Computational Intelligence in Data Mining—Volume 2

Mobile robots are now widely used in the daily life. A mobile robot is a one which allows motion in different directions and it can be used as a prototype in certain applications. By controlling the mobile robot using a sensor it can be used as a prototype for wheelchairs and thus they can assist the physically disabled people in their movement. This is very useful for them in their personnel as well as professional life. Thus the mobile robots are entered into the human day-to-day life. The sensor can capture the electrical impulses during the brain activity. And they are converted into commands for the movement of mobile robot. ZigBee is a wireless protocol used for the interaction between the computer and the mobile robot. Brain-computer interface is the communication system that enables the interaction between user and mobile robot. Electroencephalogram signals are used for controlling the mobile robots.

The objective of the proposed paper is to develop an optimal policy of an inventory model that minimizes the total relevant cost per unit time. In this model, a two-warehouse system considers an owned warehouse (OW) with limited storage capacity and a rented warehouse (RW) with unlimited storage capacity. The demand rate is an exponential function of time, the rate of deterioration of OW is more than that of RW and the supplier provides the purchaser a permissible delay of payment. The results have been validated with the help of numerical examples.

Data clustering is one of the major areas of research in data mining. Of late, high dimensionality dataset is becoming popular because of the generation of huge volumes of data. Among the traditional partitional clustering algorithms, the bisecting K-Means is one of the most widely used for high dimensional dataset. But the performance degrades as the dimensionality increases. Also, the task of selection of cluster for further bisection is a challenging one. To overcome these drawbacks, we incorporate two constraints namely, stability-based measure and Mean Square Error (MSE) on the novel partitional clustering method, CHB-K-Means algorithm. In the experimental analysis, the performance is analyzed with respect to computation time and clustering accuracy as the number of outliers detected varies. We infer that an average clustering accuracy of 75 % has been achieved and the computation time taken for cluster formation also decreases as more number of outliers is detected.

Last-level cache is a high level cache memory shared by all core in a multi-core chip to improve the overall efficiency of modern processors. Most of the cache management policies are based on the residency of the cache block, access time, frequency tour, recency and reuse distance to lower down the total miss count. However, in a contemporary multi level cache processor a bypass replacement reduces both miss penalty and processor stall cycles to accelerate the overall performance. This paper has analyzed different bypass replacement approaches. It also gives motivation behind different bypass replacement techniques, used in the last-level cache with inclusive and non-inclusive behavior with improved placement and promotion.

Kinase phosphorylates specific substrates by transferring phosphate from ATP. These are important targets for the treatment of various neurological disorders, drug addiction and cancer. To organize the kinases diversity and to compare distantly related sequences it is important to classify kinases with high precision. In this study we made an attempt to classify kinases using four different classification algorithms with three different physiochemical features. Our results suggest that Random Forest gives an average precision of 0.99 for classification of kinases; and when amphiphilic pseudo amino acid composition was used as feature, the precision of the classifier was much higher than compared to amino acid composition and dipeptide composition. Hence, Random forest with amphiphilic pseudo amino acid composition is the best combination to achieve classification of kinases with high precision. Further the same can be extended for subfamilies, which can give more insight into the predominant features specific to kinases subfamilies.

Wireless Sensor Networks (WSNs) have a wide range of applications in the real world. They consist of a large number of small, inexpensive, limited energy and low-cost sensor nodes which are deployed in vast geographical areas for remote sensing and monitoring operations. The sensor nodes are mostly deployed in harsh environments and unattended setups. They are prone to failure due to battery depletion, low-cost design or malfunctioning of some components. The paper proposes a two-level fuzzy knowledge based sensor node appraisal technique (NAT) in which the cluster head (CH) assesses the health status of each non-cluster head (NCH) node by the application of fuzzy rules and challenge-response technique. The CH then aggregates data from only the healthy NCHs and forwards it to the base station. It is a pro-active approach which prevents faulty data from reaching the base station. The simulation is carried out with injected NCH faults at a specified rate. The simulation results show that the proposed NAT technique can significantly improve the throughput, network lifetime and quality of service (QoS) provided by WSNs.

