Search

[EN] This paper presents a new method for fraud detection in credit cards based on exploiting the dynamics of the card transactions. We hypothesize different behavior models in the use of the card between legitimate clients and fraudsters that are registered in the sequential pattern that follows the transactions. The method considers analyses in subspaces defined by two or three variables recorded in the transactions. From these subspaces, several dynamic features, such as transaction velocity and acceleration, are estimated as input vectors for a classification process. Linear and quadratic discriminant analysis and random forest are implemented as single classifiers. All the single classification results obtained for each of the subspaces are late fused to obtain an overall result using alpha integration algorithm. The proposed method was evaluated using a subset of real data with a very low fraud to legitimate transaction ratio. We demonstrated that the temporal dependence of card transactions exploited in different subspaces and fused to give an overall result improves the detection accuracy of fraud detection in credit cards.

[EN] The illegal dumpings in sewerage can cause problems in the Wastewater treatment plants. In this paper, we propose a system for detecting these illegal dumpings. We use conductivity sensor for detecting the change in the conductivity of water. Because this change may be due to a dump. The system is based on two coils, a coil is powered by a sinus-wave and induced the other coil. To prevent damaged for water in the copper we encapsulate the coils in a PVC tube. These coils are connected to flyport to send the values and generate alarms. We test with different configurations of coils with encapsulation of 3 and 1 mm. In the different prototypes tested have been seen that with the 3 mm encapsulation no significant differences are observed. The better prototypes are based on 40 spires of the powered coil and 80 spires in the induced coil. The best difference between samples of 0 and 40 g/l of table salt are in the prototype when the spires are distributed in 1 layer (6.06 V). Another prototype test with the same number of turns but distributed in 4 layers has a difference of 4.10 V. But during the verification phase we verified that this last prototype presents a relative error of 2.54% compared to 6.55% of the prototype distributed in a layer

[EN] Monitoring the execution of real-time systems has many advantages, it is not only useful to understand the behaviour of an application but also to find unfulfilled timing constraints in an implementation. However, real-time operating systems usually do not include the tracing tools to observe the behaviour during the execution. This paper presents the art2kitekt runtime monitoring tool, used to measure and to visualise the temporal characteristics of a real-time application. To demonstrate the functionality of the tool, the behaviour of an RTEMS-based application running over a Xilinx Zynq UltraScale+ is observed.

[EN] An optimization procedure for the design of miniaturized substrate integrated quasi-lumped filters based on aggressive space mapping techniques is presented in this paper. A gradient-descent approach based on a lossy coarse model is employed. Thus, a 3-pole bandpass filter response centered at 10 GHz is designed, manufactured and measured, showing the validity of the technique even if strong dependencies between the different electrical and physical parameters are present.

[EN] Electric Vehicles (EVs) are growing attention fortheir higher efficiency and less-polluting specifications. However,a massive introduction of EVs could lead to several issues inpower systems. Several authors have proposed various smartcharging approaches. But, these approaches could not be ap-propriately implemented without knowing the charging behaviorof the EV customers. Thus, this paper proposes an EV chargingload prediction for the particular case of Quito, Ecuador. Thisforecasting is performed based on data obtained from a GPSsystem and on statistical methods.

This paper analyzes the feasibility of using a simple diffraction model to compute the blocking of the human body to millimeter wave radio frequencies in indoor environments. The model makes a set of approximations that are evaluated to determine the applicability limits of such simplified approach, in particular for the human body blockage case. The work presented here: (1) describes briefly the mathematical support that is used to model the concealment using the Knife-Edge model, (2) identifies the potential simplifications applicable to the mathematical model implementation that allow a 3D geometric human body to be modelled with simple 2D shapes, (3) characterizes the polarization influence on the mm-wave blocking for such simplified human body models.

Aerial base stations have been recently considered in the deployment of wireless networks. Finding the optimal position for one or multiple aerial base stations is a complex problem tackled by several works. However, just a few works consider the mobility of the users which makes necessary an online optimization to follow the changes in the scenario where the optimization is performed. This paper deals with the online optimization of an aerial base station placement considering different types of users mobility and three algorithms: a Q-learning technique, a Gradient-based solution and a Greedy-search solution. Our objective is to minimize in an urban environment the path loss of the user at street level with the highest path loss. Simulation results show that the performance of the three methods is similar when a high number of users move randomly and uniformly around the scenario under test. Nevertheless, in some situations when the number of users is reduced or when the users move together in a similar direction, both Gradient and Greedy algorithms present a significantly better performance than the Q-learning method.

