Contemporary Computing

In this talk, we consider the problem of data propagation in wireless sensor networks and revisit the family of mixed strategy routing schemes. We will argue that maximizing the lifespan, balancing the energy among individual sensors and maximizing the message flow in the network are equivalent. We note that energy balance, although implying global optimality, is a local property that can be computed efficiently and in a distributed manner. We will then review some distributed, adaptative and on-line algorithms for balancing the energy among sensors.

By considering a simple model of the network and using a linear programming description of the message flow, we will show the strong result that an energy-balanced mixed strategy beats every other possible routing strategy in terms of lifespan maximization. We finalize by remarking that although the results discussed in this talk have a direct consequence in energy saving for wireless networks they do not limit themselves to this type of networks neither to energy as a resource. As a matter of fact, the results are much more general and can be used for any type of network and different type of resources.

Research in intelligent autonomous systems has historically been constrained by access to scalable high-performance computing. The challenges associated with scalability and tractability have significantly influenced analytical approaches and may have led to analysis in the lower dimensional spaces. Such reductions may overly simplify approaches to model and exploit redundancy in manners that human perception is able to in a bidirectional inferencing process. Evidence indicates that human perception involves different methods of classifying, indexing and associating information according to different spatio-temporal and saliency metrics. Recent surge in high performance computing with multicore processors and wide spread access to large disk-space at finer granularity, have triggered a renewed assessment of existing approaches. A survey of the techniques lead to a compelling need for efficient methods for capturing and exploiting “context” and “context-specific information” to swiftly change the behaviors of intelligent systems. The talk will highlight the evolution of signal processing, linguistic hierarchy, computing models where information fusion is considered, and draw some parallels, to make a case for a layered architecture for context-adaptive autonomous systems. Interactive access to petascale supercomputing has given a newer framework to incorporate context in such studies at scale that is required to perform complex intelligence tasks. The talk will highlight our current research in this direction including our vision of the future of intelligent autonomous systems with embedded high-performance computing with reach back compute power.

Here we focus on research in Intelligent MultiMedia or MultiModal computing concerning the computer processing and understanding of perceptual signal and symbol input from at least speech, text and visual images, and then reacting to it, involving signal and symbol processing techniques from engineering, computer science, artificial intelligence and cognitive science. With IntelliMedia systems, people can interact in spoken dialogues with machines, querying about what is being presented and even their gestures and body language can be interpreted. Of particular interest is the mapping of inputs into, and outputs out of, semantic representations and this is what distinguishes Intelligent MultiMedia from traditional MultiMedia. We will demonstrate here software prototypes such as ‘PlayPhysics’, which uses computer games to teach physics to first year university students. PlayPhysics is a virtual learning environment for teaching physics which integrates research in Intelligent Tutoring Systems (ITSs), where students learn about concepts such as momentum by trying to get an astronaut back to his craft in time by determining optimal mass and velocity. PlayPhysics also gives detailed feedback online. Another prototype we have developed is ‘MemoryLane’, a mobile digital storytelling companion for older people. Reminders of the person’s past, such as photos, video, favourite songs or poems (provided by the individual or their family) are input as text, image, moving image and sound, creating the material from which multimodal stories can be generated. Each story is different and MemoryLane can factor in any problems with the person’s eyesight, hearing or dexterity and adapt the presentation accordingly (like making the text larger or reducing the amount of sound or images). Based also on preferences it allows the holder to select aspects they like and reject anything they wish to forget. All of this work falls within ‘The Imagineering Quarter’ within the City of Derry/Londonderry, Northern Ireland, comprising 5 neighbouring buildings of the North West Regional College (NWRC) (‘Foyle’, ‘Strand’, ‘Lawrence’) & University of Ulster (‘Foyle Arts’, ‘Computing’) focussed on teaching, research & technology transfer with software demonstrators in Digital Creativity (digital storytelling, music, film, theatre, dance, art, design; games, virtual worlds) linking to The Nerve Centre, Verbal Arts Centre community centres, cross-border Letterkenny Institute of Technology (LYIT), local software industry & access to Project Kelvin – a secure high capacity dedicated broadband link (10 G. LanPhy) direct to Canada, USA, Europe & rest of the island with a delay of only 2 ms. Derry/Londoderry is First UK City of Culture, 2013 and a key contributor is The Imagineering Quarter

.

The problem of identifying meaningful patterns (i.e., motifs) from biological data has been studied extensively due to its paramount importance. Motifs are fundamental functional elements in proteins vital for understanding gene function, human disease, and identifying potential therapeutic drug targets. Several versions of the motif search problem have been identified in the literature. Numerous algorithms have been proposed for motif search as well. In this talk we survey some of these algorithms. We also summarize our contributions to motif search and related problems. In addition, we will summarize MnM, a web based system built by us for motif search that is used by biologists widely.

A lot of research effort has been put into community detection from all corners of academic interest such as physics, mathematics and computer science. In this paper I have proposed a Bi-Objective Genetic Algorithm for community detection which maximizes modularity and community score. Then the results obtained for both benchmark and real life data sets are compared with other algorithms using the modularity and MNI performance metrics. The results show that the BOCD algorithm is capable of successfully detecting community structure in both real life and synthetic datasets, as well as improving upon the performance of previous techniques.

The applications of biologically-inspired computing models into the distributed computing paradigm have potential benefits offering self-detection and self-reconfiguration capabilities of the computing systems. In the large scale distributed systems, the arbitrary failure of nodes and network partitions are the major challenges in terms of failure detection, fault-tolerance and maintainability. This paper proposes a novel distributed algorithm for self-detection and self-reconfiguration of distributed systems on the event of arbitrary node failures resulting in network partitioning. The algorithm is designed based on the hybridization of biological membrane computing model and cell-signaling mechanisms of biological cells. This paper presents the problem definition, design as well as performance evaluation of the proposed algorithm.

