Search

Longitudinal linked data are RDF descriptions of observations from related sampling frames or sensors at multiple points in time, e.g., patient medical records or climate sensor data. Observations are expressed as measurements whose values can be repeated several times in a sampling frame, resulting in a considerable increase in data volume. We devise a factorized compact representation of longitudinal linked data to reduce repetition of same measurements, and propose algorithms to generate collections of factorized longitudinal linked data that can be managed by existing RDF triple stores. We empirically study the effectiveness of the proposed factorized representation on linked observation data. We show that the total data volume can be reduced by more than 30 % on average without loss of information, as well as improve compression ratio of state-of-the-art compression techniques.

We consider a fading version of the dirty-paper problem, as proposed by Grover and Sahai. In this formulation, the various signals involved are complex-valued, and the interference (known only to the transmitter) is multiplied by a random complex-valued coefficient, whose phase is known only to the receiver. We focus on a compound channel formulation, and seek to maximize the worst-case performance. We present an achievable strategy modeled on the lattice-based approach of Erez, Shamai and Zamir and propose heuristic methods to optimize its parameters. We also derive an upper bound on the maximum achievable transmission rates. Our bounds are shown to be tight in some settings, yielding a complete characterization of capacity. We also provide simulation results, indicating the practical effectiveness of our approaches.

In this paper, in order to improve rejection capabilities of the paper currency recognition system for unknown currency patterns on promise of ensuring recognition capabilities for known currency patterns, a feed-forward neural network (FNN) with Gaussian activation function is proposed. The proposed activation function is a ridge-like function. Moreover, a hybrid-learning algorithm for optimizing the width parameters of the Gaussian function is proposed. In the network the Gaussian activation function instead of the sigmoid function is employed in all units of hidden and output layers. The algorithm consists of two steps, one is exploring local minima by employing the gradient descent search, and the other is extricating the search from local minima, in which a random search with the downhill simplex method is employed. The results of simulation reveal the potential effectiveness of the proposed activation function and the algorithm. The system with the proposed activation function and the proposed algorithm can recognize known currency patterns and reject the unknown currency patterns effectively.

To study any electromagnetic system, the geometry model must be discretized into elements with an appropriate size for the working frequency. The discretization of a system must be transparent to the user of electromagnetic computing tools. A mesher is presented based on the paving algorithm. The algorithm has been modifled to allow triangular elements and has been accelerated by distributing the load on multiple processors simultaneously. Also, a multilevel mode has been implemented. With this tool, any geometry deflned by NURBS (Non Uniform Rational B-Spline) surfaces can be decomposed into triangular and quadrangular curved elements.

Hypercomputation is about the feasibility of machines and systems that are either more expressive or computationally more powerful than the Turing machine. A number of researchers and thinkers have put forth a number of supposedly knock-out arguments against hypercomputation. Nevertheless, these arguments are not unwavering as they seem to be and here I explain why.

Generation of question papers through a question bank is an important activity in learning management systems and educational institutions. The quality of question paper is based on various design constraints such as whether a question paper assesses different problem solving skills as per Bloom's taxonomy, whether it covers all units from the syllabus of a course and whether it covers various difficulty levels. In this paper, we present the design of an algorithm to generate question paper template which satisfies the above mentioned constraints. The algorithm is illustrated in the paper by using four constraints namely question paper format, coverage of syllabus, coverage of difficulty levels and coverage of cognitive levels according to Bloom's taxonomy. The algorithm presented is extensible to support any number of user defined constraints.

We study a spatial evolutionary rock-paper-scissors game with synchronized strategy updating. Players gain their payoff from games with their four neighbors on a square lattice and can update their strategies simultaneously according to the logit rule, which is the noisy version of the best-response dynamics. For the synchronized strategy update two types of global oscillations (with an ordered strategy arrangement and periods of three and six generations) can occur in this system in the zero noise limit. At low noise values, all nine oscillating phases are present in the system by forming a self-organizing spatial pattern due to the comprising invasion and speciation processes along the interfaces separating the different domains.

