Pub Date : 2014-06-15DOI: 10.1109/MECO.2014.6862670
A. Roenko, V. Lukin, S. Abramov, I. Djurović
Knowledge of noise properties is important for many applications. It is often necessary to estimate characteristics of noise, in particular, spatially correlated noise automatically. For this purpose, it is desirable to have methods that allow obtaining preliminary estimates of spatial correlation degree. Here we show how this can be done by processing discrete cosine transform (DCT) coefficients calculated in 8×8 pixel image blocks. Moreover, we test the proposed method for synthetic aperture radar (SAR) images with different number of looks. The performed tests demonstrate that for TerraSAR-X data the noise correlation degree increases for larger number of looks. Besides, it is shown that the approach earlier tested for additive noise is applicable to pure multiplicative noise (speckle) and, in general to other kinds of signal-dependent noise.
{"title":"Automation of analysis for correlated noise in images based on DCT coefficient statistics","authors":"A. Roenko, V. Lukin, S. Abramov, I. Djurović","doi":"10.1109/MECO.2014.6862670","DOIUrl":"https://doi.org/10.1109/MECO.2014.6862670","url":null,"abstract":"Knowledge of noise properties is important for many applications. It is often necessary to estimate characteristics of noise, in particular, spatially correlated noise automatically. For this purpose, it is desirable to have methods that allow obtaining preliminary estimates of spatial correlation degree. Here we show how this can be done by processing discrete cosine transform (DCT) coefficients calculated in 8×8 pixel image blocks. Moreover, we test the proposed method for synthetic aperture radar (SAR) images with different number of looks. The performed tests demonstrate that for TerraSAR-X data the noise correlation degree increases for larger number of looks. Besides, it is shown that the approach earlier tested for additive noise is applicable to pure multiplicative noise (speckle) and, in general to other kinds of signal-dependent noise.","PeriodicalId":416168,"journal":{"name":"2014 3rd Mediterranean Conference on Embedded Computing (MECO)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132359429","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-15DOI: 10.1109/MECO.2014.6862714
Nenad Soskic, Dejan Popov Tapavicki, D. Rapic, S. Medic, Nikola Kuzmanovic
Living room is a shared space, in which families carry out their daily routines. The proliferation of personal devices (handhelds or laptops) has modified traditional family behaviors including watching television. Great improvement of hardware performance for handheld devices such as tablets and smartphones is the main reason of their increased usage as a second screen device and their newly acquired popularity with digital TV vendors. In this paper, one proposal of system architecture and SW framework for a second screen concept is presented. The paper focuses on implementation of DVB multimedia second screen broadcasting, providing live TV and mosaic content as well as other digital television services such as: EPG, teletext, subtitles, PVR, and remote controller for the TV or STB on the second screen.
{"title":"Remote control android application for second screen DTV environment","authors":"Nenad Soskic, Dejan Popov Tapavicki, D. Rapic, S. Medic, Nikola Kuzmanovic","doi":"10.1109/MECO.2014.6862714","DOIUrl":"https://doi.org/10.1109/MECO.2014.6862714","url":null,"abstract":"Living room is a shared space, in which families carry out their daily routines. The proliferation of personal devices (handhelds or laptops) has modified traditional family behaviors including watching television. Great improvement of hardware performance for handheld devices such as tablets and smartphones is the main reason of their increased usage as a second screen device and their newly acquired popularity with digital TV vendors. In this paper, one proposal of system architecture and SW framework for a second screen concept is presented. The paper focuses on implementation of DVB multimedia second screen broadcasting, providing live TV and mosaic content as well as other digital television services such as: EPG, teletext, subtitles, PVR, and remote controller for the TV or STB on the second screen.","PeriodicalId":416168,"journal":{"name":"2014 3rd Mediterranean Conference on Embedded Computing (MECO)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116807701","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-15DOI: 10.1109/MECO.2014.6862685
Dimitrija Angelkov, N. Koceska, S. Koceski
This work presents description of an active solar tracking system, which computes the solar position as a function of topo-centric position and time. A micro-controller based design of an automatic solar tracker is presented in this paper. The Sun positioning testing procedures conducted included two primary motor calibration tests and a heliostat positioning test with two micro switches. The measured errors from the motor positioning tests were in the range of 0.22°, whereas solar positioning tests conducted showed an error of 0.21°, which is satisfying the main market needs for decreased costs and increased accuracy.