This paper presents the results from a research to design and develop a bank cheque recognition system for Indian banks. Geometrical features of the bank cheque is the one which is the outcome of the type or category of the cheque which belongs to a specific bank. In this research, an attempt is made to design a model to classify Indian bank cheques according to the features extracted scanned cheque images by applying classification level decision using Feed forward artificial Neural Network (NN). The proposed paper contains a feed forward propagation Neural networks system designed for classification of bank cheque images. Six groups of bank cheque images including SBI, Canara, Axis, Corporation, SBM and Union bank cheques are used in the classification system. The accuracy of the system is analyzed by the variation on the range of the cheque image with different orientation and locations of bank logo and trained. The efficiency of the system is demonstrated through the experimental results extensively.

The key principles involved in abstraction, encapsulation, design and development phases of the software structures of a system is, management of their complexities. These structures consist of the necessary components defining the concept of architecture of the system. Complex embedded systems, evolving with time, comprise of complex software and hardware units for its execution. This requires effective and efficient software model-based engineering practices. The complex systems are evolving in-terms of its resources and contemplating the operational dynamics. In this paper we emphasize on the formal foundations of Architecture Analysis and Design Language (AADL) for model-based engineering practices. This engineering process involves models as the centralized and the indispensible artifacts in a product’s development life-cycle. The outcome of the approach features the techniques along with the core capabilities of AADL and managing the evolving resources considering impact analysis. A suitable case study, Power Boat Autopilot is considered. The details on the use of AADL capabilities for architectural modeling and analysis are briefly presented in this paper.

Intrusion Detection Systems (IDS) are intended to protect computer networks from malicious users. Several data mining techniques have been used to build intrusion detection models for analyzing anomalous behavior of network users. However, the performance of such techniques largely depends on their ability to analyze intrusion data and raise alarm whenever suspicious activities are observed. In this paper, some rule based classification techniques, viz., Decision Table, DTNB, NNGE, JRip, and RIDOR have been applied to build intrusion detection models. Further, in order to improve the performance of the classifiers, six rank based feature selection methods, viz., Chi squared attribute evaluator, One-R, Relief-F, information Gain, Gain Ratio, and Symmetrical Uncertainty have been employed to select the most relevant features. Performance of different combinations of classifiers and feature selection techniques have been studied using a set of performance criteria, viz., accuracy, precision, detection rate, false alarm rate, and efficiency.

Large amount of data is available in the form of reviews, opinions, feedbacks, remarks, observations, comments, explanations and clarifications. In Education system, main focus is given to quality of teaching. That quality depends on coordination among teacher and student. Feedback analysis is more important to measure the faculty performance. Performance of faculty should be evaluated so that we can enhance our education quality. To measure the performance of faculty, we use classification technique by using opinion mining. We also use this technique on GPU architecture using CUDA-C programming to evaluate performance of a faculty in very less time. This paper uses opinion mining concept with GPU to extract performance of a faculty.

The proposed system aims at extracting meaning from the natural language query for querying the semantic knowledge sources. Semantic knowledge sources are systems conceptualized with Ontology. Characterization of a concept is through other concepts as a constraint over other. This very method to extract meaning from the natural language query has been experimented in this system. Constraints and entities from the query and the relationship between the entities is capable of transforming natural language query to a SPARQL (a query language for Semantic Knowledge sources). Further the SPARQL query is generated through recursive procedure from the intermediate query which is more efficient that mapping with patterns of the question. The system is compared with other systems of QALD (Question Answering over Linked Data) standard.

In a communication network the biggest challenge with multicasting is minimizing the amount of network resources employed. This paper proposes an ant colony optimization (ACO) and neural network (NN) based novel ACONN implementation for an efficient use of multicast routing in a communication network. ACO globally optimize the search space where as NN dynamically determine the effective path for multicast problem. The number of iteration and complexity study shows that the proposed hybrid technique is more cost effective and converges faster to give optimal solution for multicast routing in comparison to ACO and Dijkstra’s algorithm.

Now-a-days an object detection and classification is a unique perplexing difficulties. In the meantime the morphology and additional topographies of the defected objects are unlike from normal or defect free object, so it is possible to classify them using such artificial intelligence (AI) based structures. Here an substitute methodology followed by several AI procedures are established to categorize the defective object and defect free object by means of principal image texture topographies of various defective object like a soft drinks or cold drinks bottle and applying the pattern recognition techniques after that the successful accomplishing the image spitting, quality centered parameter abstraction as well as successive sorting of substandard and defect free bottles. Our results validated that Least Square support vector machine, linear kernel and radial function has maximum overall performance in terms of Classification Ratio (CR) is about 96.35 %. Thus, the proposed setup model is proved as a best choice for classification of an object.