Grid facilitates global computing infrastructure for user to consume the services over the network. To optimize the workflow grid execution, a robust multi-objective scheduling algorithm is needed. In this paper, we considered two conflicting objectives like execution time (makespan) and total cost. We propose a multi-objective scheduling algorithm, using

ε

–MOEA approach based on evolutionary computing paradigm. Simulation results show that the proposed algorithm generates multiple scheduling solutions near the Pareto optimal front with uniform spacing and better convergence in small computation time.

Wheat grain quality assessment is important in meeting market requirements. The quality of the wheat can be judge byits length, thickness, width, area, etc. In this paper on the basis of simple mathematical calculations different parameters of a number of wheat grains are calculated. The present paper focused on the classification of wheat grains using morphological. The grain types used in this study were Hard Wheat, Tender Wheat. In this paper the application of neural network is used for assessment of wheat grain. The contours of whole and broken grains have been extracted, precisely normalised and then used as input data for the neural network. The network optimisation has been carried out and then the results have been analysed in the context of response values worked –out by the output neurons.

The Paper presents a simple computer-aided design approach for designing Finite Impulse Response (FIR) digital filters. FIR filter is essentially a digital filter with non Recursive responses. Since the error surface of digital FIR filters is generally nonlinear and multimodal, global optimization techniques are required in order to avoid local minima. There are many ways for the design of FIR Digital filters. This Paper Presents soft computing technique for the design of FIR filters. In this Paper, Genetic Algorithm (GA) base evolutionary method is proposed for design of FIR digital filter. GA is a well-known powerful global optimization algorithm introduced in combinatorial optimizations problems. The Simulation result for the employed example is presented in this paper and can be efficiently used for FIR digital filter design.

In this paper, we present some observations on the various algorithms proposed to find a Minimum Independent Dominating Set (MIDS). MIDS is proven to be an NP-hard problem. We compared an exact algorithm based on intelligent subset enumeration with another exact algorithm based on matching in graphs. We found that the former performs better than the latter for small graphs despite having a worse asymptotic complexity. There is only one Polynomial Time Approximation Scheme (PTAS) proposed in literature for computing MIDS which works for polynomially bounded growth graphs. We observed that changing the

ε

value in the PTAS reduces the running time quite drastically but does not increase the cardinality returned significantly. We compared the cardinality of the IDS returned by various heuristics for grid, unit disk graph and general graph topologies. The results show that the highest degree heuristic returns the best cardinality amongst all these algorithms in literature for all graphs except grid graphs for which the inter-dominator 3-hop distance heuristic performs better. To the best of our knowledge, this is the first empirical study where the exact, PTAS and heuristic solutions to the MIDS problem have been compared in terms of the quality of the solution returned as well as provide insights into the behavior of these approaches for various types of graphs.

This paper proposes a distributed solution for fault diagnosis in wireless sensor networks (WSN). Fault diagnosis is achieved by comparing the heartbeat sequence number generated by neighbouring nodes and dissemination of decision made at each node. The proposed protocol considers channel error where the channel is modelled as two state Markov model. Theoretical analysis and simulation results show that both message and time complexity of proposed protocol is

O

(

n

) for an n-node WSN. This work investigates the energy consumed in diagnosing a fault event. The per-node energy overhead is substantially reduced and becomes scalable.

Recently Usama et al. proposed a chaos-based satellite image cryptosystem, which employed multiple one-dimensional chaotic maps in novel manner to enhance the robustness, security and efficiency of sensitive satellite imagery. It is very efficient in terms of encryption time. The authors of the cryptosystem under study claimed that it has high level of security and can be applied to transmit confidential multimedia images over Internet/shared network. Unfortunately, the security analysis of the cryptosystem reveals that it has serious security flaws. Consequently, it is susceptible to a number of attacks. In this paper, the cryptanalysis of original cryptosystem is presented and it is shown that the attacker can recovers the plain-image from given cipher-image under three types of classical cryptographic attacks without knowing the secret key. The simulation results of cryptanalysis demonstrate that the cryptosystem highly lacks security and cannot be utilized for the protection of confidential/sensitive multimedia images such as the satellite imagery.

Regular expression signature matching has been used increasingly in network security applications like intrusion detection systems, virus scanners, network forensics, spam filters etc. However, signature matching causes decrease in performance on the host when load increases due to the large requirements in terms of memory and processing power. This is mainly because every byte and possibly a combination of bytes of the input have to be matched against a large set of regular expressions. Modern Graphics Processing Units (GPUs) are capable of high performance computing and recently are being used for general purpose computing. The large performance throughput and data parallelism of these modern GPUs is used to perform matching on the input data in parallel. Experimental results show that our GPU implementation is up to 12 times faster than the traditional CPU implementation while being up to 4 times faster than the GPU implementation using texture memory.

Differential Evolution (DE) and Particle Swarm Optimization (PSO) algorithms have gained a lot of popularity in the last few years for solving complex optimization problems. Several variants of both the algorithms are available in literature. One such variation is combining the two algorithms in a manner so as to develop an algorithm having positive features of both the algorithms. In the present study we propose a hybrid of DE and PSO algorithm called Mixed Particle Swarm Differential Evolution Algorithm (MPDE) for solving global optimization algorithms. The numerical and statistical results evaluated on a set of benchmark functions show the competence of the proposed algorithm. Further, the proposed algorithm is applied to a practical problem of determining the location of the earthquakes in the Northern Himalayan and Hindu Kush regions of India.

