Electrical and Computer Engineering

As a structured document, Braille is the most common means of reading and study for visually handicapped people. The need for converting Braille documents into a computer-readable format has motivated research into the implementation of Braille recognition systems. The main theme of this research is to propose robust probabilistic approaches to different steps of Braille Recognition. The method is meant to be very general in terms of being independent of those parameters of the Braille document such as skewness, scale, and spacing of the page, lines, and characters. For a given Braille document, a statistical method is proposed for estimating the scaling, spacing, and skewness parameters, whereby the detected dots of the Braille document are modeled using a parameterized probability density function. Skewness, scaling, and line spacing are estimated as a solution of a maximum-likelihood (ML) problem, which is solved using expectation maximization. Based on those parameters, each line of the Braille document is extracted, and each of three rows of individual lines is separated based on the vertical projection of the Braille dots. Finally, a scale-independent automatic document gridding procedure is proposed for dot localization and character detection based on a hidden Markov model.

Classi¯cation is an important ¯eld in machine learning and pattern recognition. Amongst various types of classi¯ers such as nearest neighbor, neural network and Bayesian classi¯ers, support vector machine (SVM) is known as a very powerful classi¯er. One of the advantages of SVM in comparison with the other methods, is its e±cient and adjustable generalization capability. The performance of SVM classi¯er depends on its parameters, specially regularization parameter C, that is usually selected by cross-validation. Despite its generalization, SVM su®ers from some limitations such as its considerable low speed training phase. Cross-validation is a very time consuming part of training phase, because for any candidate value of the parameter C, the entire process of training and validating must be repeated completely. In this paper, we propose a novel approach for early stopping of the SVM learning algorithm. The proposed early stopping occurs by integrating the validation part into the optimization part of the SVM training without losing any generality or degrading performance of the classi¯er. Moreover, this method can be considered in conjunction with the other available accelerator methods since there is not any dependency between our proposed method and the other accelerator ones, thus no redundancy will happen. Our method was tested and veri¯ed on various UCI repository datasets and the results indicate that this method speeds up the learning phase of SVM without losing any generality or a®ecting the ¯nal model of classi¯er.

— Ant meeting problem is a distributed problem that its purpose is to find an efficient path for ants to meet each other. Actually, ant robots suffer from the low memory and their restricted computational power. They communicate with their environment by leaving some symptoms like pheromones in their surroundings. In the problem of ants meeting, the ants should reach together in order to cooperate in an intelligent algorithm. In the common type of this problem, they start to search the environment without any knowledge about their locations, directions, obstacles and any other information. They can just leave some signs on the environment which shows their tracks. These marks are the only way of communication between ants. Also, these marks show the information about the ant which has passed the track, and every ant knows whose mark is that. In this paper, a new heuristic approach is proposed for the distributed ants meeting problem based on A* algorithm which guaranties the ants meeting in a finite time. Experimental results show that the presented algorithm is more efficient than previous ones.

It is well known that in many situations combining diverse classifiers can improve the performance of a classification system. In this paper, a new histogram based lip segmentation technique is proposed considering local kernel histograms in different illumination invariant colour spaces. The histogram is computed in local areas using two Gaussian kernels; one in the colour space and the other in the spatial domain. Using the estimated histogram, the posterior probability associated to non-lip class is then computed for each pixel. This process is performed considering different colour spaces. A weighted averaging method is then used for fusing the posterior probability values. As the result a new score is obtained which is used for labelling the pixels as lip or non-lip. The advantage of the proposed method is that the segmentation process is totally unsupervised. So, the method is robust against different variations such as variation in lip shape, skin colour, facial hair, illumination, etc. Moreover, an improved performance is achieved by fusing colour information.

In software production process, quality assurance resources are limited by time and cost. In order to achieve high quality, managers need to detect the defect prone parts of code and allocate the resources to them. So far, researchers have used some methods such as complexity metrics, design metrics and network measures for software defect prediction. Although these methods are somewhat efficient, still there isn't a global method for it. In this paper we have presented two new definitions which are dependency tree and circular dependency. Dependency tree covers all of the direct and indirect dependencies circular dependency evaluates all of the cyclic chains of dependency for each component. What has been differed our work from the previous related works is that we have presented that using transitive dependencies is efficient in defect prediction.