{"title":"Low-cost dual-axis system for solar tracking","authors":"Dimitrija Angelkov, N. Koceska, S. Koceski","doi":"10.1109/MECO.2014.6862685","DOIUrl":"https://doi.org/10.1109/MECO.2014.6862685","url":null,"abstract":"This work presents description of an active solar tracking system, which computes the solar position as a function of topo-centric position and time. A micro-controller based design of an automatic solar tracker is presented in this paper. The Sun positioning testing procedures conducted included two primary motor calibration tests and a heliostat positioning test with two micro switches. The measured errors from the motor positioning tests were in the range of 0.22°, whereas solar positioning tests conducted showed an error of 0.21°, which is satisfying the main market needs for decreased costs and increased accuracy.","PeriodicalId":416168,"journal":{"name":"2014 3rd Mediterranean Conference on Embedded Computing (MECO)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126886104","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-15DOI: 10.1109/MECO.2014.6862723
Orges Cico, Z. Dika
Migration from the classic server platforms to the newly emerging cloud technologies has been the most common challenges during the past years. Many businesses and users are yet concerned about what the cloud has to offer. Moving their businesses and data to a new platform yet has some challenges to overcome, such as safety, integrity, reliability and scalability. Most of the companies offering cloud solutions claim to have overcome many obstacles and provide a fully robust infrastructure, but these claims are yet to be proven with real case studies and software engineering approaches. In this paper we will provide real data measurements, obtained from load and performance testing of both infrastructures, that will present a good comparison between Server and Cloud platforms when providing the same service to the end users. We have chosen an adequate application, a university social network, which can provide a good estimation of the scalability of the Cloud vs. Server infrastructure. This case study has not only proven that some of the claims made so far are right but it has also raised several other questions in terms of investments costs, reliability and future implementations approaches.
{"title":"Performance and load testing of cloud vs. classic server platforms (Case study: Social network application)","authors":"Orges Cico, Z. Dika","doi":"10.1109/MECO.2014.6862723","DOIUrl":"https://doi.org/10.1109/MECO.2014.6862723","url":null,"abstract":"Migration from the classic server platforms to the newly emerging cloud technologies has been the most common challenges during the past years. Many businesses and users are yet concerned about what the cloud has to offer. Moving their businesses and data to a new platform yet has some challenges to overcome, such as safety, integrity, reliability and scalability. Most of the companies offering cloud solutions claim to have overcome many obstacles and provide a fully robust infrastructure, but these claims are yet to be proven with real case studies and software engineering approaches. In this paper we will provide real data measurements, obtained from load and performance testing of both infrastructures, that will present a good comparison between Server and Cloud platforms when providing the same service to the end users. We have chosen an adequate application, a university social network, which can provide a good estimation of the scalability of the Cloud vs. Server infrastructure. This case study has not only proven that some of the claims made so far are right but it has also raised several other questions in terms of investments costs, reliability and future implementations approaches.","PeriodicalId":416168,"journal":{"name":"2014 3rd Mediterranean Conference on Embedded Computing (MECO)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127810746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-15DOI: 10.1109/MECO.2014.6862721
S. Latifi, A. Durresi, B. Çiço
Software Defined Networks are emerging as a new paradigm in data networks. While most corporate networks and datacenters (like Google, Cisco, Big Switch, etc.) are already on OpenFlow protocol utilizing SDN, a number of emulation platforms have emerged, such as Mininet. In this paper we propose an emulated enterprise/university Software-Defined network using Python and Mininet; we have also built a conventional network using ns-3 with the same number of nodes, topology, etc. Our tests show that our proposed network model outperforms the conventional/existing model in parameters like minimum and mean delay for ICMP and HTTP packets, as well as minimum and maximum convergence time, and therefore are a best solution to transition corporate networks with hundreds or more nodes, mainly due to the fast propagation of routing policies from the central controller.
{"title":"Emulating enterprise network environments for fast transition to software-defined networking","authors":"S. Latifi, A. Durresi, B. Çiço","doi":"10.1109/MECO.2014.6862721","DOIUrl":"https://doi.org/10.1109/MECO.2014.6862721","url":null,"abstract":"Software Defined Networks are emerging as a new paradigm in data networks. While most corporate networks and datacenters (like Google, Cisco, Big Switch, etc.) are already on OpenFlow protocol utilizing SDN, a number of emulation platforms have emerged, such as Mininet. In this paper we propose an emulated enterprise/university Software-Defined network using Python and Mininet; we have also built a conventional network using ns-3 with the same number of nodes, topology, etc. Our tests show that our proposed network model outperforms the conventional/existing model in parameters like minimum and mean delay for ICMP and HTTP packets, as well as minimum and maximum convergence time, and therefore are a best solution to transition corporate networks with hundreds or more nodes, mainly due to the fast propagation of routing policies from the central controller.","PeriodicalId":416168,"journal":{"name":"2014 3rd Mediterranean Conference on Embedded Computing (MECO)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132765524","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-15DOI: 10.1109/MECO.2014.6862675
Y. Bekhtin, Damiao P. Malebo
The suggested method is aimed to be realized in different on-board embedded systems for images obtained by synthetic aperture radar (SAR) and featured with the problem of speckle. The conducted analysis of real SAR images has shown that the statistical properties of speckle strongly depend on the texture of the irradiated surface. The main idea of the method is to select the optimum despeckling algorithms from the recommended database of the algorithms using texture analysis and apply the selected algorithms to various segments of SAR images having different variances and probability density functions of speckle. The results of modeling have shown the advantage of the combined filtering in comparing with some famous despeckling procedures under criteria such as PSNR, SSIM, etc.