An offline signature verification system based on feature extraction from signature images is introduced. Varieties of features such as geometric features, topological features and statistical features are extracted from signature images using Gabor filter technique. As all the features are not relevant, only the salient features are selected from the extracted one by a Rough Set Theory based reduct generation technique. Thus only the relevant features of the signatures are retained to reduce the dimension of feature vector so as to reduce the computation time and are used for offline signature verification. The experimental results are expressed using few parameters such as False Rejection Rate (FRR), False Acceptance Rate (FAR).

The paper proposes classification of protein sequences using K-Nearest Neighbor (KNN) algorithm. Motif extraction method N-gram is used to encode biological sequences into feature vectors. The N-gram generated is represented using Boolean data representation technique. The experiments are conducted on dataset consisting of 717 sequences unequally distributed into seven classes with a sequence identity of 25 %. The number of neighbors in the KNN classifier is varied from 3, 5, 7, 9, 11, 13 and 15. Euclidean distance and Cosine coefficient similarity measures are used for determining nearest neighbors. The experimental results revealed that the procedure with Cosine measure and the number of neighbors as 15 gave the highest accuracy of 84 %. The effectiveness of the proposed method is also shown by comparing the experimental results with those of other related methods on the same dataset.

With rapid increase in private power producers to meet the increasing power demand, results in the congestion problem. As the power transfer is increasing the operation of power systems is becoming difficult due to higher scheduled and unscheduled power flows. Interline Power Flow Converter (IPFC) is a most flexible device and effective in reducing the congestion problem. In this paper line utilization factor (LUF) is used for finding the best location to place the IPFC. The Harmony Search (HS) Algorithm is used for proper tuning of IPFC for a multi objective function which reduces active power loss, total voltage deviations, security margin and the capacity of installed IPFC of installed IPFC capacity. Simulation is carried out on IEEE-30 bus test system and the results are presented and analyzed to verify the proposed method.

In this work, a novel broadband circularly polarized spidron fractal slot antenna has been considered. The design and implementation of single Spidron antenna, arrays of 2 × 2 and 4 × 4 are considered along with 4 × 4 array with the modified ground has been considered. HFSS13.0 has been used for the design and simulation, and fabrication FR4 epoxy material has been used. The broadband characteristics are examined simulation, and experimental results were presented. Theoretically a 4 × 4 array antenna offers the gain of is 5 dB and wide band ranging from 12.17 to 14.98 GHz. Its bandwidth is 2813 MHz. Practically a 4 × 4 array antenna operates from 9.9 to 14 GHz. The grid Spidron array performance evolution and comparison with 2 × 2 and 4 × 4 arrays is presented.

Six axis industrial robots are widely used for carrying out various operations in industry and in the process it is subjected to varying payload conditions. This paper shows a methodology to determine the optimum value of pay load vis-à-vis the design parameters considering the criteria of reducing the material used to build the structure of industrial robot based on the finite element method (FEM). Different loads are applied at gripper and the total deformation is calculated. Finally, the weak area of the robot arm is found out and relative improving suggestions are put forward, which leads to the foundation for the optimized design. In the fuzzy-based method, the weight of each criterion and the rating of each alternative are described by using different membership functions and linguistic terms. By using this technique, the deformation is determined in accordance to load to the robotic gripper. The results of the analysis are presented and it is found that the triangular membership function is the effective one for deformation measurement as its surface plot shows a good agreement with the output result.

In modern receiver control applications achieving real time response is of main importance. The paper is a part of software development for CSM controller system. In this paper a real time application, for controlling SCAN DF receiver simulator will be implemented using POSIX socket APIs on VxWorks RTOS and process the data from the scanDF receiver subsystem. The processed data can be used to develop a database for electronic order of battle. This application will be able to interact with receiver using socket APIs, to exchange command and data packets. TCP/IP based socket APIs will be used to realize the application.

Mathematical techniques are easily understood by using Numerical analysis. This type of analysis is used for solving complex problems to do complicated calculations with the help of Computers. Numerical methods are used for finding the approximate solution to solve the mathematical problems. Modular design process increases debugging and testing process of a program. Additional tasks are performed by using new modules. In this paper object oriented design method is applied for Euler method.