Aerial base stations have been recently considered in the deployment of wireless networks. Finding the optimal position for one or multiple aerial base stations is a complex problem tackled by several works. However, just a few works consider the mobility of the users which makes necessary an online optimization to follow the changes in the scenario where the optimization is performed. This paper deals with the online optimization of an aerial base station placement considering different types of users mobility and three algorithms: a Q-learning technique, a Gradient-based solution and a Greedy-search solution. Our objective is to minimize in an urban environment the path loss of the user at street level with the highest path loss. Simulation results show that the performance of the three methods is similar when a high number of users move randomly and uniformly around the scenario under test. Nevertheless, in some situations when the number of users is reduced or when the users move together in a similar direction, both Gradient and Greedy algorithms present a significantly better performance than the Q-learning method.

This paper analyzes the feasibility of using a simple diffraction model to compute the blocking of the human body to millimeter wave radio frequencies in indoor environments. The model makes a set of approximations that are evaluated to determine the applicability limits of such simplified approach, in particular for the human body blockage case. The work presented here: (1) describes briefly the mathematical support that is used to model the concealment using the Knife-Edge model, (2) identifies the potential simplifications applicable to the mathematical model implementation that allow a 3D geometric human body to be modelled with simple 2D shapes, (3) characterizes the polarization influence on the mm-wave blocking for such simplified human body models.

Wireless Body Area Networks (WBANs) have increased the attention of the research community for the next generation wireless medical devices. Among others, Wireless Capsule Endoscopy (WCE) aims to transmit better quality images. For this, the Ultra Wideband (UWB) frequency band is becoming a good alternative to currently allocated frequencies for in-body networks, allowing higher data rate and having a low power transmission. Common channel characterization in WBANs are performed in frequency domain, i.e., analyzing the received power as a function of frequency. Nevertheless, indepth studies in delay domain analyzing the impulse response of the channel are barely considered in current literature. In this paper, an initial study in delay domain, i.e., the Power Delay Profile (PDP) characteristics, is performed. Moreover, a comparison between the channel response in frequency and delay domain is performed. This work gives an insightful view of the impulse response of the channel for in-body to on-body communications. For that, an extensive campaign of phantom measurements and software simulations are conducted.

Wireless Body Area Networks (WBANs) have increased the attention of the research community for the next generation wireless medical devices. Among others, Wireless Capsule Endoscopy (WCE) aims to transmit better quality images. For this, the Ultra Wideband (UWB) frequency band is becoming a good alternative to currently allocated frequencies for in-body networks, allowing higher data rate and having a low power transmission. Common channel characterization in WBANs are performed in frequency domain, i.e., analyzing the received power as a function of frequency. Nevertheless, indepth studies in delay domain analyzing the impulse response of the channel are barely considered in current literature. In this paper, an initial study in delay domain, i.e., the Power Delay Profile (PDP) characteristics, is performed. Moreover, a comparison between the channel response in frequency and delay domain is performed. This work gives an insightful view of the impulse response of the channel for in-body to on-body communications. For that, an extensive campaign of phantom measurements and software simulations are conducted.

Localization for Wireless Capsule Endoscopy (WCE) in the Ultra-Wideband frequency band is a very active field of investigation due to its potential advantages in future endoscopy applications. Received Signal Strength (RSS) based localization is commonly preferred due to its simplicity. Previous studies on Ultra-Wideband (UWB) RSS-based localization showed that the localization accuracy depends on the average ranging error related to the selected combination of receivers, which not always is the one experiencing the highest level of received power. In this paper the tendency of the localization error is further investigated through supplementary software simulations and previously conducted laboratory measurements. Two-dimensional (2D) and three-dimensional (3D) positioning are performed and the trend of the localization error compared in both cases. Results shows that the distribution of the selected path loss values, corresponding to the receivers used for localization, around the in-body position to estimate also affects the localization accuracy.

[EN] Optical coherence tomography (OCT) is an important imaging modality that is used frequently to monitor the state of retinal layers both in humans and animals. Automated OCT analysis in rodents is an important method to study the possible toxic effect of treatments before the test in humans. In this paper, an automatic method to detect the most significant retinal layers in rat OCT images is presented. This algorithm is based on an encoder-decoder fully convolutional network (FCN) architecture combined with a robust method of post-processing. After the validation, it was demonstrated that the proposed method outperforms the commercial Insight image segmentation software. We obtained results (averaged absolute distance error) in the test set for the training database of 2.52±0.80 µm. In the predictions done by the method, in a different database (only used for testing), we also achieve the promising results of 4.45 ± 3.02 µm.

[EN] Optical coherence tomography (OCT) is an important imaging modality that is used frequently to monitor the state of retinal layers both in humans and animals. Automated OCT analysis in rodents is an important method to study the possible toxic effect of treatments before the test in humans. In this paper, an automatic method to detect the most significant retinal layers in rat OCT images is presented. This algorithm is based on an encoder-decoder fully convolutional network (FCN) architecture combined with a robust method of post-processing. After the validation, it was demonstrated that the proposed method outperforms the commercial Insight image segmentation software. We obtained results (averaged absolute distance error) in the test set for the training database of 2.52±0.80 µm. In the predictions done by the method, in a different database (only used for testing), we also achieve the promising results of 4.45 ± 3.02 µm.