Measures of vertex similarity have been incorporated in graph matching algorithms. Graph matching tries to retrieve a 1-1 correspondence between vertices of two given graphs. In this paper, the vertex similarity measure of Blondel et al. is studied for its usefulness in detecting graph isomorphism. Firstly, the applicability of this measure to distinguish similar pairs from dissimilar pairs is shown to be limited in scope even for small graphs. In a preliminary experiment, we show that Blondel’s vertex similarity measure does not retrieve the isomorphism within a graph of 14 nodes. We propose a refinement of Blondel’s measure. Zager et al. also refine Blondel’s measure and further propose a graph matching algorithm. We propose a graph matching algorithm based on the lines of Zager et al. and test our algorithm against Zager’s as well as Blondel’s and show that the proposed refinement performs better than both the measures with regard to graph isomorphism problem. The performance is evaluated systematically on a large bench mark data set made available by Foggia et al. The proposed algorithm performs with 90.10% accuracy on all of the 18,200 pairs of isomorphic graphs available in the benchmark dataset.

Internet routers play an important role during the time of network congestion. All the Internet routers have some buffer at input and output ports, which hold the packets at the time of congestion. Many queue management algorithms have been proposed but they focus on fixed queue limit. Recognizing the fact that active queue management algorithms have fixed maximum queue limit, we direct our attention to variable length queue limit for Combined Input Output Queued (CIOQ) switches. We incorporated our proposed technique, which is a fuzzy logic control based generic variable length active queue management scheme in TCP/IP networks, to the drop-tail and the Adaptive RED (A-RED) algorithm. The empirical results show low packet loss and high queue utilization in modified algorithms (augmented with variable length active queue management scheme) in comparison to the original droptail, RED and A-RED algorithms.

Graph Coloring Problem is a well-studied classical NP-hard combinatorial problem. Several well-known heuristics and evolutionary approaches exist to solve single-objective graph coloring problem. We have considered a bi-objective variant of graph coloring problem, in which the number of colors used and the corresponding penalty which is incurred due to coloring the end-points of an edge with same color, are simultaneously minimized. In this paper, we have presented an evolutionary approach with Multipoint Guided Crossover (MPGX) to minimize both objectives simultaneously. On applying proposed evolutionary algorithm over standard graph coloring problem instances, a guaranteed solution to the single-objective graph coloring problem is achieved. We have adapted a few well-known heuristics which are evolved for single-objective graph coloring problem to generate set of solutions for bi-objective graph coloring problem and obtained Pareto fronts. Empirical results show that proposed evolutionary algorithm with simple Multipoint Guided Crossover generates superior or (near-) equal solutions in comparison with the adapted heuristic solutions as well as with evolutionary algorithm solutions using a few crossover (Penalty-based Color Partitioning Crossover (PCPX) and Degree Based Crossover (DBX)) operators across entire Pareto front for considered bi-objective variant of graph coloring problem.

Timeout based models are an important class of design models for discrete event modeling and simulation of real-time systems and protocols. In this work, we define a framework to graphically represent timeout based models with synchronous communication. The formalism offers system designers an expressive graphical language with well defined semantics to model their system designs and reason about their behavior. For actual implementation, these models are expressed using GraphML standard with support for embedded ANSI C code. We further devise an automated translation technique (and develop corresponding prototype tool support) to translate the GraphML designs into SAL (Symbolic Analysis Laboratory) model specifications, which in turn, can be formally verified using the SAL verification engine.

In this paper we propose a mechanism for enhancing the performance of the Artificial Bee Colony Algorithm (ABCA) by making use low discrepancy Sobol sequence. The performance of the proposed Sobol sequence guided ABC (S-ABC) is analyzed over several benchmark functions and also compared to that of basic ABC. The empirical results show that the presence of low discrepancy sequence like that of Sobol, significantly improves the performance of the basic ABCA.

The spanning tree problem is a well-studied problem and Evolutionary Algorithms (EAs) have been successfully applied to a large variants of the spanning tree problem. The behavior of an evolutionary algorithm depends on the interaction between the encoding and the genetic operators that act on that encoding. Various encodings and operators have been proposed for the spanning tree problems in the literature. The edge-set encoding has been shown very effective for such problems as it shows high locality and high heritability. However, it requires effective genetic operators to exploit favorable characteristics of an encoding to guide the search and obtain high quality results. In this work, we consider bounded-diameter minimum spanning tree (BDMST) problem and improve upon the crossover operator for edge-set encoding. The empirical results show the effectiveness of our approach. Finally, based on the simulation results, we highlight interesting properties of the new recombination operator which helps it find better trees compared to the previous one.

Over the past two decades, the scenario in the computing world has evolved from client-server to distributed systems and then to central virtualization called as cloud computing. Computing world is moving towards Cloud Computing and it remains as buzzword of the current era. Earlier, users had complete control over their processes and data stored in personal computer where as in cloud, cloud vendor provides services and data storage in remote location over which the client has no control or information. As application and data processing takes place in public domain outside the designated firewall, several security concerns and issues arise. The main objective of the paper is to provide an overall security perspective in cloud Computing and highlight the security concerns and other issues. The paper also highlights few technical security issues in cloud computing.