This paper discusses the enhancement of voltage stability and damping in isolated hybrid system. Fuzzified compensation has been carried out with TCSC for proper enhancement of stability and reactive power. The wind energy conversion system with linear approximation is experimented with different loading conditions. Overall compensation has been achieved with TCSC Controller for different loadings and uncertain wind power inputs. Both proportional and integral constants are properly tuned and the optimized results are achieved with Fuzzy controller. The results vindicate the performance of the proposed compensator and desired results are achieved.

In this paper, we introduce a new high payload reversible data hiding scheme using weighted matrix. First, we enlarged the size of original image by interpolation. Then we perform modular sum of entry-wise multiplication with original image block and predefine weighted matrix. We subtract the modular sum from secret data and store the positional value by modifying three least significant bits of interleaved pixel. The original pixels are not affected during data embedding which assures reversibility. The proposed scheme provides average embedding payload 2.97 bits per pixel (bpp) and PSNR 37.97 dB. We compare our scheme with other state-of-the-art methods and obtain reasonably better results. Finally, we have tested our scheme through some attacks and security analysis which gives promising results.

In this paper, an Inter-cluster Ant Colony Optimization algorithm has been used that relies upon ACO algorithm for routing of data packets in the network and an attempt has been made to minimize the efforts wasted in transferring the redundant data sent by the sensors which lie in the close proximity of each other in a densely deployed network. ACO is being widely used in optimizing the network routing protocols. This work has focused on evaluating the performance of iLEACH protocol. The overall goal is to find the effectiveness of the iLEACH when ACO inter-cluster data aggregation is applied on it.

In this grand era of technology the wireless sensor networks outstand to serve multidisciplinary fields of technology whether it be technology, agriculture, healthcare, logistics monitoring etc. Agriculture field has recently adopted the wireless sensor technology to enhance the production and monitoring of crops in a more efficient manner. This paper presents the various applications of wireless sensor networks adopted for monitoring and measuring various parameters for crop and seed germination for effective yield generation. In this paper various parameters are monitored by using wireless sensor networks. Also in this paper a survey of most used sensor nodes and the features that they comprise of are depicted to analyze their usage in the various domains. This paper discusses the advantages of sensor nodes such as cheaper cost, compact size and how they help in managing the agricultural application yielding to best quality crops. The work in implementing wireless sensor nodes is ongoing and a survey in various dimensions is discussed in this paper.

In this paper, an enhanced version of Harmony Search (HS), called Tournament Selective Harmony Search (TSHS) is used to obtain an optimal set of weights for Functional Link Artificial Neural Network (FLANN) with Gradient Descent Learning (GDL) for the task of classification in data mining. The TSHS performs better than HS and Improved HS (IHS) by avoiding random selection of harmonies for their improvisation by introducing tournament selection strategy. This approach of TSHS to acquire optimal harmony in a population of harmony memory is adopted to find out optimal set of weights for FLANN model. The proposed TSHS-GDL-FLANN is compared with other alternatives by examining on various benchmark datasets from UCI Machine Learning repository. In order to get statistical correctness of results, the proposed method is analyzed by using ANOVA statistical test under null-hypothesis.

Cardiovascular disease (CVD) is a big reason of morbidity and mortality in the current living style. Identification of Cardiovascular disease is an important but a complex task that needs to be performed very minutely and accurately and the correct automation would be very desirable. Every human being cannot be equally skilful and so as doctors. All doctors cannot be equally skilled in every sub specialty and at many places we don’t have skilled and specialist doctors available easily. An automated system in medical diagnosis would enhance medical care and it can also reduce costs. In this study, we have designed a system that can efficiently discover the rules to predict the risk level of patients based on the given parameter about their health. Then we evaluate and compare this system using C45 rules and partial tree. The performance of the system is evaluated in terms of different parameter like rules generated, classification accuracy, classification error, global classification error and the experimental results shows that the system has great potential in predicting the heart disease risk level more efficiently.

Recurrent Neural Network (RNN) has been known to be very useful in predicting software reliability. In this paper, we propose a model that explores the applicability of Recurrent Neural Network with Back-propagation Through Time (RNNBPTT) learning to predict software reliability. The model has been applied on data sets collected across several standard software projects during system testing phase. Though the procedure is relatively complicated, the results depicted in this work suggest that RNN exhibits an accurate and consistent behavior in reliability prediction.