Localization for Wireless Capsule Endoscopy (WCE) in the Ultra-Wideband frequency band is a very active field of investigation due to its potential advantages in future endoscopy applications. Received Signal Strength (RSS) based localization is commonly preferred due to its simplicity. Previous studies on Ultra-Wideband (UWB) RSS-based localization showed that the localization accuracy depends on the average ranging error related to the selected combination of receivers, which not always is the one experiencing the highest level of received power. In this paper the tendency of the localization error is further investigated through supplementary software simulations and previously conducted laboratory measurements. Two-dimensional (2D) and three-dimensional (3D) positioning are performed and the trend of the localization error compared in both cases. Results shows that the distribution of the selected path loss values, corresponding to the receivers used for localization, around the in-body position to estimate also affects the localization accuracy.

[EN] In recent years, localization for capsule endoscopy applications using Ultra-Wideband (UWB) technology has become an attractive field of study due to its potential benefits for patients. Performance analysis of RF-based localization techniques are very limited in literature. Most of the available studies rely on software simulations using digital human models. Nonetheless, no realistic studies based on in-vivo measurements has been reported yet. This paper investigates the performance of RSS-based technique for three-dimensional (3D) localization in the UWB frequency band. Impact of receivers selection as well as of the evaluated path loss model on the localization accuracy is investigated. Results obtained through CST-based simulations and from recently conducted in-vivo measurements are presented and compared.

[EN] Wireless Body Area Networks (WBANs) are a promising technology for medical purposes. Currently the WBAN are classified into: implanted (in-), surface (on-) or outside (off-) body communications regarding the location of the devices with reference to the human body. The Ultra Wide-Band (UWB) frequency band is growing as a band of interest for implanted communications because of its high data rate and low power consumption among other benefits. Software simulations, in-vivo measurements and experimental phantom measurements are common methods to properly characterize the propagation channel. Nevertheless, up to now, experimental phantoms measurements presented in the literature show some inconveniences, i.e., the accuracy of the phantoms compared with the real human tissues or the testbed used for the measurements. This paper aims at overcoming these issues using accurate phantoms designed for the purpose of implanted communications in the UWB frequency band. In addition, a multilayer phantom container was developed. This container has capacity for two different phantoms, emulating a heterogeneous propagation medium for in-body measurements. Moreover, a novel setup was built for in-body phantom measurements. As a result, an experimental path loss model is presented from the measurements obtained with phantoms. Besides, software simulations mimicking the experimental setup are performed in order to validate the previous results obtained

[EN] Wireless Body Area Networks (WBANs) are a promising technology for medical purposes. Currently the WBAN are classified into: implanted (in-), surface (on-) or outside (off-) body communications regarding the location of the devices with reference to the human body. The Ultra Wide-Band (UWB) frequency band is growing as a band of interest for implanted communications because of its high data rate and low power consumption among other benefits. Software simulations, in-vivo measurements and experimental phantom measurements are common methods to properly characterize the propagation channel. Nevertheless, up to now, experimental phantoms measurements presented in the literature show some inconveniences, i.e., the accuracy of the phantoms compared with the real human tissues or the testbed used for the measurements. This paper aims at overcoming these issues using accurate phantoms designed for the purpose of implanted communications in the UWB frequency band. In addition, a multilayer phantom container was developed. This container has capacity for two different phantoms, emulating a heterogeneous propagation medium for in-body measurements. Moreover, a novel setup was built for in-body phantom measurements. As a result, an experimental path loss model is presented from the measurements obtained with phantoms. Besides, software simulations mimicking the experimental setup are performed in order to validate the previous results obtained

[EN] In recent years, localization for capsule endoscopy applications using Ultra-Wideband (UWB) technology has become an attractive field of study due to its potential benefits for patients. Performance analysis of RF-based localization techniques are very limited in literature. Most of the available studies rely on software simulations using digital human models. Nonetheless, no realistic studies based on in-vivo measurements has been reported yet. This paper investigates the performance of RSS-based technique for three-dimensional (3D) localization in the UWB frequency band. Impact of receivers selection as well as of the evaluated path loss model on the localization accuracy is investigated. Results obtained through CST-based simulations and from recently conducted in-vivo measurements are presented and compared.

[EN] In recent years, localization for capsule endoscopy applications using Ultra-Wideband (UWB) technology has become an attractive field of study due to its potential benefits for patients. Performance analysis of RF-based localization techniques are very limited in literature. Most of the available studies rely on software simulations using digital human models. Nonetheless, no realistic studies based on in-vivo measurements has been reported yet. This paper investigates the performance of RSS-based technique for three-dimensional (3D) localization in the UWB frequency band. Impact of receivers selection as well as of the evaluated path loss model on the localization accuracy is investigated. Results obtained through CST-based simulations and from recently conducted in-vivo measurements are presented and compared.