This work forms part of a general research effort into developing a methodology for reliability estimates of results obtained in numerical hydrodynamic computations. We consider the possibility of overtopping at the crest of a dyke of height h0positioned on a river with several uncertain parameters, such as discharge, Manning coefficient, bathymetry, etc. The estimation model is based either on the first-order reliability method (FORM), on the multi-form method, or on the importance sampling methods and is coupled with algorithms for the resolution of an adjoint optimization problem. Numerical tests are carried out on flows over a channel with a bump and on a river. The results obtained with our algorithms are compared with those obtained with the commercial software Nessus®and with the Monte Carlo method. The proposed multi-form approach combined with a robust optimization algorithm provides reliable results within reasonable computation times.

References to Thue’s 1914 paper on string transformation systems are based mainly on a small section of that work defining Thue systems. A closer study of the remaining parts of that paper highlight a number of important themes in the history of computing: the transition from algebra to formal language theory, the analysis of the “computational power” (in a pre-1936 sense) of rules, and the development of algorithms to generate rule-sets.

The measurement of stacked-sheet quantity is an essential step in packaging and printing production, and its counting accuracy has a direct impact on economic efficiency of related companies. With its noncontact, non-destructivity and real-time measurement merits, the machine vision method has been widely applied to quality control for high-end printing products. In this paper, we aim to circumvent the fringe detection problem in stacked-sheet images by introducing a level line guided line-segment growing algorithm. Then, a high-accuracy measurement of stack quantity can be realized with the improvement of precision and completeness on fringe identification.Our work mainly consists of three parts: 1) A unidirectional gradient operator is adopted to eliminate multiple responses on a single fringe. 2) The gradient magnitude and level-line direction are combined to improve the growth of line support regions in noisy environment. 3) To completely identify each sheet fringe, a connected component analysis algorithm is integrated to remedy the local gap in line detection. The performance of our algorithm has been verified in experiments using various kinds of printed-papers with a large number. It is shown that the long-term measurement error is less than 0.75‰ and is sufficient to meet the requirement of factory applications.

In this paper, the Two-Way Channel (TWC) with noncausal Channel State Information (CSI) is investigated. For the discrete memoryless channel an achievable rate region is derived. Then by extending the result to the Gaussian TWC with additive interference noise, it is shown that the capacity region of the latter channel is the same as the capacity when there is no interference, i.e., a two-way version of Costa's writing on dirty paper problem is established.

We present a new class of irregular low-density parity-check (LDPC) codes for moderate block lengths (up to a few thousand bits) that are well-suited for rate-compatible puncturing. The proposed codes show good performance under puncturing over a wide range of rates and are suitable for usage in incremental redundancy hybrid-automatic repeat request (ARQ) systems. In addition, these codes are linear-time encodable with simple shift-register circuits. For a block length of 1200 bits the codes outperform optimized irregular LDPC codes and extended irregular repeat-accumulate (eIRA) codes for all puncturing rates 0.6~0.9 (base code performance is almost the same) and are particularly good at high puncturing rates where good puncturing performance has been previously difficult to achieve.

Vehicular traffic congestion is a well-known economic and social problem generating significant costs and safety challenges, and increasing pollution in the cities. Current intelligent transport systems and vehicular networking technologies rely heavily on the supporting network infrastructure which is still not widely available. This paper contributes towards the development of distributed and cooperative vehicular traffic congestion detection by proposing a new vehicle-to-vehicle (V2V) congestion detection algorithm based on the IEEE 802.11p standard. The new algorithm allows vehicles to be self-aware of the traffic in the street, performing congestion detection based on speed monitoring and cooperation with the surrounding vehicles. Cooperation is achieved using adaptive single-hop broadcasting which depends on the level of congestion. The paper presents the congestion detection algorithm and the cooperative communication in detail, and presents performance evaluation using large-scale simulation in Veins framework based on OMNeT++ network simulator and SUMO vehicular mobility simulator. Results show that precise congestion detection and quantification can be achieved using a significantly decreased number of exchanged packets.