Soft set theory is a new mathematical approach to vagueness introduced by Molodtsov. This is a parameterized family of subsets defined over a universal set associated with a set of parameters. In this paper, we define membership function for fuzzy soft sets. Like the soft sets, fuzzy soft set is a notion which allows fuzziness over a soft set model. So far, more than one attempt has been made to define this concept. Maji et al. defined fuzzy soft sets and several operations on them. In this paper we followed the definition of soft sets provided by Tripathy et al. through characteristic functions in 2015. Many related concepts like complement of a fuzzy soft set, null fuzzy soft set, absolute fuzzy soft set, intersection of fuzzy soft sets and union of fuzzy soft sets are redefined. We provide an application of fuzzy soft sets in decision making which substantially improve and is more realistic than the algorithm proposed earlier by Maji et al.

Composite operations of arithmetic are extensively used in the applications of Digital Signal Processing (DSP). An optimized Multiply Accumulator Unit using fused Add-Multiply (FAM) operator by exploring structured and proficient recoding methods utilizing them. This paper deals with the study of performance comparisons of 16-bit and 32-bit MAC design based on EMBR techniques in terms of look up tables and power utilization with 8-bit and 16-bit recoding form of Modified Booth (MB) multiplier.

Our research work elaborated in the design and construction of a method that bring sustenance for a reduction in false assumptions during the detection of breast cancer. Our key drive of this research work was to elude the false assumptions in the detection practice in a cost effective manner. We proposed a unique method to decrease false assumption in breast cancer detection cases and split this method in three different modules as preprocessing, formation of homogeneous blocks and color quantization. The preprocessing convoluted in eradicating the extraneous slices. The formation homogeneous blocks sub-method was to do segmentation of the image. The task of the third sub-method (i.e. color quantization) was to break the colors amid different regions.

Regression Testing plays a vital role for the improvement in quality of product during software maintenance phase. This phase ensures that modification made to the system under test doesn’t adversely affect the performance of the existing features. Hence regression testing incurs more cost and time. Test case prioritization, which is one of the techniques of regression testing, is an efficient technique to minimize the cost and time of testing. In this paper, the author has proposed an approach for test case prioritization by maintaining information of previous and current release of the project in a repository. To represent the behavioral aspect of the system, it is modeled using both UML activity and sequence diagram. Then frequent pattern is generated by applying Association Rule Mining on the information stored in the repository. Finally the prioritization is carried out using the generated frequent patterns and Business Criticality Value of the different features.

Group Recommender System is one of the categories of recommender system, where the recommendation of things is for a group of users rather than for any individual. These system combines the preferences of each user present in the group and then predicts things which are suitable for the users of the group. Various grouping strategies are available, which are used to generate to group recommendations, but most of them are suitable when used for specific purpose only. In this paper we have proposed a novel approach to group recommender system using collaborative filtering technique, which can be applicable to all the real world scenarios where the data set uses rating system to distinguish among users’ preferences. We have made use of nearest neighbor algorithm to create a group of users with similar likeness. We have also applied the priority among users of the group as there are some members whose preferences might affect the whole group. We have validated our results with the movie lens data set which is the standard data set for recommender system testing.

Super resolution became one of the best techniques to obtain high resolution images as of a number of low-resolution images because of its simplicity and wide range of application in many fields of science and technology. There are several methods exist for super resolution but, wavelet transformation is chosen because of its minimalism and the constraints used to get better image restoration result. In this paper first Wavelet Transformation is considered to restore better image. Further Genetic algorithm is used to smooth the noise and better frequency addition into the image to get an optimum super resolution image.

Operating system selection problem is to choose the optimum OS based on user preferences on different factors. Functionality is one such attribute selected from software quality model which comprises several other sub factors preferred to analyse the quality of system. Analysis is done using AHP approach for weight calculations and results are validated using Entropy method on Functionality factors of Software quality model ISO 9126. This analysis helps in decision making for users which concerns about how well outcomes of any execution are achieved. A model AHP-Entropy is proposed which can be used to rank alternatives on both subjective and quantitative factors.

Firefly algorithm optimization is based on the attractiveness/brightness of the firefly. In firefly algorithm, a lighter (lesser fitness function) firefly move towards the brighter firefly (higher fitness function) with amplitude proportional to Euclidean distance between the lighter and brighter firefly. If no such brighter firefly is found then it moves randomly is search space. This random move causes chance of decrement in brightness of the brighter firefly depending on the direction in which it is move. We proposed a modified firefly algorithm in which movement of brighter fireflies is towards the direction of brightness instead of random move. If this direction of brightness is not in the process then firefly is in same position. We call this novel algorithm as MFA-LBG. Experimental results shows that modified firefly algorithm reconstructed image quality and fitness function value is better than the standard firefly algorithm (FA-LBG) and LBG algorithms. It is observed that that modified firefly algorithm convergence time is less than the standard firefly algorithm.