[EN] Wireless Body Area Networks (WBANs) are a promising technology for medical purposes. Currently the WBAN are classified into: implanted (in-), surface (on-) or outside (off-) body communications regarding the location of the devices with reference to the human body. The Ultra Wide-Band (UWB) frequency band is growing as a band of interest for implanted communications because of its high data rate and low power consumption among other benefits. Software simulations, in-vivo measurements and experimental phantom measurements are common methods to properly characterize the propagation channel. Nevertheless, up to now, experimental phantoms measurements presented in the literature show some inconveniences, i.e., the accuracy of the phantoms compared with the real human tissues or the testbed used for the measurements. This paper aims at overcoming these issues using accurate phantoms designed for the purpose of implanted communications in the UWB frequency band. In addition, a multilayer phantom container was developed. This container has capacity for two different phantoms, emulating a heterogeneous propagation medium for in-body measurements. Moreover, a novel setup was built for in-body phantom measurements. As a result, an experimental path loss model is presented from the measurements obtained with phantoms. Besides, software simulations mimicking the experimental setup are performed in order to validate the previous results obtained

[Otros] ATSC 3.0, the latest Digital Terrestrial Television (DTT) standard, allows a higher spectral efficiency and/or a transmission robustness with Multiple-Input Multiple-Output (MIMO) technology compared to existing single-antenna DTT networks. Regarding MIMO channel estimation, two pilot encoding algorithms known as Walsh-Hadamard encoding and Null pilot encoding are possible in ATSC 3.0. The two MIMO pilot algorithms are standardized so as to have the same pilot positions and the same pilot boosting as SISO, but the performance has not been evaluated. This paper focuses on the performance evaluation of the two MIMO pilot encoding algorithms in ATSC 3.0 using physical layer simulations. Results can be used as guidelines or recommended practices to broadcasters to select the MIMO pilot encoding algorithm that better suits their service requirments. Several channel estimation algorithms have been evaluated in both mobile and fixed reception conditions. The simulation results show that Null pilot encoding provides slightly better performance than WalshHadamard encoding for fixed reception but worse performance for mobile reception, especially at high signal-to-noise ratios.

[Otros] ATSC 3.0, the latest Digital Terrestrial Television (DTT) standard, allows a higher spectral efficiency and/or a transmission robustness with Multiple-Input Multiple-Output (MIMO) technology compared to existing single-antenna DTT networks. Regarding MIMO channel estimation, two pilot encoding algorithms known as Walsh-Hadamard encoding and Null pilot encoding are possible in ATSC 3.0. The two MIMO pilot algorithms are standardized so as to have the same pilot positions and the same pilot boosting as SISO, but the performance has not been evaluated. This paper focuses on the performance evaluation of the two MIMO pilot encoding algorithms in ATSC 3.0 using physical layer simulations. Results can be used as guidelines or recommended practices to broadcasters to select the MIMO pilot encoding algorithm that better suits their service requirments. Several channel estimation algorithms have been evaluated in both mobile and fixed reception conditions. The simulation results show that Null pilot encoding provides slightly better performance than WalshHadamard encoding for fixed reception but worse performance for mobile reception, especially at high signal-to-noise ratios.

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., [EN] Two methods are presented to improve word confidence scores for Line-Level Query-by-String Lexicon-Free Keyword Spotting (KWS) in handwritten text images. The first one approaches true relevance probabilities by means of computations directly carried out on character lattices obtained from the lines images considered. The second method uses the same character lattices, but it obtains relevance scores by first computing frame-level character sequence scores which resemble the word posteriorgrams used in previous approaches for lexicon-based KWS. The first method results from a formal probabilistic derivation, which allow us to better understand and further develop the underlying ideas. The second one is less formal but, according with experiments presented in the paper, it obtains almost identical results with much lower computational cost. Moreover, in contrast with the first method, the second one allows to directly obtain accurate bounding boxes for the spotted words.

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., [EN] The H-KWS 2016, organized in the context of the ICFHR 2016 conference aims at setting up an evaluation framework for benchmarking handwritten keyword spotting (KWS) examining both the Query by Example (QbE) and the Query by String (QbS) approaches. Both KWS approaches were hosted into two different tracks, which in turn were split into two distinct challenges, namely, a segmentation-based and a segmentation-free to accommodate different perspectives adopted by researchers in the KWS field. In addition, the competition aims to evaluate the submitted training-based methods under different amounts of training data. Four participants submitted at least one solution to one of the challenges, according to the capabilities and/or restrictions of their systems. The data used in the competition consisted of historical German and English documents with their own characteristics and complexities. This paper presents the details of the competition, including the data, evaluation metrics and results of the best run of each participating methods.