Effective recognition and localization of paper defect based on machine vision is the key issue for paper defect detection system. This paper proposed an improved algorithm by combination with Blob analysis algorithm and image preprocessing approach to detect the paper defects which exist in captured images by a linear charge coupled device (CCD) camera. First, the defected images are preprocessed, such as image denoising, image segmentation, connectivity analysis, and then extract effective paper textures: defect amount, regional area, long axis, short axis, central position and so on, meanwhile draw the minimum bounding rectangles. Compared with the traditional morphology algorithm and threshold segmentation and fractal feature algorithm, the improved algorithm is validated by a great deal of experimental results with high detection efficiency and defects localization accuracy.

In this paper, we consider some practical scheduling problems under unavailability constraints (break-down periods, maintenance durations and/or setup times). Such problems can be met in different industrial environments and be associated to numerous real-life applications. The studied criteria are related to the maximum lateness minimization. The paper is organized in two main parts. The first part focuses on some special cases (makespan minimization when tails are equal and maximum lateness minimization when release dates are equal (see Section 2). The second part is devoted to the maximum lateness minimization (see Section 3). In each part, we present the main contributions and explain their principles (complexity results, heuristic algorithms and their worst-case performance, existing approximation schemes, exact methods, dynamic programming,...), Finally, Section 4 concludes the paper.

Nowadays volume images are frequently used in many applications. Volume images of paper can be analyzed to increase the understanding of the complexity of the fiber network in paper and its effect on the optical and mechanical properties of paper. We show how curve and surface representations of the fiber, the fiber wall, and the fiber lumen can be computed using distance transform based algorithms. These representations are used for easy computation of wall thickness, degree of collapse, fiber length, slenderness ratio, fiber curl, and torsional resistance for the fibers. Free-fiber segments are identified, again using a distance transform based algorithm. Finally, we show tools for qualitative visual inspection of the fibers. The methods are evaluated and illustrated using both sets of synthetic and of real data. In addition to the analysis of the fiber network in paper, they can be used in many other applications where shape analysis of elongated objects is to be performed.

In common with many other materials, paper tends to shrink as it dries. Although every attempt is made to restrain paper, some shrinkage occurs on all paper machines in the direction perpendicular to that of manufacture and this shrinkage is always much higher at the edges of the machine than in the centre. Measurement of the profile of this cross-machine shrinkage is possible using the fast Fourier transform to locate and measure periodic elements imprinted by the filtration fabrics used during the formation of the paper web. This paper describes a new method which allows the geometrical relationships within the fabric to be used along with dimensional changes to estimate shrinkage. The method has the advantages over previous methods of more tolerant sampling protocols, operator independent analysis and improved accuracy.

Elasticsearch is a web search tool in view of Lucene. Apache Lucene is a free and open-source data retrieval programming library. Versatile Search gives a conveyed, multitenant-fit full-content web search tool with a HTTP web interface and pattern free JSON archives. It is created in Java and has been released as open source under the terms of the Apache License. Elasticsearch can be utilized to pursuit a wide range of records. It gives adaptable hunt, has close continuous pursuit, and backings multitenancy. It is appropriated, which implies that records can be partitioned into shards and every shard can have zero or more duplicates. Every hub has one or more shards, and goes about as a facilitator to delegate operations to the right shard(s). Elasticsearch is like a wrapper on top of Lucene. In this paper a detailed description of how lucene’s scoring algorithm works and how elasticsearch uses it as “similarity algorithm”

Through the needs of Loading-Machine storage scheduling in forestry-pulp-paper production logistics intelligent distribution system, analysis the Loading-Machine scheduling model of problems and improvement measures, put forward Loading-Machine intelligent production scheduling algorithm, based on ensuring the feasibility of scheduling, scheduling to rationalization, equalization, execute only, production process optimization, and realize the Loading-Machine intelligent production scheduling through the computer programming.

