA) regarding the updates with maximation-minimisation. Moreover, distinctive hyperspectral mixture models also presented by adopting the considerations like NEs, mismodelling effects (MEs) and endmember variability (EV). The extensive simulation analysis by the implementation of proposed models with their estimation approaches tested on both the synthetic and real-time images. Further, it is also shown that the comparative analysis with the conventional approaches.

192

Authors: Sanjive Tyagi, Rakesh Kumar Dwivedi, Ashendra K. Saxena

In this paper, a novel approach of PDF-based text steganography is presented. A covert communication is achieved by embedding secret information at between-characters positions of words within cover PDF text. A stego-encoding technique is being designed to improve the embedding capacity along with sufficiently reduced overheads of stego-cover PDF file. The proposed method constructs multilevel embedding capacities which can be used according to requirements from higher embedding capacity to typical embedding capacity. Encoding and decoding of secret text is being obtained by implementing quadratic probing of hashing technique that builds up a direct encoding table which provides better performance in term of time complexity. Exploratory outcome exhibits that the proposed technique gives productive algorithms in term upgraded security and high payload of concealed data. Experiential verification and examination of the proposed technique with certain prominent text steganographic techniques demonstrate that proposed technique performs better than current methodologies and free of requirements like shading scheme, languages, wordlist, compression and so forth.

214

Authors: Mouna Berquedich, Oualid Kamach, Malek Masmoudi

The management of emergency services within hospitals typically requires an effective manipulation and capitalization of data sources comprising worth information useful for hospitals’ security, safety and efficiency. To meet these requirements, we dispose an agile approach of decision-making that addresses massive crowds occurring at the emergency departments. This approach considers the major constraints that might face these departments, as such lacks of human resources, high costs, and confusedness of patient’s prioritization cases, non-sufficient capacity and logistics. We, basically, inspired from the biological immune defence processes to illustrate a piloting emergency mean known as the artificial immune system (AIS). This system intelligently assists hospital’s decision-makers to enhance their supplying strategies, provides relevant traces from tracking previous information that assist hospital’s staffs that face the massive patient flow, and fosters the execution of efficient solutions. This paper investigates AIS data and relevant traces of overcrowding dilemmas, in which these data could be exploited to increase the reception planning, decrease tension-related anomalies within the emergency department, and supplying hospital chiefs working under stress. The main objective is to present the basic concepts of a new approach that aims to provide hospital decision-makers with relevant traces to assist them in their decision-making in the difficult situations. The expected benefits (main findings) for using the considered approach are not only to provide crisis managers with a relevant computerized decision support system, but also to minimize financial costs, reduce the response time and positively impact the crisis management.

244

Authors: Ahmed Abdelaziz, Nagy Ramadan Darwish, Hesham Ahmed Hefny

The management of agile software projects plays an important role in the success or failure of projects. There are many factors of success or failure that clearly affect agile software projects. There is a big problem in not having a formal study to identify critical failure factors that assess the success or failure of projects accurately. This paper focuses on failure factors of agile software projects. This paper seeks to configure a formal study by using multiple regression analysis for identifying critical failure factors to avoid projects failure, thus saving time and money for the stakeholders. The proposed model is implemented by using R programming. The proposed model also is implemented by two models. The first model based on critical failure factors (CFF) of agile software projects. The second model based the failure factors of the proposed in the initial list (FFIL). The model of CFF outperforms on the model of FFIL in order to accuracy by 10.7% and random error ratio by 4.8%. The results revealed that only 10 out of 19 hypotheses were supported, identifying seven critical failure factors for Agile software development projects: (a) Organizational culture too traditional, (b) Lack of agile logistical arrangements,(c) Lack of project management competence, (d) Lack of team work, (e) Bad customer relationship, (f) defined project scope, and (g) defined project planning.

268

Authors: Dipti Yashodhan Sakhare

Nowadays, Motion Estimation (ME) plays a vital role in traffic signals, movie making, and security purpose to identify the motion of an object. In this paper, Smart and Low power (SLP) based ME is introduced with Level Convertor and Carry SeLect Adder (LC-CSLA-ME). Normal SLP architecture requires more power for performing ME, but in the LC-CSLA-ME architecture, 0.9v supply voltage is used to carry-out the ME. The LC is used to supply the required voltage to the respective sub blocks, which helps to reduce the power consumption of the overall architecture. Instead of using normal adder, CSLA adder is used in prediction and classification unit for improving the hardware utilization. The performance of Application Specified Integrated Chips (ASIC) and Field Programmable Gate Array (FPGA) analysed for both existing and proposed methods. In 180nm technology, LC-CSLA-ME architecture occupied 0.065mm2 area, 2.10mW power, and 104ms delay. The different dataset such as Johnny, Vidyo 1, and Vidyo 4 is analysed for ME. The Coding Time Saving (CTS) evaluated for both existing and LC-CSLA-ME architectures. In the LC-CSLA-ME, average CTS of Operating Power Points (OPP1, OPP2, OPP3, OPP4, and OPP5) are 34.021, 54.316, 67.021, 75.027, and 80.071 that is high compared to the existing methods.

292

Authors: Kavikondala Praveen Kumar Rao, Tamilarasan Senthil Murugan

Mobile Ad-hoc Network (MANET) dynamically designed by the self-organization of the mobile nodes connected through wireless links without using any centralized administration. Energy consumption is the most significant issue in the MANETs, because most of the mobile host’s functions have inadequate battery resources. Reduction in the energy consumption increases the lifetime and throughput of the network. The performance of existing techniques is less in terms of energy conservation issues. To overcome this limits, this paper suggested an energy conservation mechanism incorporated with a pro-active MANET routing scheme. The routing schema is employed in the energy level and the movement of the nodes. Calculation of energy cost is accomplished based on the assessment of the energy consumption level of the node, using K-Means clustering along with AODV (Ad-hoc On Demand Distance Vector). The data packets are secured using (AES), while sending data from source to destination. The proposed methodology “K-Means-AODV-ACO-AES” model is compared with the existing Optimal Key Management for Secure Data Transmission (OKMSDT) and Secured AODV (Sec-AODV). From the comparison result, it is observed that the K-Means-AODV-ACO-AES” increases in Packet Delivery Ratio (5%), along with the decrease in end-to-end-delay (6%), and energy consumption (7%) and drop (8%) in secured environment.

319

Authors: G. R. Sakthidharan, P. Chandra Sekhar Reddy

There has been a tremendous growth in research concern in Wireless Sensor Networks (WSNs), due to the broad range of applications in distinctive areas such as environmental observation, security domains, home automation, etc. Despite advantages, the prevalent adoption of WSN, specifically in remote and isolated regions, has become the major challenge, because of this battery constrained network with limited energy source. As advancement, the nodes are framed under Wireless Rechargeable Sensor Network (WRSN) to enhance the lifetime of sensor nodes with the establishment of charging mechanisms. With those considerations, this paper presents a Schedule-based Optimized Node Recharging Model (SONRM), in accordance to the condition of limited capability of mobile charging. In the process of the construction of charging circuit, the SONRM effectively considers the dynamic changes of energy utilization among nodes at different factors and develops a Recharge Scheduling Algorithm (RSA) that helps in finding the next Energy Starving Sensor Node (ESSN) to be recharged. Simultaneously, a model for the establishment of feasible circuit for recharging has also been done for ensuring the effectiveness of charging. During the operation, the node mortality can be reduced by the proposed model with the incorporation of dynamic charging threshold. When compared with the existing mechanisms, the SONRM attains better results on analysing decisive factors such as network longevity, service time, efficient charging, throughput and recharging time.