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., [EN] The H-KWS 2016, organized in the context of the ICFHR 2016 conference aims at setting up an evaluation framework for benchmarking handwritten keyword spotting (KWS) examining both the Query by Example (QbE) and the Query by String (QbS) approaches. Both KWS approaches were hosted into two different tracks, which in turn were split into two distinct challenges, namely, a segmentation-based and a segmentation-free to accommodate different perspectives adopted by researchers in the KWS field. In addition, the competition aims to evaluate the submitted training-based methods under different amounts of training data. Four participants submitted at least one solution to one of the challenges, according to the capabilities and/or restrictions of their systems. The data used in the competition consisted of historical German and English documents with their own characteristics and complexities. This paper presents the details of the competition, including the data, evaluation metrics and results of the best run of each participating methods.

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., [EN] This paper describes the Handwritten Text Recognition (HTR) competition on the READ dataset that has been held in the context of the International Conference on Frontiers in Handwriting Recognition 2016. This competition aims to bring together researchers working on off-line HTR and provide them a suitable benchmark to compare their techniques on the task of transcribing typical historical handwritten documents. Two tracks with different conditions on the use of training data were proposed. Ten research groups registered in the competition but finally five submitted results. The handwritten images for this competition were drawn from the German document Ratsprotokolle collection composed of minutes of the council meetings held from 1470 to 1805, used in the READ project. The selected dataset is written by several hands and entails significant variabilities and difficulties. The five participants achieved good results with transcriptions word error rates ranging from 21% to 47% and character error rates rating from 5% to 19%.

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Photonic interconnects are a promising solution for the so-called communication bottleneck in current Chip Multiprocessor (CMPs) architectures. This technology presents an inherent low-latency and power consumption almost independent of communication distance, which are really desirable features in future Networks on Chip for next CMPs generations. However, since nanophotonic technology is still growing and therefore in an immature state, current simulators of detailed systems may not provide accurate models of photonic components. In this context, non-representative results are obtained when unaccurate photonic models are assumed. This paper summarizes all of the components that conform a fully operative photonic NoC and presents their current state of the art. Moreover, we evaluate a realistic photonic network that consists of two photonic rings and a token-based arbitration mechanism and compare it against a non-realistic model. In addition, both realistic and non-realistic schemes are valuated under different configurations varying the number of wavelengths that photonic waveguides employ. The experimental results show that the non-realistic NoC presents up 6× network latency deviation with respect to the accurate model. This deviation is translated into a performance deviation higher than 10% in several applications studied, which demonstrates the importance of accurate models when simulating current technologies under development like nanophotonics. Finally, a power consumption model of the realistic photonic network is presented. The results show that the overall photonic network power consumption grows with the number of wavelengths per waveguide since the number of required modulators and receivers becomes higher. In this way, the proposed realistic photonic network, which employs only two wavelengths for arbitration and destination selection tasks, increases its power consumption up to 3%, so network designs with more complex arbitration mechanisms must ta

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., [EN] This paper describes the Handwritten Text Recognition (HTR) competition on the READ dataset that has been held in the context of the International Conference on Frontiers in Handwriting Recognition 2016. This competition aims to bring together researchers working on off-line HTR and provide them a suitable benchmark to compare their techniques on the task of transcribing typical historical handwritten documents. Two tracks with different conditions on the use of training data were proposed. Ten research groups registered in the competition but finally five submitted results. The handwritten images for this competition were drawn from the German document Ratsprotokolle collection composed of minutes of the council meetings held from 1470 to 1805, used in the READ project. The selected dataset is written by several hands and entails significant variabilities and difficulties. The five participants achieved good results with transcriptions word error rates ranging from 21% to 47% and character error rates rating from 5% to 19%.

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., [EN] Two methods are presented to improve word confidence scores for Line-Level Query-by-String Lexicon-Free Keyword Spotting (KWS) in handwritten text images. The first one approaches true relevance probabilities by means of computations directly carried out on character lattices obtained from the lines images considered. The second method uses the same character lattices, but it obtains relevance scores by first computing frame-level character sequence scores which resemble the word posteriorgrams used in previous approaches for lexicon-based KWS. The first method results from a formal probabilistic derivation, which allow us to better understand and further develop the underlying ideas. The second one is less formal but, according with experiments presented in the paper, it obtains almost identical results with much lower computational cost. Moreover, in contrast with the first method, the second one allows to directly obtain accurate bounding boxes for the spotted words.