Many researchers have compared results from experiments using paper ballots and computer ballots, yet few have studied the interchangeability of the two data collection methods. If computers fail between sessions of an experiment, one would like to be able to use paper ballots for that session with some confidence that the experimental results will not be affected. The objective of this study was to determine if ballot type had a significant influence on descriptive analysis results. Multivariate analysis of variance indicated no significant differences (P

The calibration curve for any device or method to infer suction is the most important characteristic to evaluate for reliable measurement to be obtained. Some published literature (e.g. Houston et al. 1994; Bulut et al. 2001; Leong et al. 2002) have presented results in which the calibration for the filter paper method is not unique in relation to the type of suction (i.e., total or matric). A review of the data from the literature is presented, together with data showing the reason that justifies a unique calibration curve for the filter paper. It is also shown that it is necessary to check the calibration curve by performing a quick calibration check. A procedure for calibrating the filter paper for this check is given. Fundamental matters necessary for understanding the concept on which the filter paper method is based are also presented.

Error correction codes are used in semiconductor memories to protect information against errors. Simple error correction codes are preferred due to their low redundancy and encoding/decoding latency. Hamming codes are simple and can be easily built for any word length. They only allow single error correction, so a multiple error can lead to a wrong decoding. Multiple errors often manifest as burst errors, and they are becoming more frequent as integration scale increases. This paper proposes modified Hamming codes, with the same redundancy and coverage as the original versions, but adding short burst error detection. Three code examples, with different error correction and detection capabilities, are presented. They are especially well-suited for memories, where the length of the data word is commonly a power of 2, and low redundancy and fast and simple encoder and decoder circuits are required.

The paper gives a review of already existing 3-T XOR cells and provides an optimized value of (W/L) on the basis of simulation results obtained, so as to improve the threshold loss problems present in the existing designs of 3-T XOR cells thus helping improve the driving capability, however the driving capability is not sufficient for large circuits like multipliers, hence has a scope for further improvement. Using the best improved version of 3-T XOR cell a Full Adder Circuits is designed. All the basic circuits and their improved versions have been implemented in Cadence Virtuoso for 180nm technology and 1.8v sources.

Background and AimsThe critical appraisal of the evidence presented in research papers is an early step in the process of evidence-based design (EBD), which aims at transferring research findings into design practice (Stichler, 2010a). However, because architects and designers are often not trained in research methods, they may feel somewhat overwhelmed by the terms used in the description of the methodology of scientific papers. Therefore, when designing healthcare facilities, they may not feel confident discussing the evidence available from research studies with members of the medical professions, who often display a profound knowledge of research methods. Also, architects and designers may encounter difficulties in evaluating the quality of the evidence resulting from numerous research papers, which sometimes arrive at conflicting results.The purpose of this article is to develop a simple method to appraise the quality of research papers suitable for EBD. The most common study designs will be explained, and an algorithm, which could be helpful for guiding architects and designers when assessing the quality of evidence, will be introduced.Appraising the Quality of Evidence in the Process of EBDIn building design, there is a broad spectrum of evidence sources, such as experts' opinions, guidelines, observational studies, and systematic reviews. In the design process, it is necessary to determine which sources are scientifically valid and how the resulting findings are to be incorporated best into design decisions. The methodology of EBD offers a way of acquiring, appraising, weighting, and transferring the existing evidence into design practice (Rosswurm & Larrabee, 1999). This circular process is shown in Figure 1.The circle of the evidence-based design process starts with the definition of the research question and the development of a research strategy in step one. In the second step, relevant papers are acquired and selected. In the third step, which comprises the critical appraisal of the research studies' quality, this article comes into use by providing a simple algorithm for architects, designers, and other practitioners. It should be noted that this appraisal only rates the methodological quality of a given study. When "weighing the evidence" in step four, the results of several studies on the chosen research question are combined. Additionally, external factors, such as an estimation of the evidence's consistency, relevance, and external validity, are taken into account (SIGN 50, 2008). In step five, design recommendations, which are often scaled between weak and strong (Guyatt et al., 2008a), are developed. This judgment is made on the basis of the quality of the evidence (see step three) and its assigned weight (see step four). In the sixth and last step, the design decisions made on the basis of the design recommendations are implemented and evaluated.As it has been outlined, the critical appraisal of the quality of scientific evidence is a crucial step in EBD, answering the question to what extent the results of a research study are valid. However, what determines the "quality of evidence"? In general, it is defined as "the extent to which a study's design, conduct, and analysis has minimized selection, measurement, and confounding bias" (West et al., 2002). The better a study manages to protect against bias and error, the higher the quality of the results produced. Methods to protect against bias are, among others, group allocation, randomization, and blinding. Group allocation means dividing the participants into at least two groups: a treatment group and a control group. Randomization is the masked, random allocation of participants to the treatment or the control group. Blinding also refers to the group allocation: in the best case, during the intervention, neither the participant nor the researcher knows which group a participant belongs to (a doubleblind study). …