This paper is summery for experimental research work carried out on internet usage activities of students on social network sites. The summarized research work has proposed a novel and efficient method for discovery of user latent behavior from the student’s datasets. A comparison with existing standard method is also presented which show that the proposed approach is better on the basis of accuracy, error rates, time sharing etc. features.

This present paper focuses on the design of proportional-derivative-integral controller with derivative filter (PIDF) for automatic generation control problem. A two-area, six-unit reheat thermal-hydro restructured power system is considered with nonlinearities such as time delay (TD) and generation rate constraint (GRC). The gain parameters PIDF controllers are optimized by using differential evolution (DE) algorithm employed with integral of time multiplied absolute error (ITAE) as an objective function. The performance of proposed controller is investigated under all the possible scenarios that take place in a restructured power market. From simulation results reveals that DE optimized PIDF controller minimizes the errors effectively.

Artificial neural network (ANN) based forecasting models have been established their efficiencies with improved accuracies over conventional models. Evolutionary algorithms (EA) are used most frequently by the researchers to train ANN models. Population initialization of EA can affect the convergence rate as well as the quality of optimal solution. Random population initialization of EAs is the most commonly used technique to generate candidate solutions. This paper presents an opposition-based genetic algorithm (OBGA) learning to generate initial candidate solutions for ANN based forecasting models. The present approach is based on the concept that, it is better to begin with some fitter candidate solutions when no a priori information about the solution is available. In this study both GA and OBGA optimizations are used to optimize the parameters of a multilayer perceptron (MLP) separately. The efficiencies of these methods are evaluated on forecasting the daily closing prices of some fast growing stock indices. Extensive simulation studies reveal that OBGA method outperforms other with better accuracies and convergence speed.

Multilayer neural network based classifiers have been proven their better approximation and generalization ability in medical data classification. However they are characterize with both computational and structural complexities. This article proposes an Evolving Functional Link Network (EFLN) for medical data classification. First, the input signals are mapped from lower to higher dimensional feature space applying some trigonometric expansion functions. Then the optimal number of expanded input signals, weight vectors and network parameters are obtained by an evolutionary search technique. Therefore the optimal network structure can be achieved on fly by evolving a set of FLNs during training rather fixing it earlier. The proposed EFLN classifiers are validated with some benchmark data sets from UCI machine learning repository. The performances are compared with that of a gradient descent based FLN (GDFLN), multiple linear regressions (MLR) and a multilayer perceptron (MLP) and found to be superior.

Electrocardiogram signal is principally used for the interpretation and assessment of heart’s condition. The main criteria in ECG signal analysis is interpretation of QRS complex and obtaining its feature information. R wave is the most significant segment of this QRS complex, which has a prominent role in finding HRV (Heart Rate Variability) features and in determining its characteristic features. This paper intends to propose a novel approach for the analysis of ECG signals. The ECG signal is preprocessed using stationary wavelet transform (SWT) with interval dependent thresholding integrated with the wiener filter and is then subjected to Hilbert transform along with a window to enhance the presence of QRS complexes, to detect R-Peaks by setting a threshold. The proposed algorithm is validated with different parameters like Sensitivity, +Predictivity and Accuracy. The proposed method yields promising results with 99.94 % Sensitivity, 99.92 % +Predictivity, 99.87 % Accuracy. Finally the proposed method is compared with other methods to show the efficiency of the proposed technique for the analysis of ECG Signal.

The computation speed is restricted with the binary number system by generation and transmission particularly for the carry, as the bit size increments. Using a quaternary Signed Digit number system one may achieve an adder, subtraction which can be of carry free, borrow free respectively for multiplication. However QSD number system necessitates a different set of major modulo based logic aspects for every arithmetic operation. QSD is a superior radix number system, which is used for every arithmetic operation with generate carry free operation and the number scope for QSD is −3 to 3. Proposed QSD multipliers with generating partial products which perform very high speed operation as compared to QSD adders and also used for number of times addition process increasing. The addition process is a carry free operation and other processes for the digits of large numbers like 126,256 or further can be exerted with persistent delay and low difficulty. Tools Modelsim 6.0, Microwind are used.