This paper presents the experimental evaluation for recognition of consonant-vowel (CV) units under noise. Noise is one of the common degradation in real environments which strongly effects the performance of speech recognition system. In this work, initially effect of noise on recognition of CV units is studied by using two-stage CV recognition system proposed in our earlier studies. Later spectral processing based speech enhancement methods such as spectral subtraction and minimum mean square error (MMSE) are used for preprocessing to improve the CV recognition performance under noise. Performance of the CV recognition is studied on Telugu broadcast database for white and vehicle noise. Experimental results show that the speech enhancement techniques gives the improvement in the CV recognition performance under noise case.

This paper discuss the effect of noise on vowel onset point (VOP) detection performance. Noise is one of the major degradation in real-time environments. In this work, initially effect of noise on VOP detection is studied by using recently developed VOP detection method. In this method, VOPs are detected by combining the complementary evidence from excitation source, spectral peaks and modulation spectrum to improve VOP detection performance. Later spectral processing based speech enhancement methods such as spectral subtraction and minimum mean square error (MMSE) are used for preprocessing to improve the VOP detection performance under noise. Performance of the VOP detection is analyzed by using TIMIT database for white and vehicle noise. In general, performance of VOP detection is degraded due to noise and in particular performance is effected significantly due to spurious VOPs introduced at low SNR values. Experimental results indicate that the speech enhancement techniques provides the improvement in the VOP detection performance by eliminating spurious VOPs under noise.

Spatial data clustering groups similar objects based on their distance, connectivity, or their relative density in space whereas in the real world, there exist many physical constraints e.g. highways, rivers, hills etc. that may affect the result of clustering. Therefore, these obstacles when taken into consideration render the cluster analysis a hopelessly slow exercise. In this paper, a clustering method is being proposed that considers the presence of physical obstacles and uses obstacle modeling as a preprocessing step. With a view to prune the search space and reduce the complexity at search levels, the work further incorporates the hierarchical structure into the existing clustering structure. The clustering algorithm can detect clusters of arbitrary shapes and sizes and is insensitive to noise and input order.

In this paper, we propose a framework for modeling goal driven requirements of web applications. Web engineers mostly focus on design aspects only overlooking the real goals and expectations of the user. Goal oriented Requirement Engineering is a popular approach for Information system development but has not been explored much for Web applications. However, in today’s times Web is dominating in every business making it imperative that its requirements are analyzed carefully and in profundity. Goal driven requirements analysis helps in capturing stakeholders’ goals very finely, by choosing between alternatives and resolving conflicts. Detailed classification of both functional and non-functional requirements specific to web applications is discussed in the presented work. A framework, GOREWEB (Goal oriented Requirements Engineering for Web Applications) is proposed for analyzing goals and translating them into functional and non-functional web requirements.

Smart phones nowadays, with colossally large number of users have become very prominent[1]. Furthermore, this increasing prominence goes arm in arm with the rising number of malwares[2], thus making it inevitable to take cognizance of the need for an efficient malware detection mechanism. However, sundry former associated works like [3] & [4] for malware detection, have not cited a novel strategy which we feel can be attributed to the lack of malware classification in them. Fundamentally, classification of malwares provides a head start to the detection mechanism by curtailing the search space & the processing time of the detection mechanism. So in order to accomplish the malware classification, we develop few malwares and discuss their behavior and aftereffects on the device. And then we utilize the resource victimized by these malware on the phone as base for classification and allocate same class to those malwares that affect same resource. Finally by employing the aforementioned malware classification, we outline a strategy for their detection. Experimentation of the detection scheme on the malwares with and without classification reveals that with classification the real and CPU time consumed by detection process are almost 45% and 22% of the respective times without classification, which thus elucidates the fact that classification based malware detection in future can be employed as a propitious tool.

Transmission Control Protocol (TCP) is known to suffer from performance degradation in mobile wireless environments, as such environments are prone to packet losses due to high bit error rates and mobility induced disconnections, because TCP is well reliable protocol for wired networks. In wired network packet loss due to congestion. Proposed scheme is a true end-to-end approach, based on the idea of exclusive handover message and is used for alleviating the degrading effect of host mobility on TCP performance. Experiments are performed using the Network Simulator (NS-2). The simulator has been extended to incorporate wireless link characteristics.

Remote sensing image classification in recent years has been a proliferating area of global research for obtaining geo-spatial information from satellite data. In Biogeography Based Optimization (BBO), knowledge sharing between candidate problem solutions or habitats depends on the migration mechanisms of the ecosystem. In this paper an extension to Biogeography Based-Optimization is proposed for image classification by incorporating the non-linear migration model into the evolutionary process. It is observed in recent literature that sinusoidal migration curves better represent the natural migration phenomenon as compared to the existing approach of using linear curves. The motivation of this paper is to apply this realistic migration model in BBO, from the domain of natural computing, for natural terrain features classification. The adopted approach calculates the migration rate using Rank- based fitness criteria. The results indicate that highly accurate land-cover features are extracted using the extended BBO technique.

Single Electron Transistor is the most prominent nanoelectronic device that will dominate the operations of nanoscaled integrated circuits. Molecules, especially DNA is prophesized to be integral part of the futuristic ICs. In this paper the current voltage characteristics of DNA base Cytosine are obtained by non-equilibrium Green’s function combined with density functional theory. The pattern of current flow for an applied voltage sweep of 0-5 V is plotted. The phenomenon of tunneling is exhibited in the characteristics of molecules. The DNA base cytosine displays a typical surge of current in the voltage sweep section of 0.4V-0.6V, indicating single electron effects. The effect of gate voltage on the current-voltage characteristics of cytosine was studied in the gated two-probe setup. The typical section of characteristics of cytosine was re-drawn by varying the gate potential. The application of gate bias exhibits excellent ON/OFF switching for combinations of the two applied voltages- source voltage and gate voltage. Repetitive peaks are also observed in current when gate voltage is varied, fixing source potential. In this paper the cytosine molecule is proposed as a switch, AND gate and OR gate in this paper that can be used in DNA based molecular electronic devices.