With the increasing use of CiteSeer academic search engines, the accuracy of such systems has become more and more important. The paper adopts the improved particle swarm optimization algorithm for training conditional random field model and applies it into the research papers’ title and citation retrieval. The improved particl swarm optimization algorithm brings the particle swarm aggregation to prevent particle swarm from being plunged into local convergence too early, and uses the linear inertia factor and learning factor to update particle rate. It can control algorithm in infinite iteration by the iteration between particle relative position change rate. The results of which using the standard research papers’ heads and references to evaluate the trained conditional random field model shows that compared with traditionally conditional random field model and Hidden Markov Model, the conditional random field model ,optimized and trained by improved particle swarm, has been better ameliorated in the aspect of F1 mean error and word error rate. DOI: http://dx.doi.org/10.11591/telkomnika.v11i3.2188 Full Text: PDF

Most traditional scientific papers are unstructured documents, which are difficult to meet the requirement of structured retrieval, statistical classification and association analysis and other high-level application, how to extract and analyze the structured information of the papers becomes a challenging problem. A structured information extraction algorithm for unstructured and/or semi-structured machine-readable documents is proposed, it according to the extracted rules after feature learning on the basis of analyzing the basic structure and format features of traditional scientific papers, which extracts the title, author, abstract, keywords, text and other elements of paper from the unstructured documents, such as Word, then exports the structured text from the traditional scientific papers with the format required by multi-dimensional scientific papers, it can meet the requirements of structured retrieval, statistical classification and other high-level applications of scientific papers.

In order to keep up with the requirements of College English Diagnostic Test System developed by our team, after analyzing the pros and cons of current test paper generating algorithms, the MCKP algorithm, which is based on maximum coverage of knowledge points (MCKP), is proposed. In this paper, the knowledge hierarchy of College English Test (CET4) of China is constructed to provide the foundation for MCKP. Then given three main parameters, including the required items number, the mastery rates of knowledge-point and the testing frequencies of knowledge points, the theory and implementation of the MCKP algorithm is elaborated. At last, experiments indicate relatively high success rate and effectiveness of test paper generating process. Therefore, MCKP algorithm can provide an essential improvement in test paper generating process. Keywordsmaximum covarage of knowledge points (MCKP); knowledge hierarchy of CET-4;mastery rates ;testing frequencies; success rate

The paper put forward a kind of test bank group volume algorithm which is based on the balance of the strategy. This algorithm is based on topics, to grasp the requirements as constraint conditions of small set of item bank division, and adjust the difficulty of all kinds of questions to the corresponding balance coefficient approximation, In order to achieve overall control of the difficulty of examination paper. Through The Higher Algebra group roll verification. This algorithm is based on the volume efficiency of higher to achieve control of the examination paper difficulty.

