Pub Date : 2014-12-01DOI: 10.1109/GlobalSIP.2014.7032200
Tarek A. Lahlou, A. Oppenheim
Signal processing techniques exploiting natural and efficient representations of a class of signals with an underlying parametric model have been extensively studied and successfully applied across many disciplines. In this paper, we focus attention to the representation of one such class, i.e. transient structured signals. The class of transient signals in particular often results in computationally ill-conditioned problems which are further degraded by the presence of noise. We develop the Discrete Transient Transform, a biorthogonal transform to a basis parameterized by decay rate, along with algorithms for its implementation which mitigate these numerical issues and enable a spectral approach to parameter identification, estimation, and modeling for signals with transient behavior. The three algorithms developed have varying degrees of numerical robustness for generating the biorthogonal transient basis. Issues pertaining to transient spectral leakage and resolution are characterized and discussed in the context of an example related to Vandermonde system inversion.
{"title":"Spectral representation of transient signals","authors":"Tarek A. Lahlou, A. Oppenheim","doi":"10.1109/GlobalSIP.2014.7032200","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032200","url":null,"abstract":"Signal processing techniques exploiting natural and efficient representations of a class of signals with an underlying parametric model have been extensively studied and successfully applied across many disciplines. In this paper, we focus attention to the representation of one such class, i.e. transient structured signals. The class of transient signals in particular often results in computationally ill-conditioned problems which are further degraded by the presence of noise. We develop the Discrete Transient Transform, a biorthogonal transform to a basis parameterized by decay rate, along with algorithms for its implementation which mitigate these numerical issues and enable a spectral approach to parameter identification, estimation, and modeling for signals with transient behavior. The three algorithms developed have varying degrees of numerical robustness for generating the biorthogonal transient basis. Issues pertaining to transient spectral leakage and resolution are characterized and discussed in the context of an example related to Vandermonde system inversion.","PeriodicalId":362306,"journal":{"name":"2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131342337","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 : 2014-12-01DOI: 10.1109/GlobalSIP.2014.7032303
Shiying Han, Ying-Chang Liang, B. Soong, Fengye Hu
Spectrum refarming (SR) allows different generations of cellular networks to operate in the same frequency band. Infrastructure sharing involves the sharing of passive or active elements of mobile infrastructure within or among operators. Both of them are important as the cellular technology evolves. In this study, we consider an uplink OFDMA/CDMA SR system with passive infrastructure sharing, i.e., the OFDMA system and CDMA system share a common cell site while each adopting different receive antennas. Due to the cross channel state information (C-CSI) is difficult to be obtained in practice, we propose a resource allocation scheme with which the OFDMA system can provide sufficient protection to the CDMA system without C-CSI. We further propose to increase the interference margin for OFDMA resource allocation by exploiting the gap between the actual and the predicted interference suffered by CDMA users, which permits higher transmission power from OFDMA users. Simulation results have validated that the CDMA services can be protected by the proposed schemes, while the OFDMA throughput can be improved.
{"title":"Resource allocation for OFDMA/CDMA spectrum refarming system with passive infrastructure sharing","authors":"Shiying Han, Ying-Chang Liang, B. Soong, Fengye Hu","doi":"10.1109/GlobalSIP.2014.7032303","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032303","url":null,"abstract":"Spectrum refarming (SR) allows different generations of cellular networks to operate in the same frequency band. Infrastructure sharing involves the sharing of passive or active elements of mobile infrastructure within or among operators. Both of them are important as the cellular technology evolves. In this study, we consider an uplink OFDMA/CDMA SR system with passive infrastructure sharing, i.e., the OFDMA system and CDMA system share a common cell site while each adopting different receive antennas. Due to the cross channel state information (C-CSI) is difficult to be obtained in practice, we propose a resource allocation scheme with which the OFDMA system can provide sufficient protection to the CDMA system without C-CSI. We further propose to increase the interference margin for OFDMA resource allocation by exploiting the gap between the actual and the predicted interference suffered by CDMA users, which permits higher transmission power from OFDMA users. Simulation results have validated that the CDMA services can be protected by the proposed schemes, while the OFDMA throughput can be improved.","PeriodicalId":362306,"journal":{"name":"2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP)","volume":"205 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121286392","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 : 2014-12-01DOI: 10.1109/GlobalSIP.2014.7032187
Larry Heck, Hongzhao Huang
This paper presents a novel method to learn neural knowledge graph embeddings. The embeddings are used to compute semantic relatedness in a coherence-based semantic parser. The approach learns embeddings directly from structured knowledge representations. A deep neural network approach known as Deep Structured Semantic Modeling (DSSM) is used to scale the approach to learn neural embeddings for all of the concepts (pages) of Wikipedia. Experiments on Twitter dialogs show a 23.6% reduction in semantic parsing errors compared to the state-of-the-art unsupervised approach.