In wireless sensor network (WSN) the accuracy of data is important to maintain the networks’ performance. Therefore detecting nodes which either provide faulty readings or do not provide any information is an essential issue in sensor network management. As a whole the solution is to detect nodes with data and function faults, this paper proposes a novel method to detect nodes with both types of faults without assuming a particular sensing model. The performance of proposed fault model is more accurate and requires less communication compared to the existing methods.

This paper mainly deals with designing a biological controller for biped robot to generate biped locomotion inspired from human gait oscillation. The nonlinear dynamics of the biological controller is modeled by designing a Central Pattern Generator (CPG) which is the coupling of the Relaxation Oscillators. In this work the CPG consists of four Two-Way coupled Rayleigh Oscillators. The four major leg joints (e.g. two knee joints and two hip joints) are being considered for this modeling. The CPG parameters are optimized using Genetic Algorithm (GA) to match an actual human locomotion captured by the Intelligent Gait Oscillation Detector (IGOD) biometric device. The Limit Cycle behavior and the dynamic analysis on the biped robot have been successfully simulated on Spring Flamingo robot in YOBOTICS environment.

A secure routing mechanism is the basis of security in mobile ad hoc networks. Seeing that nodes have to share routing information in order to find the route to the destination, further an ad hoc network is an open setting where everyone can participate, trust is a key concept in secure routing mechanisms. In this paper, we propose a novel opinion based trust evaluation model to detect misbehaving nodes and further securing data transmission by avoiding such node from path selected. Misbehaving nodes are detected based on the node own direct observation, but in the circumstances of network failure as node congestion or collusion a well-behaving node can be misidentified. Therefore we require opinion of other nodes before proceeding to isolation of such misbehaving node. Source nodes maintain a path reliability index associated with each path to the intended destination. The most trustworthy and reliable path is selected by source node for forwarding data packets to the intended destination. We have evaluated the misbehaving node detection rate and the performance of our method along a number of parameters through simulation. Results show that our method increases the throughput of the network while also discovering a secure route.

This paper presents our experimental work on analysis of sentiments and mood from a large number of Weblogs (blog posts) on two interesting topics namely ‘Women’s Reservation in India’ and ‘Regionalism’. The experimental work involves transforming the collected blog data into vector space representation, doing Parts of Speech Tagging to extract opinionated words and then applying semantic orientation approach based SO-PMI-IR algorithm for mining the sentiment and mood information contained in the blog text. We obtained interesting results, which have been successfully evaluated for correctness through both manual tagging and by cross-validating the outcomes with other machine learning techniques. The results demonstrate that these analytical schemes can be successfully used for blog post analysis in addition to the review texts. The paper concludes with a short discussion of relevance of the work and its applied perspective.

In this paper, we put forward a technique for optimization of the search results obtained in response to an end user’s query. With the enormous volume of data present on the web, it is relatively easy to find matched documents containing the given query terms. The difficult part is to select the best from the possible myriad of matching pages. Moreover, most Web search engines perform very well for a single keyword query but fail to do so in case of multiple terms. In this paper by using the concept of Meta search engines we propose a suitable query processing and optimization algorithm for giving the best possible result for multiple term keywords in the ranked order.

Cloud computing is on-demand computing in which the computing resources are owned and managed by a service provider and the users access the resources via the Internet. But cloud computing potential doesn’t begin and end with the personal computer’s transformation into a thin client. The mobile platform is going to be heavily impacted by this technology as well. License management is a major issue faced by mobile cloud computing paradigm. This paper presents an agent-based license management approach, the MobiLim, for mobile cloud computing. The mobile devices access the services of cloud and pay for the usage to service provider. Independent software vendors (ISVs) control the access to the resource provider resources. MobiLim provides a secure and robust license management solution to the service provider.

Data mining algorithms are widely used today for the analysis of large corporate and scientific datasets stored in databases and data archives. Industry, science and commerce fields often need to analyze very large datasets maintained over geographically distributed sites by using the computational power of distributed and parallel systems. Grid computing emerged as an important new field of distributed computing, which could support the distributed knowledge discovery applications. In this paper, we have proposed a method to perform Data Mining on Grids. The Grid has been setup using Foster and Kesselman’s Globus Toolkit, which is the most widely used middleware in scientific and data intensive grid applications. For the development of data mining applications on grids we have used Weka4WS. Weka4WS is an open source framework extended from the Weka toolkit for distributed data mining on Grid, which deploys many of machine learning algorithms provided by Weka Toolkit. To evaluate the efficiency of the proposed system, a performance analysis of Weka4WS by executing distributed data mining tasks, namely clustering and classification, in grid scenario has been performed. At last, a study on the speed up obtained by doing data mining on grids is done.

This paper presents text independent emotion recognition from speech using mel frequency cepstral coefficients (MFCCs) along with their velocity and acceleration coefficients. In this work simulated Hindi emotion speech corpus, IITKGP-SEHSC is used for conducting the emotion recognition studies. The emotions considered are anger, disgust, fear, happy, neutral, sad, sarcastic, and surprise. Gaussian mixture models are used for developing emotion recognition models. Emotion recognition performance for text independent and text dependent cases are compared. Around 72% and 82% of emotion recognition rate is observed for text independent and dependent cases respectively.