With various kinds of intelligent metering equipment coming into service, there has been an urgent need for a set of examination system in the electric power industry to check the staff's level of measuring electric energy. In this paper, we design and implement a random test paper generation algorithm for this examination system, and analyze the experiment's data, based on the practical requirement of a training system for the electric energy measurement in a certain electric company. At the same time, in order to get a set of test papers to satisfy the given conditions, this paper discusses how to use optimized genetic algorithm to generate test papers from the question bank. This paper introduce fishnet algorithm to generate test papers automatically for getting a better, more fair and more objective test paper.

Item bank system has been developed with Java and MySQL. The system contains course management module, examination questions management module, test paper management module and examination management module. The test paper generation algorithm used in this system was also discussed.

We propose a density evolution based dirty-paper code design framework that combines trellis coded quantization with multi-level low-density parity-check (LDPC) codes. Unlike existing design techniques based on Gaussian approximation and EXIT charts, the proposed framework tracks the empirically collected log-likelihood ratio (LLR) distributions at each iteration, and employs density evolution and differential evolution algorithms to design each LDPC component code. In order for the approximated LLR distributions obtained using density evolution to better match the true LLR distributions, a novel decorrelator is added to the decoder to make channel LLRs and check node LLRs almost independent of each other. Simulation results show that at 1 bit per sample transmission rate, the dirty-paper codes designed using the proposed method operate within 0.53 dB of the theoretical limit, reducing the best known result (with a 0.58 dB gap to the limit) by 0.05 dB at the same complexity and block length, while this improvement can be as large as 0.21 dB (corresponding to a 0.37 dB gap to the limit) when we use higher complexity encoder/decoder with longer block length.

The paper presents and evaluates the power of a new scheme that generates search heuristics mechanically for problems expressed using a set of functions or relations over a finite set of variables. The heuristics are extracted from a parameterized approximation scheme called Mini-Bucket elimination that allows controlled trade-off between computation and accuracy. The heuristics are used to guide Branch-and-Bound and Best-First search. Their performance is compared on two optimization tasks: the Max-CSP task defined on deterministic databases and the Most Probable Explanation task defined on probabilistic databases. Benchmarks were random data sets as well as applications to coding and medical diagnosis problems. Our results demonstrate that the heuristics generated are effective for both search schemes, permitting controlled trade-off between preprocessing (for heuristic generation) and search.

Several models have been proposed to represent conditional executions and dependencies among real-time concurrent tasks for the purpose of schedulability analysis. Among them, task graphs with cyclic recurrent behavior, i.e., those modeled with a single source vertex and a period parameter specifying the minimum amount of time that must elapse between successive activations of the source job, allow for efficient schedulability analysis based on the periodicity of the request and demand bound functions (em rbf and dbf). We leverage results from max-plus algebra to identify a recurrent term in rbf and dbf of general task graph models, even when the execution is neither recurrent nor controlled by a period parameter. As such, the asymptotic complexity of calculating rbf and dbf is independent from the length of the time interval. Experimental results demonstrate significant improvements on the runtime for system schedulability analysis.

In this paper, a new data hiding method - writing on wet paper using multi-pixel differencing is presented in order to provide large embedding capacity and improve further steganographic embedding efficiency for the stego-image. It takes into consideration four pixels of a block, and the differences between the lowest gray-value pixel and its surrounding pixels are used to embed the secret data. The receivers can extract secret bits from carrier images only by some matrix multiplications no matter how we embed the secret message. Thus, the invisibility of this steganographic method is achieved in process of secret information transmission. The experimental results show that our scheme has a larger embedding capacity and better robust performance against the methods of active attacks such as noise addition.