Waste stemmed from inappropriate management is a major challenge for perishable resources. Improvement of the inappropriate management has great potential to improve the efficiency of the resources. This research aims to maximize profit and reduce resource spoilage through a fitness value approach based on the decay rate of the perishable resources. A particular type of resource whose decay rate is uniform with time is considered here and is defined as uniform perishable resource. But here in this paper it is shown that the best way to utilize those resources is to follow the first method (i.e. to pick up the best resource first).

Cloud computing is very promising technology to deliver computing resources like infrastructure (IaaS), platform (PaaS), software (SaaS) etc. in form of services over the Internet. Server consolidation is the mechanism to minimize active (running) physical servers in the data center of cloud service provider. Consolidation helps the service provider to make efficient usage data center resources and in turn to minimize running cost of data center. Users have to pay as per service usage. SLA is like contract between cloud service provider and users. Service Level Agreement (SLA) is the mechanism of providing guaranteed services to cloud users. SLA Violation may cause high penalties to service provider. In this paper it is shown that if consolidation is maximized, it may cause more number of SLA violation. A balanced approach of server consolidation with minimal SLA violations is suggested in this paper.

Smart homes form one of the major components leveraging demand response within the smart grid paradigm. Flexible pricing policies along with the capability of scheduling power among many homes form the crux of a wide variety of smart home power management controllers. However leveraging power scheduling for smart homes while keeping user costs minimal is a challenging proposition and involves complex multistage, stochastic, non-linear optimization techniques. For ease of computation, heuristic algorithms can be employed that require consumer load corresponding to smart homes which are not available a priori. The efficiency of power scheduling heuristics, however depend on the accuracy of the consumer loads forecasted. In this paper, we focus on developing a technique that can efficiently forecast consumer loads and thereafter the predicted load is fed to a GA heuristic based power scheduling algorithm for smart homes. Detailed procedure for the aforementioned forecasting has been presented and the results obtained are analyzed.

Mobile ad hoc network (MANET) has attracted many security attacks due to its characteristics of dynamic topology, limited resources and decentralize monitoring. One of these vulnerable attack is wormhole in which two or more malicious nodes form a tunnel like structure to relay packets among themselves. This type of attack may cause selective forwarding, fabrication and alteration of packets being sent. In this paper, we have proposed a way to protect network from wormhole attack by using identity based signature scheme on cluster based ad hoc network. Proposed scheme does not require distribution of any certificate among the nodes so it decreases computation overhead. We have also discussed existing work that either require certifcates or does not accomplish all the security requirements of network. Our simulation results show that attack is prevented successfully and it outperforms other schemes.

Optimization problem of an important traditional machining process namely surface grinding is considered in this work. The performance of machining processes in terms of cost, quality of the products and sustainability of the process is largely influenced by its process parameters. Thus, choice of the best (optimal) combination machining parameters is vital for any machining process. Hence, in present work a new algorithm is used for solving the considered optimization problem. The Jaya algorithm is a simple yet powerful algorithm and is a algorithm-specific parameter-less algorithm. The comparison of results of optimization show that the results of Jaya algorithm are better than the results reported by previous researchers using GA, SA, ABC, HS, PSO, ACO and TLBO.

Image segmentation is a process of elevating the objects by partitioning a digital image into multiple segments. To analyze an image, segmentation is the best process to follow. Especially, for detecting tumours from medical images such as brain, skin and breast in the field of medicine. To improve the results of brain images PSNR, ENTROPY image fusion technique is applied. Here the segmentation process is carried out by k-means clustered model algorithm. Then the entire data base is subjecting to classification mode under multi class svm sigmoid classifier. This generates a descent output of 96 % accurate results using various texture features they are contrast, energy, area of the tumour detecting by the cluster model and entropy. These parameters helps in identifying the tumour detection from brain MRI, CT scanned images.

In this paper, a higher order neural network called Pi-Sigma neural network with an improved Particle swarm optimization has been proposed for data classification. The proposed method is compared with some of the other classifiers like PSO-PSNN, GA-PSNN and only PSNN. Simulation results reveal that, the proposed IPSO-PSNN outperforms others and has better classification accuracy. The result of the proposed method is tested with the ANOVA statistical tool, which proves that the method is statistically valid.