This paper proposes a new method of concatenation cost calculation for enhancing the optimality in unit selection. Instead of defining same set of concatenation costs for all types of speech unit transitions, costs are defined based on the type of unit transitions. Different types of unit transitions that can occur mainly in an utterance are voiced to voiced, voiced to unvoiced and unvoiced to unvoiced transitions. Natural measure of continuity is identified for each of these transitions, and costs are defined accordingly. For voiced to voiced transitions, in addition to spectral continuity, pitch and energy continuity metrics are proposed. In case of voiced to unvoiced and unvoiced to unvoiced transitions, silence duration embedded in the unvoiced region is proposed as the continuity metric. This approach of segment specific concatenation cost calculation improves the quality of syllable based text to speech synthesis. Listening tests provide a proof on the effectiveness of proposed methodology which has clearly shown the decrease in perceptual discontinuity at joins, and improvement in the overall quality of the synthesised speech.

Devising a scheme for evaluating the service quality of an institutional electronic library is a difficult and challenging task. The challenge comes from the fact that the services provided by an institutional electronic library depend upon the contents requested by the users and the contents housed by the library. Different types of users might be interested in different types of contents. In this paper, we propose a technique for evaluating the service quality of an institutional electronic library. Our scheme is based on the service profiles of contents requested by the users at the server side which is hosted at the library. Further, we propose models to analyze the service quality of an electronic library. For analyzing the service quality, we present two analytical models. The first one is based on the number of days by which the item to be served by the library is delayed and the penalty points per day for the duration for which the item is delayed. The second model is based on the credits earned by the library if the item is served in a timely fashion, and the penalties, thereof, if the item is delayed. These models may help in evaluating the service quality of an electronic library and taking the corrective measures to improve it.

In this paper, we propose a cooperative caching scheme that exploits an energy efficient model in Wireless Sensor Network. Cooperative caching is done in the form of concentric circular cache layers around the sink. A Circular cache layer is a group of nodes falls under circumference of the circle formed from the sink as center with a certain radius. At a time only a single cache layer becomes active which caches the data in it. A fading-aware routing is utilized to minimize the total power consumption during packet transfer. The scheme performs well in the multiple sink environments. Proposed scheme is compared with the existing Fading-aware energy efficient routing approach. Simulation results show the performance and better efficiency of the network.

A major hindrance in the way of reliable and lossless communication is the inter symbol interference (ISI). To counter the effects of ISI and to have proper & reliable communication an adaptive equalizer can be employed at the receiver end. This paper considers the applications of artificial neural network structures (ANN) to the channel equalization problem. The problems related with channel nonlinearities and can be effectively subdued by application of ANNs. This paper contains a new approach to channel equalization using functional link artificial neural network (FLANN). In this paper we have incorporated the novel idea of utilizing an evolutionary technique called Differential Evolution (DE) for the training of FLANN we have compared the results with back propagation (BP) and Genetic Algorithm (GA) trained FLANNs. The comparison has been drawn based upon the minimum Mean Square Error (MSE) and Bit Error Rate (BER) performances. From this study it is evident that the DE trained FLANN performs better than the other types of equalizers.

In the vehicular ad hoc networks several types of data are disseminated and transmission protocol changes with change in the type of data. Some information is useful for only those vehicles which are in certain specific range or location. Dissemination of traffic update message to all the vehicles is wastage of channel bandwidth. In our proposed method, vehicles which are already trapped in the traffic jam are transmitting traffic update message that is disseminated to the vehicles which might be trapped to the traffic jam. Simulation results demonstrate that the proposed protocol reduces congestion in the dense traffic region and efficiently utilizes the bandwidth in stressful road scenarios.

Providing quality of service (QoS) in a mobile ad hoc network is a challenging task due to its peculiar characteristics. This paper aims at presenting a routing protocol which identifies a path between a given source and a destination in TDMA-based ad hoc networks. This path is examined for satisfying QoS in terms of end-to-end delay. For that purpose, we introduce an analytical model for end-to-end delays incurred by a packet from a given source to a destination. We have evaluated the performance of our protocol through simulation. The proposed protocol performs better in term of QoS satisfaction ratio as compared to existing protocols.

In past few years, GIS systems have marked a significant development in India. Governments are now planning to utilize the GIS data collected in E-Governance Systems. But when it comes to actual implementation it is often realized that the geographical data still lacks significant comprehension. Example when it was decided to utilize the global imagery of India in Land Record Management, authorities felt that no matter how efficient the data is but it is still inferior to the imagery done by Google and Yahoo. The matter of fact that Google Maps indexes precise data about roads, colonies and even buildings and its all due to people’s participation from apps like Wikimapia etc. In this paper we are presenting a Referenced Approach to merge the data from different GIS systems and implement them in e-governance systems. It also describes the setup of Land record Management system, Survey Management and Personal Identity System based on this approach.

At present, secondary distribution network is controlled manually in most of the places. The limitation of the system is that there is no feedback on the operation. Maintenance staff has to physically check the correct operation. The system has an isolator / changeover switch at the incomer end and a contactor for switching the power. There are no protections against overloads and short circuit faults. There is no metering facility in the system and the revenue calculations are done manually. In this paper a control system is designed for remote operation of LV feeders for continuous monitoring and load shedding with GSM interface with central control station SCADA with two way communication using SMS messages. In this paper a prototype model is developed by integrating SCADA control with GSM technology using the logic Twidosoft v.3 for continuous monitoring and obtaining parameters from the field and remote control of LV feeders. The proposed scheme is intended to give better and more control options, performance, safety to equipment & people and help in reducing downtime by way of suggesting preventive maintenance automatically by the system. Offered system includes the metering functions, so that the meter readings can be downloaded from remote Central Control Station using the same GSM Modem and GSM Network. Automatic re-switching by remote control to enable supplies to be restored quickly to all consumers is a current feature of this scheme. In this project various hardware equipments are configured using IFIX software. Many screens are created for sending and receiving reports. A program is written in MATLAB Editor for determining the channel capacity of GSM network and then the optimum message transfer rate is calculated for this type of application. The scheme has been successfully tested and the analytical discussions and results are presented for the same.