{"title":"Deep learning of knowledge graph embeddings for semantic parsing of Twitter dialogs","authors":"Larry Heck, Hongzhao Huang","doi":"10.1109/GlobalSIP.2014.7032187","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032187","url":null,"abstract":"This paper presents a novel method to learn neural knowledge graph embeddings. The embeddings are used to compute semantic relatedness in a coherence-based semantic parser. The approach learns embeddings directly from structured knowledge representations. A deep neural network approach known as Deep Structured Semantic Modeling (DSSM) is used to scale the approach to learn neural embeddings for all of the concepts (pages) of Wikipedia. Experiments on Twitter dialogs show a 23.6% reduction in semantic parsing errors compared to the state-of-the-art unsupervised approach.","PeriodicalId":362306,"journal":{"name":"2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121194314","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 : 2014-12-01DOI: 10.1109/GlobalSIP.2014.7032326
Xinzhi Zhang, Rong Chai, F. Gao
In this paper, we study the matched filter based spectrum sensing in a more reasonable cognitive radio (CR) scenario when the primary user (PU) could possibly work on more than one transmit power levels. Besides the traditional sensing target where the secondary user (SU) should decide the presence of PU, an additionally new target here could be also recognizing the power levels of PU, which achieves more "cognition" for CR. We derive the closed form solutions for decision regions and several performance metrics, from which some interesting phenomenons are observed and the related discussions are provided. Numerical examples are presented to corroborate the proposed studies.
{"title":"Matched filter based spectrum sensing and power level detection for cognitive radio network","authors":"Xinzhi Zhang, Rong Chai, F. Gao","doi":"10.1109/GlobalSIP.2014.7032326","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032326","url":null,"abstract":"In this paper, we study the matched filter based spectrum sensing in a more reasonable cognitive radio (CR) scenario when the primary user (PU) could possibly work on more than one transmit power levels. Besides the traditional sensing target where the secondary user (SU) should decide the presence of PU, an additionally new target here could be also recognizing the power levels of PU, which achieves more \"cognition\" for CR. We derive the closed form solutions for decision regions and several performance metrics, from which some interesting phenomenons are observed and the related discussions are provided. Numerical examples are presented to corroborate the proposed studies.","PeriodicalId":362306,"journal":{"name":"2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121796416","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 : 2014-12-01DOI: 10.1109/GlobalSIP.2014.7032209
Ahmed G. Khodary, H. A. Aly
Poor visibility due to haze poses a challenge for driving and can significantly compromise safety. In this paper, we present an algorithm and an implementation that assists the driver by providing an electronic view that improves visibility via haze removal. Our optimized implementation on the Texas Instruments TMS320DM6446EVM DSP based evaluation board provides full D1 video resolution and frame processed for haze removal every 1.7 seconds. This near real-time implementation can significantly augment the drivers vision by providing updates at a rate comparable with the rate at which the driver looks at the rear-view mirror. The algorithm uses a dark channel prior with an edge-guided filter for fast implementation; both components are specifically modified for improving performance on the DM6446 processor. The system provides the first DSP-based embedded implementation of haze removal processing with near real-time performance and represents a significant speed-up and reduction in memory requirements over previously reported algorithms.