International audience; The goal of this study was to evaluate the areal mass distribution (defined as the X-ray transmission image) of paper from its optical transmission image. A Bayesian inversion framework was used in the related deconvolution process so as to combine indirect optical information with a priori knowledge about the type of paper imaged. The a priori knowledge was expressed in the form of an empirical Besov space prior distribution constructed in a computationally effective way using the wavelet transform. The estimation process took the form of a large-scale optimization problem, which was in turn solved using the gradient descent method of Barzilai and Borwein. It was demonstrated that optical transmission images can indeed be transformed so as to fairly closely resemble the ones that reflect the true areal distribution of mass. Furthermore, the Besov space prior was found to give better results than the classical Gaussian smoothness prior (here equivalent to Tikhonov regularization).

Algorithm and strategy directly affect the success rate, efficiency and quality of generating test paper. From the analysis of the usual requirement of the user proposition and the result of test paper, the algorithm of generating test paper based on the degree of difficulty is put forward as the strategy of step deals from the and the weight average model and discrete variable random binomially distributed model.

In this paper we introduce a novel and efficient approach to detect topics in a large corpus of research papers. With rapidly growing size of academic literature, the problem of topic detection has become a very challenging task. We present a unique approach that uses closed frequent keyword-set to form topics. Our approach also provides a natural method to cluster the research papers into hierarchical, overlapping clusters using topic as similarity measure. To rank the research papers in the topic cluster, we devise a modified PageRank algorithm that assigns an authoritative score to each research paper by considering the sub-graph in which the research paper appears. We test our algorithms on the DBLP dataset and experimentally show that our algorithms are fast, effective and scalable.

In order to improve the auto-generating test paper's quality at the cost of low time, an intelligent algorithm based on immune genetic algorithm was proposed. In the algorithm immune process was added into the basic framework of genetic algorithm, and the algorithm based on vaccination was put forward and used for test paper auto-generating. The problem that genetic algorithm is precocity and easy to fall into a local optimization was resolved by this algorithm. The feasibility and validity of the algorithm was proved by test data compared with other algorithms. At last the validating system enhances the users' condition constraints for test paper through manual inching, thus the system is more simple and practical.

This is the third special issue published in this journal with selected contributions from the VI Southern Conference on Programmable Logic (SPL, www.splconf.org). SPL is one of the first-class conferences on programmable logic around the world. Started in 2005, SPL has since 2007 the technical cosponsorship of the IEEE Circuits and Systems Society (CAS). The conference has been always organized in the south hemisphere. The selection of papers presented in this special issue is coming from the last 6th edition, held in Ipojuca, Brazil, from March 24 to 26, 2010. This issue is focused on speech and video acceleration applications. The first paper by P. Eh Kan et al., “FPGA implementation for GMM-based speaker identification” in the area of biometric-based speaker identification, describes the hardware implementation for classification of a text-independent (Gaussian mixture model) GMMbased speaker identification system. Authors show that the proposed hardware implementation allows simultaneous identification processing for a large number of voice streams in real time. In “FPGA implementation of a pipelined Gaussian calculation for HMM-based large-vocabulary speech recognition,” R. Veitch et al. present several alternatives for a Gaussian core as part of a larger speech recognition system. These cores are developed as an embedded peripheral and as a scalable, parallel hardware accelerator. Next, in “A high throughput hardware architecture for the H.264/AVC half-pixel motion estimation targeting high definition videos,” M. M. Correa et al. present a highperformance hardware implementation for the H.264/AVC half-pixel motion estimation that targets high-definition videos. This design can process very high-definition videos like QHDTV (3840 × 2048) in real time (30 frames per second). E. C. Pedrino et al. in the area of computer vision presented “A genetic programming approach to reconfigure a morphological image processing architecture.” They report the development of a reconfigurable architecture using logical, arithmetic, and morphological instructions generated automatically by a genetic programming approach. Related to cryptographic algorithms, G. Perin et al., in “Montgomery modular multiplication on reconfigurable hardware: systolic versus multiplexed implementation,” develop two Montgomery modular multiplication architectures: a systolic and a multiplexed circuit with competitive results. Finally, A. G. Silva-Filho et al. presented an advanced EDA tool in the paper “An ESL approach for energy consumption analysis of cache memories in SoC platforms” in the area of design space exploration. They propose an electronicsystem-level (ESL) approach for system modeling and cache energy consumption analysis of SoCs.