Test Effort Estimation is an important activity in software development because on the basis of effort cost and time required for testing can be calculated. Various models are available for estimating effort but to some extent all models result in erroneous effort estimation. So there is a need to optimize the effort estimated. Meta heuristic techniques can be used for this purpose, to optimize a problem by iteratively trying to improve a solution, using some computational methods. Particle Swarm Optimization is one such technique which have been incorporated in this work to get good test effort estimates.

A popular trend in data dissemination involves online data sources that are hidden behind query forms, which are part of the

deep web

. Extracting information across multiple deep web sources in a domain is challenging, but increasingly crucial in many areas. Keyword search, a popular information discovery method, has been studied extensively on the surface web and relational databases. Keyword-based queries can provide a powerful yet intuitive means for accessing data from the deep web as well. However, this involves many challenges. For example, deep web data is hidden behind query interfaces, deep web data sources often contain redundant and/or incomplete data, and there is often inter-dependence among data sources. Thus, it is very hard to automatically execute cross-source queries.

This paper focuses on answering

cross-source

queries over deep web data sources. In our approach, we model a list of deep web data sources using a

graph

to capture the dependencies among them, and we consider the problem of answering cross-source queries over these deep web data sources as a graph search problem. We have developed a bidirectional query planning algorithm to generate query plans for two types of cross-source queries, which are

entity-attributes

queries and

entity-entity relationship

queries.

This paper describes a unified modeling technique applied after threat identification step of threat modeling process, for ranking the causes of a threat which is then used for threat mitigation and testing. The paper presents a unique approach that starts with enumeration of causes for each possible threat over the system with construction of threat cause model that diagrammatically describes the causes and sub-causes responsible for the occurrence of a threat. The paper suggests an approach for the ranking of both threats and their causes for effective mitigation. After applying threat cause mitigation strategy, testing of system towards a threat is verified by checking the security at the perimeter of the cause model for that threat. This unique technique assures that ensuring all sub-causes at lowest level of abstraction impossible will make the system safe towards a particular threat. Unlike other techniques this technique is unified as it starts with a threat model for each individual threat, that enumerates causes of their occurrence and then the same is used for mitigation and testing. Hence this strategy can ensure security when applied to all threats over the system.

Face recognition system has been evolving as a convenient biometric mode for human authentication. Face recognition is the problem of searching a face in the reference database to find a face that matches a given face. The purpose is to find a face in the database, which has highest similarity with a given face. The task of face recognition involves the extraction of different features of the human face from the face image for discriminating it from other persons. Many face recognition algorithms have been developed and have been commercialized for applications such as access control and surveillance. For enhancing the performance and accuracy of biometric face recognition system, we use a multi-algorithmic approach, where in a combination of two different individual face recognition techniques is used. We develop six face recognition systems based on the six combinations of four individual techniques namely Principal Component Analysis (PCA), Discrete Cosine Transform (DCT), Template Matching using Correlation and Partitioned Iterative Function System (PIFS). We fuse the scores of two of these four techniques in a single face recognition system. We pperform a comparative study of recognition rate of these face recognition systems at two precision levels namely at top- 5 and at top-10. We experiment with a standard database called ORL face database. Experimentally, we find that each of these six systems perform well in comparison to the corresponding individual techniques. Overall, the system based on combination of PCA and DCT is giving the best performance among these six systems.

Assignment Problem was well studied in the past 50 years, and is of great value in operations research and engineering. With the growing size of these problems and the new complexities introduced over the years, multi robot task assignment problems have become an important focus of assignment research. The work presented in this paper considers the scenario where multiple destination sites are available. The task of the controller is to assign a robot to each site as soon as possible in such a way so that robots can reach their destination with minimum travelling distance. Efficient algorithms for solving problem of this type have important applications in industries and home automation. Our main contributions are twofold: (a) A wave front based path planning method to compute the cost for the performance of each robot on each target (destination); and (b) An assignment algorithm for the assignment of robots to targets so that the sum of the total cost so obtained is as minimum as possible. The proposed approach has been tested through computer simulation. Experimental results demonstrate that our algorithm runs fast and produces near-optimal solutions on randomly generated instances.

In 2005, Liao et al. pointed out some weaknesses in Das et al.’s dynamic ID-based scheme. They proposed a slight modification to Das et al.’s scheme to improve its weaknesses. In 2008, Gao-Tu, and in 2010, Sood et al., found vulnerabilities in Liao et al.’s scheme; and independently proposed its security enhanced versions. However, we identify that Gao-Tu’s scheme is insecure against user impersonation attack, server counterfeit attack, man in the middle attack, server’s resource exhaustion attack and does not provide session key agreement. We also demonstrate that Sood et al.’s scheme is still vulnerable to malicious user attack in different ways and user’s password is revealed to the server. Besides both the schemes have no provision for revocation of lost or stolen smart card. Our cryptanalysis results are important for security engineers, who are responsible for the design and development of smart card-based user authentication systems.

Service Discovery is an important element which requires finding a set of suitable webservice candidates faster for the service requester among those published by the service provider. Among large number of functionally-equivalent, it is difficult for users to choose a best service to be invoked. This paper proposes a new webservice reference platform which has service discovery element behaving as a fuzzy expert system. The proposed webservice reference model makes the service discovery element automatic .