{"title":"A new image-sequence haze removal system based on DM6446 Davinci processor","authors":"Ahmed G. Khodary, H. A. Aly","doi":"10.1109/GlobalSIP.2014.7032209","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032209","url":null,"abstract":"Poor visibility due to haze poses a challenge for driving and can significantly compromise safety. In this paper, we present an algorithm and an implementation that assists the driver by providing an electronic view that improves visibility via haze removal. Our optimized implementation on the Texas Instruments TMS320DM6446EVM DSP based evaluation board provides full D1 video resolution and frame processed for haze removal every 1.7 seconds. This near real-time implementation can significantly augment the drivers vision by providing updates at a rate comparable with the rate at which the driver looks at the rear-view mirror. The algorithm uses a dark channel prior with an edge-guided filter for fast implementation; both components are specifically modified for improving performance on the DM6446 processor. The system provides the first DSP-based embedded implementation of haze removal processing with near real-time performance and represents a significant speed-up and reduction in memory requirements over previously reported algorithms.","PeriodicalId":362306,"journal":{"name":"2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP)","volume":"19 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113963233","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 : 2014-12-01DOI: 10.1109/GlobalSIP.2014.7032281
Joseph Santarcangelo, Xiao-Ping Zhang
This paper develops dynamic prediction hidden Markov models for arousal time curve estimation in sports videos. The method determines the arousal time curve by selecting a state sequence that maximizes the joint probability density function between the states and the arousal time curve. We derive the parameters using the expected maximization algorithm. Experiments were performed on several types of sports videos. Test measures include squared residual error and criteria derived from psychology. The experimental results show that the novel method performed better in estimating the arousal time curve than state of the art linear regression methods on most of the tested sports videos.
{"title":"Arousal content representation of sports videos using dynamic prediction hidden Markov models","authors":"Joseph Santarcangelo, Xiao-Ping Zhang","doi":"10.1109/GlobalSIP.2014.7032281","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032281","url":null,"abstract":"This paper develops dynamic prediction hidden Markov models for arousal time curve estimation in sports videos. The method determines the arousal time curve by selecting a state sequence that maximizes the joint probability density function between the states and the arousal time curve. We derive the parameters using the expected maximization algorithm. Experiments were performed on several types of sports videos. Test measures include squared residual error and criteria derived from psychology. The experimental results show that the novel method performed better in estimating the arousal time curve than state of the art linear regression methods on most of the tested sports videos.","PeriodicalId":362306,"journal":{"name":"2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP)","volume":"80 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131416370","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 : 2014-12-01DOI: 10.1109/GlobalSIP.2014.7032331
Daquan Feng, Guanding Yu, Y. Yuan-Wu, Geoffrey Y. Li, G. Feng, Shaoqian Li
Mode switching is one of the most important features of device-to-device (D2D) communications since it can bring more freedoms for potential D2D pairs. In this paper, we investigate optimal D2D mode switching to maximize the network spectrum-efficiency (SE). We formulate the optimal SE problems in three D2D transmission modes, dedicated mode, reusing mode and cellular mode, while guaranteeing the quality-of-service (QoS) requirements for both the D2D pairs and the regular cellular users (RCUs). Bisection algorithm is adopted to solve the quasiconvex optimization problems in the dedicated and cellular modes through transforming the original problem into a sequence of convex feasibility problems. For the reusing mode, concave-convex procedure (CCCP) is used to solve the difference of convex (D. C.) optimization problem. Simulation results show that system SE can be improved significantly with the proposed mode switching algorithm compared with the single mode transmission without mode switching.
{"title":"Mode switching for device-to-device communications in cellular networks","authors":"Daquan Feng, Guanding Yu, Y. Yuan-Wu, Geoffrey Y. Li, G. Feng, Shaoqian Li","doi":"10.1109/GlobalSIP.2014.7032331","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032331","url":null,"abstract":"Mode switching is one of the most important features of device-to-device (D2D) communications since it can bring more freedoms for potential D2D pairs. In this paper, we investigate optimal D2D mode switching to maximize the network spectrum-efficiency (SE). We formulate the optimal SE problems in three D2D transmission modes, dedicated mode, reusing mode and cellular mode, while guaranteeing the quality-of-service (QoS) requirements for both the D2D pairs and the regular cellular users (RCUs). Bisection algorithm is adopted to solve the quasiconvex optimization problems in the dedicated and cellular modes through transforming the original problem into a sequence of convex feasibility problems. For the reusing mode, concave-convex procedure (CCCP) is used to solve the difference of convex (D. C.) optimization problem. Simulation results show that system SE can be improved significantly with the proposed mode switching algorithm compared with the single mode transmission without mode switching.","PeriodicalId":362306,"journal":{"name":"2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP)","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131485230","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 : 2014-12-01DOI: 10.1109/GlobalSIP.2014.7032346
Qian Wan, R. Pal
We consider a prediction problem with multiple output responses based on an ensemble of multivariate regression trees. The selection of the optimal ensemble is formulated as a multi-objective optimization problem and solved using genetic algorithms. We illustrate the application of our approach on drug sensitivity prediction problem where the proposed methodology outperforms regular multivariate random forests in terms of correlation coefficients between predicted and experimental sensitivities. We also demonstrate that generating the Pareto-optimal front provides us a choice of ensembles for different optimization objectives.