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Photonic interconnects are a promising solution for the so-called communication bottleneck in current Chip Multiprocessor (CMPs) architectures. This technology presents an inherent low-latency and power consumption almost independent of communication distance, which are really desirable features in future Networks on Chip for next CMPs generations. However, since nanophotonic technology is still growing and therefore in an immature state, current simulators of detailed systems may not provide accurate models of photonic components. In this context, non-representative results are obtained when unaccurate photonic models are assumed. This paper summarizes all of the components that conform a fully operative photonic NoC and presents their current state of the art. Moreover, we evaluate a realistic photonic network that consists of two photonic rings and a token-based arbitration mechanism and compare it against a non-realistic model. In addition, both realistic and non-realistic schemes are valuated under different configurations varying the number of wavelengths that photonic waveguides employ. The experimental results show that the non-realistic NoC presents up 6× network latency deviation with respect to the accurate model. This deviation is translated into a performance deviation higher than 10% in several applications studied, which demonstrates the importance of accurate models when simulating current technologies under development like nanophotonics. Finally, a power consumption model of the realistic photonic network is presented. The results show that the overall photonic network power consumption grows with the number of wavelengths per waveguide since the number of required modulators and receivers becomes higher. In this way, the proposed realistic photonic network, which employs only two wavelengths for arbitration and destination selection tasks, increases its power consumption up to 3%, so network designs with more complex arbitration mechanisms must ta

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., [EN] The H-KWS 2016, organized in the context of the ICFHR 2016 conference aims at setting up an evaluation framework for benchmarking handwritten keyword spotting (KWS) examining both the Query by Example (QbE) and the Query by String (QbS) approaches. Both KWS approaches were hosted into two different tracks, which in turn were split into two distinct challenges, namely, a segmentation-based and a segmentation-free to accommodate different perspectives adopted by researchers in the KWS field. In addition, the competition aims to evaluate the submitted training-based methods under different amounts of training data. Four participants submitted at least one solution to one of the challenges, according to the capabilities and/or restrictions of their systems. The data used in the competition consisted of historical German and English documents with their own characteristics and complexities. This paper presents the details of the competition, including the data, evaluation metrics and results of the best run of each participating methods.

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., [EN] This paper describes the Handwritten Text Recognition (HTR) competition on the READ dataset that has been held in the context of the International Conference on Frontiers in Handwriting Recognition 2016. This competition aims to bring together researchers working on off-line HTR and provide them a suitable benchmark to compare their techniques on the task of transcribing typical historical handwritten documents. Two tracks with different conditions on the use of training data were proposed. Ten research groups registered in the competition but finally five submitted results. The handwritten images for this competition were drawn from the German document Ratsprotokolle collection composed of minutes of the council meetings held from 1470 to 1805, used in the READ project. The selected dataset is written by several hands and entails significant variabilities and difficulties. The five participants achieved good results with transcriptions word error rates ranging from 21% to 47% and character error rates rating from 5% to 19%.

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., [EN] Two methods are presented to improve word confidence scores for Line-Level Query-by-String Lexicon-Free Keyword Spotting (KWS) in handwritten text images. The first one approaches true relevance probabilities by means of computations directly carried out on character lattices obtained from the lines images considered. The second method uses the same character lattices, but it obtains relevance scores by first computing frame-level character sequence scores which resemble the word posteriorgrams used in previous approaches for lexicon-based KWS. The first method results from a formal probabilistic derivation, which allow us to better understand and further develop the underlying ideas. The second one is less formal but, according with experiments presented in the paper, it obtains almost identical results with much lower computational cost. Moreover, in contrast with the first method, the second one allows to directly obtain accurate bounding boxes for the spotted words.

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Photonic interconnects are a promising solution for the so-called communication bottleneck in current Chip Multiprocessor (CMPs) architectures. This technology presents an inherent low-latency and power consumption almost independent of communication distance, which are really desirable features in future Networks on Chip for next CMPs generations. However, since nanophotonic technology is still growing and therefore in an immature state, current simulators of detailed systems may not provide accurate models of photonic components. In this context, non-representative results are obtained when unaccurate photonic models are assumed. This paper summarizes all of the components that conform a fully operative photonic NoC and presents their current state of the art. Moreover, we evaluate a realistic photonic network that consists of two photonic rings and a token-based arbitration mechanism and compare it against a non-realistic model. In addition, both realistic and non-realistic schemes are valuated under different configurations varying the number of wavelengths that photonic waveguides employ. The experimental results show that the non-realistic NoC presents up 6× network latency deviation with respect to the accurate model. This deviation is translated into a performance deviation higher than 10% in several applications studied, which demonstrates the importance of accurate models when simulating current technologies under development like nanophotonics. Finally, a power consumption model of the realistic photonic network is presented. The results show that the overall photonic network power consumption grows with the number of wavelengths per waveguide since the number of required modulators and receivers becomes higher. In this way, the proposed realistic photonic network, which employs only two wavelengths for arbitration and destination selection tasks, increases its power consumption up to 3%, so network designs with more complex arbitration mechanisms must ta

© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., CUDA is a technology developed by NVIDIA which provides a parallel computing platform and programming model for NVIDIA GPUs and compatible ones. It takes benefit from the enormous parallel processing power of GPUs in order to accelerate a wide range of applications, thus reducing their execution time. rCUDA (remote CUDA) is a middleware which grants applications concurrent access to CUDA-compatible devices installed in other nodes of the cluster in a transparent way so that applications are not aware of accessing a remote device. In this paper we present a demo which shows, in real time, the overhead introduced by rCUDA in comparison to CUDA when running image filtering applications. The approach followed in this work is to develop a graphical demo which contains both an appealing design and technical contents.

© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Power consumption is becoming an increasingly important component of processor design. As technology node shrinks both static and dynamic power become more relevant. This is particularly critical for the cache hierarchy. Previous implementations mainly focus on reducing only one kind of power in the cache, either static or dynamic. However, for a more robust approach that will remain relevant as technology continues to shrink, both aspects of power need to be addressed. Recent processors, e.g. Intel Core or IBM Power8, implement simultaneous multithreading (SMT) cores to hide high memory latencies. In these systems, the dynamic energy in the L1 cache is even more stressed since this cache level is shared by several threads running on the same core. This paper proposes and evaluates the use of phase adaptive caches in all structures of a 3-level cache hierarchy of a SMT cores. Compared to the use of conventional caches, our work results on significant dynamic and leakage energy savings with scarce performance impact.

© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., CUDA is a technology developed by NVIDIA which provides a parallel computing platform and programming model for NVIDIA GPUs and compatible ones. It takes benefit from the enormous parallel processing power of GPUs in order to accelerate a wide range of applications, thus reducing their execution time. rCUDA (remote CUDA) is a middleware which grants applications concurrent access to CUDA-compatible devices installed in other nodes of the cluster in a transparent way so that applications are not aware of accessing a remote device. In this paper we present a demo which shows, in real time, the overhead introduced by rCUDA in comparison to CUDA when running image filtering applications. The approach followed in this work is to develop a graphical demo which contains both an appealing design and technical contents.

© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Despite the recent technological improvements in vehicles and engines, and the introduction of better fuels, road transportation is still responsible for air pollution in urban areas due to the increasing number of circulating vehicles, and their relative travelled distances. We develop a methodology to calculate, in real-time, the consumption and environmental impact of spark ignition and diesel vehicles from a set of variables such as Engine Fuel Rate, Speed, Mass Air Flow, Absolute Load, and Manifold Absolute Pressure, all of them obtained from the vehicle’s Electronic Control Unit (ECU). Our platform is able to assist drivers in correcting their bad driving habits, while offering helpful recommendations to improve fuel economy. In this paper we will demonstrate through data mining, to what extent does the driving style really affect (negatively or positively) the fuel consumption, as well as the increase or reduction of greenhouse gas emissions generated by vehicles.

© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., The paper introduces the concept of reflection matrices in the coupling matrix filter reconfiguration. It is shown that reflection matrices are complementary to rotation matrices a useful concept that can be used alternatively to rotation matrices in the similarity transformations that are applied in order to transform the coupling matrix to a suitable form. A cross-coupled filter example is given where both concepts are used.

© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, Analyzing the fault-tolerance of interconnection networks implies checking the connectivity of each sourcedestination pair. The size of the exploration space of such operation skyrockets with the network size and with the number of link faults. However, this problem is highly parallelizable since the exploration of each path between a source–destination pair is independent of the other paths. This paper presents an approach to analyze the fault-tolerance degree of multistage interconnection networks using GPUs in order to speed-up it. This approach uses CUDA as parallel programming tool on a GPU in order to take advantage of all available cores. Results show that the execution time of the fault-tolerance exploration can be significantly reduced.

©2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Social circles detection is a special case of community detection in social network that is currently attracting a growing interest in the research community. In this paper, we propose a two-step technique, making emphasis on the mapping of the data by Restricted Boltzmann Machines (RBMs). Social circles are subsequently inferred by k-means over the preprocessed data. We define different vectorial representations from both structural egonet information and user profile features, and perform a set of tests to adjust the optimal parameters of the RBMs. We study and compare the performance on the ego-Facebook dataset of social circles from Facebook from the Stanford Large Network Dataset Collection. We compare our results with several different baselines.

[EN] Current UHD services use barely a couple of enhancement features compared to HD formats, by increasing spatial resolution and pixel bit-depth, which are insufficient to improve user experience when using large screens. This could led to a certain frustration on consumers, arising the fear that 4K technology could fail in the medium term as recently happened with 3DTV services. Aware of this, the audiovisual industry is involved in the agreement of the new generation of UHD, so-called UHD-1 Phase-2, which improves the immersive viewing experience features, such as Wider Color Gamut (WCG), High Dynamic Range (HDR), and High Frame Rate (HFR) technologies. ¿ In this paper, we focus on the encoding of HFR formats, which provides a crystal clear image, avoiding effects such as image blurring in scenes with high-motion, especially for sports and fast-action content. We describe the temporal scalability mechanism implemented in the HEVC standard, which can be useful for the encoding of HFR services in a backward-compatible manner, avoiding the need to carry out the simulcast of the same UHD service with different temporal resolutions. The HEVC coding efficiency for UHD@120Hz services is also reported, showing the bandwidth requirements for DVB-T2 and ATSC 3.0 networks. ¿ Finally, we describe the trial carried out at the University Campus, by broadcasting a scalable UHD@120Hz service using a DVB-T2 network, which proves backward-compatibility of the HFR services with the current UHD-1 TV sets, and also shows the HFR compatibility over low performance devices such as tablets and smartphones with low frame rates capabilities, by using in-home WiFi streaming.