This paper introduces a new fast method for the calculation of exponential B-splines sample at regular intervals. This new method is fast and it also considered polynomial spline as special case. This algorithm is based on a combination of FIR and IIR filters which enables a fast decomposition and reconstruction of a signal. In this paper we have tried to get the interpolation function which uses the symmetric exponential functions of 4th order. We are considering the real part of these functions which is used for interpolation of real signals corresponding to different exponential parameter that leads to less band limited signals when they are compared with polynomial B-spline counterparts. These characteristics were verified with 1-D and 2-D examples. We are also going through all the interpolation methods which are already in use.

In this paper, we present a multi-purpose density-based clustering framework. The framework is based on a novel cluster merging algorithm which can efficiently merge two sets of DBSCAN clusters using the concept of intersection points. It is necessary and sufficient to process just the intersection points to merge clusters correctly. The framework allows for clustering data incrementally, parallelizing the DBSCAN algorithm for clustering large data sets and can be extended for clustering streaming data. The framework allows us to see the clustering patterns of the new data points separately. Results presented in the paper establish the efficiency of the proposed incremental clustering algorithm in comparison to IncrementalDBSCAN algorithm. Our incremental algorithm is capable of adding points in bulk, whereas IncrementalDBSCAN adds points, one at a time.

In this work, Neural Fuzzy Inference Network (NFIN) controller is implemented that has a number of membership functions and parameters that are tuned using Genetic Algorithms. The number of rules used to define the Neuro-Fuzzy controller is then pruned. Pruning is utilized effectively to eliminate irrelevant rules in the rule base, thus keeping only the relevant rules. Pruning is performed at various threshold levels without affecting the system performance. This methodology is implemented for Water Bath System and analysis has been carried out to investigate the effect of pruning using a multi-step reference input signal. From the results, it is concluded that reasonably good performance of controller can be obtained with lesser number of rules, thus, reducing the computational complexity of the network.

The proposed work focuses on developing a methodology that promotes software development by partitioning the whole system into different independent components and aspects. This facilitates component reuse along with the ease of modeling the components separately and emphasizing on the concerns that the widely used OOP paradigm has failed to address. Identification of reusable components is carried out using the hybrid methodology and aspects are identified by the domain experts. Along with the components the platform independent models and aspects developed are stored in separate repositories so as to be used in development of other software of similar requirements and basic structure.

To enhance the efficiency of grid resource management systems, resource provisioning is required. For efficient resource provisioning, a policy based resource provisioning framework needs to developed. This paper presents Resource Provisioning framework and discusses the resource policy for better resource utilization and customer satisfaction.

PHYTODB contains a repository of phytoplasma genes and proteins. It provides a unified gateway to store, search, retrieve, update information about phytoplasma and computational resources for the analysis of nucleotide and aminoacid sequence data of phytoplasma. Server facilitates to differentiate and classify new phytoplasma for taxonomic purposes. PHYTODB database was updated by dividing the whole resources into two domains:

DataBanks

and

Tools

.

DataBanks

serve as the storage device of all information. Functional characterization of genes and protein are done. Updated Groupidentifier tool by rearrangement of RFLP classification scheme of phytoplasma and possibilities 6 new groups based on the new tool. PhytoDB can be obtained through

http://220.227.88.253/phytodb/

.

Redundant data transfer over a network is one of the important reasons of traffic congestion today. In this paper, we proposed an efficient and secure file sharing model using de-duplication technology to resolve it. A file sharing based de-duplication system reduce bandwidth and storage at both client and server machine. It does not download duplicate blocks that already have been downloaded. To achieve the security of client data, three-tier architecture is proposed in this work. For this purpose SHA-1 hash function is being used, in which 8KB block of data is converted into a 20 bytes digest. Thus the design presents a dramatic reduction in storage space requirement for various workloads and hence reduces time to perform backup in bandwidth constraint environment.

Integration of Feature-Oriented Programming (FOP) and Aspect- Oriented Programming (AOP) methodologies can overcome their individual limitations and thus superior software can be evolved. This approach has been investigated in evolving application software for a representative ‘Event Management System’ using Eclipse-based open source environment. The study concludes that this approach supports modular design and implementation of consistent, reusable, maintainable and cost effective ‘Event Management Software System’, tailored to the specific needs of the stakeholders.

In this paper we introduce a robust Bengali Automatic Speech Recognition(ASR) system which covers most of the commonly spoken words. This ASR system converts standard Bengali continuous speech to Bengali Unicode with a decent accuracy rate. The existing reported Bengali ASR system is confined within small vocabulary. The system uses triphone clustering mechanism and trigram language model to increase accuracy. For execution of training we have created Bengali Speech Corpus, corresponding Bengali Text corpus, Pronunciation Dictionary.

System on Chip (SOC) is the integration of IP core like CPU’S, DSP’s, Application Cores, memories etc. Communication between these IP cores is necessary for proper functionality of SOC. On chip communication arbiters plays an important role for the communication arbitration. In this paper, four arbitration algorithms i.e. Round Robin, Lottery Based Arbiter, FIFO (First in First out), TDMA (Time Division Multiple Access) are implemented in Hardware Description Languages. The results of four arbiters are compared the basis of area, power and delay.

Due to a serious case of plagiarism this paper has been retracted.

The paper “Performance Analysis of Handover TCP Message in Mobile Wireless Networks” <

http://www.springerlink.com/content/h12j112777784247/

> appearing on pages 254-261 of this publication has been retracted due to a severe case of plagiarism.