Design patterns describe good solutions to common and reoccurring problems in program design. Applying design patterns in software design and implementation have significant effects on software quality metrics such as flexibility, usability, reusability, scalability and robustness. Applying design patterns in software systems does not have a specific rule. It is possible to apply two design patterns for a specific problem, so a decision should be taken on choosing one of the two design patterns. In this paper we propose a solution to choose and apply patterns according to the required quality metrics.

To exploit multiuser spatial cooperative diversity, relay nodes can assist the transmission of signal from multiple sources to their corresponding destinations in order to combat the noise, the channel fading and the cross-interferences. In this work, we consider a network consisting of N source-destination pairs and a set of M relay nodes. We assume that a fixed total budget of transmit energy is available for allotment among all transmitters, i.e., the source and relay nodes. For such network, we consider the time division Amplify-and-Forward (AF) scheme in 2N time slots, and propose an algorithm to optimize the beamformers and the powers allocated to all transmit nodes. The proposed method maximizes the network sum capacity assuming that the channels state information (CSI) is known. Numerical simulations illustrate the significant improvement obtained under power constraint. This betterment is more pronounced for low power regimes.

The throughput of a multi-user broadcast opportunistic communications system is highly affected by the accuracy of the available channel state information (CSI) at the transmitter side. It is a questionable assumption to have almost perfect knowledge of the user SNRs at the transmitter due to the error in the channel measurement, or due to reporting feedback errors from users to the transmitter. As a result such errors, the system can not gain its maximum throughput. In this paper, we investigate the performance of opportunistic communications system considering imperfect SNR reports. We show that in order to maximize the throughput, the transmitter needs to estimate the channel state information. The imposition of a correlation constraint will prevent the estimate of channel from fluctuating too widely. Employ Kalman filter, we estimate the channel state information and show that the throughput is significantly improved.

In this paper, we have focused on orthogonal non-regenerating cooperation for networks with a source-destination pair and a number of relay nodes. We proposed two power allocation problems based on the instantaneous and average information about the channel gains to maximize either the instantaneous or average signal-to-noise ratio at the destination. We also assumed that the network total energy consumption is fixed. Since the suggested optimization problems are complicated to solve, we developed an iterative algorithm that converts the problems into a generalized geometric programming problem which derives the optimal solution very fast. The simulation results confirmed that the proposed power allocation schemes improve the system performance.

Imperfect spectrum sensing in overlay spectrum
sharing may cause collision between primary and secondary
transmission resulting in performance degradation for primary
users. To satisfy the quality of service requirement imposed by
the primary service, the maximum probability of collision has
to be kept below a given threshold. The collision, if occurs, also
results in increasing error rate in the secondary system. What
can be done if the secondary service needs better error rate than
the value imposed by the collision constraint in the primary
network? From secondary user perspective, this can be taken
care of by employing channel coding techniques at the expense
of effective rate reduction. Alternatively, one can increase the
sensing time to reduce the collision probability at the expense
of reducing the data transmission time. This also reduces the
effective data rate for secondary users. In this letter, we compare
these two approaches and demonstrate a situation where using
rate-compatible Low-Density Parity-Check codes, the effective
data rate for the coded case can be significantly more than that of
the case without channel coding while exhibiting a considerably
better performance. We then discuss necessary condition for such
an advantage and come up with cases where increasing sensing
time is preferred over adopting channel coding.

"In this paper we consider an overlay cognitive
network in which to guarantee a Quality of Service (QoS)f o rthe
primary users, a maximum probability of collision is enforced to
secondary service. When collision, caused by imperfect spectrum
sensing, happens, it results in increasing error rate in both
primary and secondary systems. While such degradation in
primary service conforms with QoS requirements of that service,
this may not be acceptable from secondary user perspective. Such
degradation can be dealt with in secondary service either by
employing channel coding techniques at the expense of effective
rate reduction or increasing the sensing time to reduce the
collision probability. This results in the reduction of the data
transmission time which also results in the reduction of effective
data ratef o rsecondary users. In this paper, we compare these two
cases and show that using rate-compatible Low-Density ParityCheck codes, the effective data ratef o r the coded case can be
significantly more than that of the case without channel coding
while exhibiting a considerably better performance."