{"title":"Combined textural-dependent despeckling in SAR images","authors":"Y. Bekhtin, Damiao P. Malebo","doi":"10.1109/MECO.2014.6862675","DOIUrl":"https://doi.org/10.1109/MECO.2014.6862675","url":null,"abstract":"The suggested method is aimed to be realized in different on-board embedded systems for images obtained by synthetic aperture radar (SAR) and featured with the problem of speckle. The conducted analysis of real SAR images has shown that the statistical properties of speckle strongly depend on the texture of the irradiated surface. The main idea of the method is to select the optimum despeckling algorithms from the recommended database of the algorithms using texture analysis and apply the selected algorithms to various segments of SAR images having different variances and probability density functions of speckle. The results of modeling have shown the advantage of the combined filtering in comparing with some famous despeckling procedures under criteria such as PSNR, SSIM, etc.","PeriodicalId":416168,"journal":{"name":"2014 3rd Mediterranean Conference on Embedded Computing (MECO)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133238861","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-15DOI: 10.1109/MECO.2014.6862682
S. Elesina, Olga Lomteva
Research of genetic algorithm with the aim to receive optimum settings while it being used in combined vision systems has been carried out. To decrease the complexity in global extremum search advanced angles of virtual terrain model are supposed to be used. System performance is shown to be increased 5 times using such approach.
{"title":"Increase of image combination performance in combined vision systems using genetic algorithm","authors":"S. Elesina, Olga Lomteva","doi":"10.1109/MECO.2014.6862682","DOIUrl":"https://doi.org/10.1109/MECO.2014.6862682","url":null,"abstract":"Research of genetic algorithm with the aim to receive optimum settings while it being used in combined vision systems has been carried out. To decrease the complexity in global extremum search advanced angles of virtual terrain model are supposed to be used. System performance is shown to be increased 5 times using such approach.","PeriodicalId":416168,"journal":{"name":"2014 3rd Mediterranean Conference on Embedded Computing (MECO)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133503490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-15DOI: 10.1109/MECO.2014.6862695
B. Podlaskin, E. Guk, A. Sukharev
It is shown that image processing performed by computer embedded systems with the use of 2D Hadamard transform allows indexing of the elements of 1D additive communication channel noise and its representation in a form of localized 2D distribution of its variance. A higher degree of noise variance localization compared to systems with 1D transform can be obtained this way, provided the correlation between 1D noise elements is preserved.
{"title":"Quasi-spatial properties of a communication channel temporal noise while using 2D hadamard transform","authors":"B. Podlaskin, E. Guk, A. Sukharev","doi":"10.1109/MECO.2014.6862695","DOIUrl":"https://doi.org/10.1109/MECO.2014.6862695","url":null,"abstract":"It is shown that image processing performed by computer embedded systems with the use of 2D Hadamard transform allows indexing of the elements of 1D additive communication channel noise and its representation in a form of localized 2D distribution of its variance. A higher degree of noise variance localization compared to systems with 1D transform can be obtained this way, provided the correlation between 1D noise elements is preserved.","PeriodicalId":416168,"journal":{"name":"2014 3rd Mediterranean Conference on Embedded Computing (MECO)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115660198","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-15DOI: 10.1109/MECO.2014.6862700
R. Mihajlovic, Marija Scekic, A. Draganic, S. Stankovic
As need for increasing the speed and accuracy of the real applications is constantly growing, the new algorithms and methods for signal processing are intensively developing. Traditional sampling approach based on Sampling theorem is, in many applications, inefficient because of production a large number of signal samples. Generally, small number of significant information is presented within the signal compared to its length. Therefore, the Compressive Sensing method is developed as an alternative sampling strategy. This method provides efficient signal processing and reconstruction, without need for collecting all of the signal samples. Signal is sampled in a random way, with number of acquired samples significantly smaller than the signal length. In this paper, the comparison of the several algorithms for Compressive Sensing reconstruction is presented. The one dimensional band-limited signals that appear in wireless communications are observed and the performance of the algorithms in non-noisy and noisy environments is tested. Reconstruction errors and execution times are compared between different algorithms, as well.