{"title":"Multi-objective optimization of ensemble of regression trees using genetic algorithms","authors":"Qian Wan, R. Pal","doi":"10.1109/GlobalSIP.2014.7032346","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032346","url":null,"abstract":"We consider a prediction problem with multiple output responses based on an ensemble of multivariate regression trees. The selection of the optimal ensemble is formulated as a multi-objective optimization problem and solved using genetic algorithms. We illustrate the application of our approach on drug sensitivity prediction problem where the proposed methodology outperforms regular multivariate random forests in terms of correlation coefficients between predicted and experimental sensitivities. We also demonstrate that generating the Pareto-optimal front provides us a choice of ensembles for different optimization objectives.","PeriodicalId":362306,"journal":{"name":"2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP)","volume":"207 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132553368","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 : 2014-12-01DOI: 10.1109/GlobalSIP.2014.7032333
Ahmed M. Salama, A. Zahran, T. Elbatt
In this paper, we propose fractional sequential sensing (FSS) as a novel cooperative sensing scheme for cognitive radio networks. FSS compromises a tradeoff between sensing accuracy and efficiency by formulating an optimization problem whose solution identifies FSS sensing parameters. These parameters include the sensing period and channels allocated for each user. Our simulation results show that FSS successfully improves the sensing accuracy while maintaining a low power profile. Additionally, we show that the sensing accuracy performance gap between FSS and other traditional schemes increases by optimizing decision engine.
{"title":"Fractional sequential sensing for energy efficient cooperative cognitive radio networks","authors":"Ahmed M. Salama, A. Zahran, T. Elbatt","doi":"10.1109/GlobalSIP.2014.7032333","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032333","url":null,"abstract":"In this paper, we propose fractional sequential sensing (FSS) as a novel cooperative sensing scheme for cognitive radio networks. FSS compromises a tradeoff between sensing accuracy and efficiency by formulating an optimization problem whose solution identifies FSS sensing parameters. These parameters include the sensing period and channels allocated for each user. Our simulation results show that FSS successfully improves the sensing accuracy while maintaining a low power profile. Additionally, we show that the sensing accuracy performance gap between FSS and other traditional schemes increases by optimizing decision engine.","PeriodicalId":362306,"journal":{"name":"2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133537072","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 : 2014-12-01DOI: 10.1109/GlobalSIP.2014.7032360
Yijie Wang, Xiaoning Qian
With increasingly "big" data available in biomédical research, deriving accurate and reproducible biology knowledge from such big data imposes enormous computational challenges. In this paper, we propose a highly scalable randomized coordinate descent Frank-Wolfe algorithm for convex optimization with compact convex constraints, which has diverse applications in analyzing biomédical data for better understanding cellular and disease mechanisms. We focus on implementing the derived stochastic coordinate descent algorithm to align protein-protein interaction networks for identifying conserved functional pathways based on IsoRank. The stochastic algorithm naturally leads to the decreased computational cost for each iteration. More importantly, we show that it achieves a linear convergence rate. Our numerical test confirms the improved efficiency of this technique for the large-scale biological network alignment problem.
{"title":"Stochastic coordinate descent Frank-Wolfe algorithm for large-scale biological network alignment","authors":"Yijie Wang, Xiaoning Qian","doi":"10.1109/GlobalSIP.2014.7032360","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032360","url":null,"abstract":"With increasingly \"big\" data available in biomédical research, deriving accurate and reproducible biology knowledge from such big data imposes enormous computational challenges. In this paper, we propose a highly scalable randomized coordinate descent Frank-Wolfe algorithm for convex optimization with compact convex constraints, which has diverse applications in analyzing biomédical data for better understanding cellular and disease mechanisms. We focus on implementing the derived stochastic coordinate descent algorithm to align protein-protein interaction networks for identifying conserved functional pathways based on IsoRank. The stochastic algorithm naturally leads to the decreased computational cost for each iteration. More importantly, we show that it achieves a linear convergence rate. Our numerical test confirms the improved efficiency of this technique for the large-scale biological network alignment problem.","PeriodicalId":362306,"journal":{"name":"2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP)","volume":"51 1-2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131894842","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：