Quality of a scanned ink-on-paper fingerprint (or offline fingerprint) is generally poorer than that of online fingerprint and therefore, it requires an efficient enhancement method. This paper presents an algorithm to enhance the offline fingerprints using adaptive techniques. The algorithm handles certain problems that are specific to this type of fingerprints. It involves fingerprint extraction, intra-image classification followed by adaptive region enhancement of classified regions. It has been tested on IIT Kanpur database. The results show that the proposed algorithm performs well in enhancing ink-on-paper fingerprints.

The No-Free Lunch theorem (Wolpert and Macready 1997) provides an important limitation of global optimization algorithms. Improving an algorithm for one class of problem is likely to make it perform more poorly for other problems. This may explain the much larger number of popular global optimizat ion algorithms compared to local optimization algorithms. This means that when we propose a new or improved global optimization algorithm, it should be targeted towards a particular application or set of applications rather than tested against a fixed set of problems. For a paper describing such an algorithm for Structural and Multidisciplinary Optimization, it should target an application including solid or fluid mechanics models that is not solved well by existing algorithms. This has become difficult to demonstrate because of the large number of available global optimization algorithms. Problems where the objective function and constraints are differentiable are not likely to satisfy the above requirement. This is because gradient based local optimizers are very efficient in finding solutions for such problems even with large number of design variables. Unless the number of local optima is very high, multi-start execution of local search techniques is likely to yield a solution more efficiently than a global search algorithm. Problems with discrete or combinatorial design variables are more likely to satisfy the requirement. Finally, many global search algorithms are stochastic in nature. This means that to demonstrate their effectiveness, one must run them multiple times so as to establish the statistical distribution of the solution as function of the design parameters.

Solitaire Cipher was designed by Bruce Schneier as a Pen and Paper Cipher in 1999. It is designed for field agents to communicate secret messages without having electricity or sophisticated electronics equipment. Its main advantage is that the natural randomness of a shuffled deck of cards contributes to its strength and security. However, the main disadvantage of solitaire is that its output keystream is not truly random, its bias i.e. the probability of getting two keystream characters twice in a row is about 1/22.5 instead of 1/26 as identified by Mirdek. This is due to the fact that Solitaire is a non-reversible cipher and it exhibits bias during key generation. To overcome this weakness, we have proposed a revised Solitaire algorithm to improve the randomness in the output keystream and also to reduce the element of biasness while the cipher remains non-reversible. We have also introduced a new and easy mechanism to generate initial deck setting (key) from a passphrase. The proposed cipher is implemented in MATLAB and the generated keystream is tested for randomness.

Neural networks (NNs) have proven to be a very powerful tool both for one-dimensional (1D) and two-dimensional (2D) direction of arrival (DOA) estimation. By avoiding complex and time-consuming mathematical calculations, NNs estimate DOAs almost instantaneously. This feature makes them very convenient for real-time applications. Further, unlike the well known MUSIC algorithm, neural network-based models provide accurate directions without additional calibration procedure of antenna array and a priori knowledge of the number of sources. In this review paper, the results achieved by the research group at the Faculty of Electronic Engineering in Nis are presented. The problem of DOA estimation of narrowband signals impinging upon different configurations of antenna arrays is addressed. Both Multi-Layer Perceptron (MLP) and Radial Basis Function (RBF) neural networks are considered, and their advantages and disadvantages are discussed. To improve the resolution of DOA estimates, sectorization model is introduced. As shown in this work, neural network-based models demonstrate high-resolution localization capabilities and much better efficiency than the MUSIC.