{"title":"An analysis of CS algorithms efficiency for sparse communication signals reconstruction","authors":"R. Mihajlovic, Marija Scekic, A. Draganic, S. Stankovic","doi":"10.1109/MECO.2014.6862700","DOIUrl":"https://doi.org/10.1109/MECO.2014.6862700","url":null,"abstract":"As need for increasing the speed and accuracy of the real applications is constantly growing, the new algorithms and methods for signal processing are intensively developing. Traditional sampling approach based on Sampling theorem is, in many applications, inefficient because of production a large number of signal samples. Generally, small number of significant information is presented within the signal compared to its length. Therefore, the Compressive Sensing method is developed as an alternative sampling strategy. This method provides efficient signal processing and reconstruction, without need for collecting all of the signal samples. Signal is sampled in a random way, with number of acquired samples significantly smaller than the signal length. In this paper, the comparison of the several algorithms for Compressive Sensing reconstruction is presented. The one dimensional band-limited signals that appear in wireless communications are observed and the performance of the algorithms in non-noisy and noisy environments is tested. Reconstruction errors and execution times are compared between different algorithms, as well.","PeriodicalId":416168,"journal":{"name":"2014 3rd Mediterranean Conference on Embedded Computing (MECO)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129471111","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-15DOI: 10.1109/MECO.2014.6862666
P. Meloni, Giuseppe Tuveri, L. Raffo, Igor Loi, Francesco Conti
In modern MPSoC architectures, programming to effectively exploit all the available resources becomes very challenging. Polyhedral Process Networks (PPN) are a known model of computation that represents a suitable solution for systematic mapping of parallel applications onto multiprocessor architectures. In previous works it has been shown that a given PPN program specification can be further analyzed and optimized, in order to meet the desired performance requirements. In this paper we present an online process splitting transformation that does not need a re-design of the communication patterns in network structure of the application. The novelty of our approach is that, differently from other compile-time approaches, the proposed transformation technique can be applied at run-time and followed, if needed, by the backward transformation. Using a FPGA-based MPSoC shared memory platform, we present an evaluation of the achievable performance improvements. We also discuss the overhead caused by the introduction of the run-time transformation support.
在现代MPSoC架构中，有效利用所有可用资源的编程变得非常具有挑战性。多面体进程网络(Polyhedral Process Networks, PPN)是一种已知的计算模型，它代表了将并行应用程序系统地映射到多处理器架构上的合适解决方案。在以前的工作中已经表明，给定的PPN程序规范可以进一步分析和优化，以满足期望的性能要求。本文提出了一种不需要重新设计应用程序网络结构中的通信模式的在线进程拆分转换方法。我们的方法的新颖之处在于，与其他编译时方法不同，所建议的转换技术可以在运行时应用，如果需要的话，还可以进行向后转换。使用基于fpga的MPSoC共享存储器平台，我们对可实现的性能改进进行了评估。我们还讨论了引入运行时转换支持所带来的开销。
{"title":"Online process transformation for polyhedral process networks in shared-memory MPSoCs","authors":"P. Meloni, Giuseppe Tuveri, L. Raffo, Igor Loi, Francesco Conti","doi":"10.1109/MECO.2014.6862666","DOIUrl":"https://doi.org/10.1109/MECO.2014.6862666","url":null,"abstract":"In modern MPSoC architectures, programming to effectively exploit all the available resources becomes very challenging. Polyhedral Process Networks (PPN) are a known model of computation that represents a suitable solution for systematic mapping of parallel applications onto multiprocessor architectures. In previous works it has been shown that a given PPN program specification can be further analyzed and optimized, in order to meet the desired performance requirements. In this paper we present an online process splitting transformation that does not need a re-design of the communication patterns in network structure of the application. The novelty of our approach is that, differently from other compile-time approaches, the proposed transformation technique can be applied at run-time and followed, if needed, by the backward transformation. Using a FPGA-based MPSoC shared memory platform, we present an evaluation of the achievable performance improvements. We also discuss the overhead caused by the introduction of the run-time transformation support.","PeriodicalId":416168,"journal":{"name":"2014 3rd Mediterranean Conference on Embedded Computing (MECO)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122465460","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：