©2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Social circles detection is a special case of community detection in social network that is currently attracting a growing interest in the research community. In this paper, we propose a two-step technique, making emphasis on the mapping of the data by Restricted Boltzmann Machines (RBMs). Social circles are subsequently inferred by k-means over the preprocessed data. We define different vectorial representations from both structural egonet information and user profile features, and perform a set of tests to adjust the optimal parameters of the RBMs. We study and compare the performance on the ego-Facebook dataset of social circles from Facebook from the Stanford Large Network Dataset Collection. We compare our results with several different baselines.

©2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., In Model-Driven Development (MDD), defects are managed at the level of conceptual models because the other artefacts are generated from them, such as more refined models, test cases and code. Although some studies have reported on defect types at model level, there still does not exist a clear and complete overview of the defect types that occur at the abstraction level. This paper presents a systematic mapping study to identify the model defect types reported in the literature and determine how they have been detected. Among the 282 articles published in software engineering area, 28 articles were selected for analysis. A total of 226 defects were identified, classified and their results analysed. For this, an appropriate defect classification scheme was built based on appropriate dimensions for models in an MDD context.

©2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Obtaining detailed pollution maps for urban environments is an effort that is gathering much interest by allowing to better regulate traffic and protect citizens from hazardous conditions. However, the scarcity of pollution sensors prevents obtaining the desired degree of detail, requiring alternative solutions to be deployed. In this paper we explore the concept of mobile pollution sensing by studying the feasibility of equipping buses with ozone measurement hardware to estimate ozone patterns for the city of Compiègne. Overall, we achieve accurate estimations, with error values typically ranging from 2% to 10%. Compared to solutions based on deploying static sensors on the different bus stops available, we find that the proposed mobile sensing approach is able to provide a degree of accuracy comparable to deploying tens of static sensors, substantially reducing costs and management.

© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., The paper introduces the concept of reflection matrices in the coupling matrix filter reconfiguration. It is shown that reflection matrices are complementary to rotation matrices a useful concept that can be used alternatively to rotation matrices in the similarity transformations that are applied in order to transform the coupling matrix to a suitable form. A cross-coupled filter example is given where both concepts are used.

© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Message Oriented Middleware (MOM) refers to the software infrastructure that support ubiquitous information delivery among software and hardware systems. Two of the most relevant protocols in this context are AMQP and MQTT. Lately, they have been extensively used to exchange messages conserving network bandwidth, device memory and batteries. These protocols provide an abstraction of the communication programming details of the different participating system entities, alleviating their coordination and collaboration. However, these protocols have not been thoroughly tested focused on studying the impact of node mobility. In this paper we present an experimental evaluation of both protocols quantifying the effect of the node mobility in terms of message loss, latency, jitter and saturation boundary values. Based on the results obtained, we provide criteria of applicability of these protocols. This evaluation is of interest for the upcoming applications that can be supported by MOM, and in especial for communication in Machine to Machine (M2M) and Internet of Things (IoT).

© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, Analyzing the fault-tolerance of interconnection networks implies checking the connectivity of each sourcedestination pair. The size of the exploration space of such operation skyrockets with the network size and with the number of link faults. However, this problem is highly parallelizable since the exploration of each path between a source–destination pair is independent of the other paths. This paper presents an approach to analyze the fault-tolerance degree of multistage interconnection networks using GPUs in order to speed-up it. This approach uses CUDA as parallel programming tool on a GPU in order to take advantage of all available cores. Results show that the execution time of the fault-tolerance exploration can be significantly reduced.

© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., The paper presents in premiere a simple mapping property of the sum of the reflection and transmission parameters of reciprocal two port networks. It is proved that although the reflection and transmission parameter may have very complicated paths on the Smith chart, their sum will be always moving on the unit circle if the circuit is symmetric and lossless. Further once symmetrical losses at the ports occur their sum path will switch on a family of circles through one point. Using inversive geometry we construct a new function which maps this family of circles in lines on the extended Smith chart. The proposed method for checking the symmetry uses just two parameters and avoids testing the phase of the corresponding input and output parameters. By means of the 3D Smith chart we propose in the end an alternative approach to visualize the parameters.