The purpose of the functional analysis is to define a clear picture of the scope, architecture, and functionality of substation automation systems, as addressed by this security profile. The real-world specific performance of substation automation system functions varies in terms of function, scope, and technology from device to device and component to component among different system offerings and deployments. However, this profile approaches the problem by defining a set of abstract roles that capture essential functionality that may be realized through a variety of implementations. For example, the functions of the Substation User Interface role may be performed by a stand-alone component, or rolled into a platform that also performs many of the remote access functions as defined in the Proxy role. Conversely, some implementations may have the decision-making functionality of the Control Authority role distributed among several devices that also implement the Substation User Interface, the Proxy, and possibly even a Control Application. Regardless, this profile defines roles in such a way that the logical architecture and state machine functionality may be used to represent a wide variety of real-world implementations.

Research infrastructures play an increasingly essential role in scientific research. They provide rich data sources for scientists, such as services and software packages, via catalog and virtual research environments. However, such research infrastructures are typically domain-specific and often not connected. Accordingly, researchers and practitioners face fundamental challenges introduced by fragmented knowledge from heterogeneous, autonomous sources with complicated and uncertain relations in particular research domains. Additionally, the exponential growth rate of knowledge in a specific domain surpasses human experts’ ability to formalize and capture tacit and explicit knowledge efficiently. Thus, a knowledge management system is required to discover knowledge effectively, automate the knowledge acquisition based on artificial intelligence approaches, integrate the captured knowledge, and deliver consistent knowledge to agents, research communities, and end-users. In this stud...

Software producing organisations face the challenge of including new technology in their products, such as cloud technologies and database management systems. As software architects and senior developers are not experts in this domain, they need to consult external experts or acquire the knowledge themselves. Software production, therefore, is a suitable domain to deploy decision support systems, that intelligently support these decision-makers in selecting the desirable technology for their product. We present a decision support system that supports decision-makers in choosing the most suitable database technology. The case studies and experts confirm that the approach increases insight into the selection process, provides a richer prioritised option list than if they had done their research independently, besides reduces the time and cost of the decision-making process.

Blockchain technology has received significant attention recently, as it offers a reliable decentralized infrastructure for all kinds of business transactions. Software-producing organizations are increasingly considering blockchain technology for inclusion into their software products. Selecting the best fitting blockchain platform requires the assessment of its functionality, adaptability, and compatibility to the existing software product. Novice software developers and architects are not experts in every domain, so they should either consult external experts or acquire knowledge themselves. The decision-making process gets more complicated as the number of decision-makers, alternatives, and criteria increases. Hence, a decision model is required to externalize and organize knowledge regarding the blockchain platform selection context. Recently, we designed a decision support system to use such decision models to support decision-makers with their technology selection problems in software production. In this article, we introduce a decision model for the blockchain platform selection problem. The decision model has been evaluated through three real-world case studies at three software-producing organizations. The case-study participants asserted that the approach provides significantly more insight into the blockchain platform selection process, provides a richer prioritized option list than if they had done their research independently, and reduces the time and cost of the decision-making process.

Context: Software architecture is a knowledge-intensive field. One mechanism for storing architecture knowledge is the recognition and description of architectural patterns. Selecting architectural patterns is a challenging task for software architects, as knowledge about these patterns is scattered among a wide range of literature. Method: We report on a systematic literature review, with the aim of building a decision model for the architectural pattern selection problem. Moreover, twelve experienced practitioners at softwareproducing organizations evaluated the usability and usefulness of the extracted knowledge. Results: An overview is provided of 29 patterns and their effects on 40 quality attributes. Furthermore, we report in which systems the 29 patterns are applied and in which combinations. The practitioners confirmed that architectural knowledge supports software architects with their decision-making process to select a set of patterns for a new problem. We investigate the potential trends among architects to select patterns. Conclusion: With the knowledge available, architects can more rapidly select and eliminate combinations of patterns to design solutions. Having this knowledge readily available supports software architects in making more efficient and effective design decisions that meet their quality concerns.