Pub Date : 2016-08-01DOI: 10.1109/FSKD.2016.7603289
C. Yu, Pin Wan, Yonghua Wang, Ting-Jung Liang
The maximum and minimum eigenvalue (MME) spectrum sensing algorithm with features such as low complexity, no need of the prior information of authorized users, etc. However, because of its detection distribution function is not clear, researchers have improved the MME spectrum sensing algorithm from the point of view of the distribution function, but cannot solve the insufficient detection performance issues of these algorithms in low signal noise ratio (SNR). To solve this problem, this paper proposes two joint spectrum sensing algorithms based on two improved MME algorithms and cyclic stationary feature detection algorithm. Simulation results show that the performance of these two kinds of joint spectrum sensing algorithms is superior to both individual performance. At the same time, its performance is better than the performance of the simple MME-cyclic stationary feature joint spectrum sensing algorithm.
{"title":"Spectrum sensing algorithm based on improved MME-Cyclic stationary feature","authors":"C. Yu, Pin Wan, Yonghua Wang, Ting-Jung Liang","doi":"10.1109/FSKD.2016.7603289","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603289","url":null,"abstract":"The maximum and minimum eigenvalue (MME) spectrum sensing algorithm with features such as low complexity, no need of the prior information of authorized users, etc. However, because of its detection distribution function is not clear, researchers have improved the MME spectrum sensing algorithm from the point of view of the distribution function, but cannot solve the insufficient detection performance issues of these algorithms in low signal noise ratio (SNR). To solve this problem, this paper proposes two joint spectrum sensing algorithms based on two improved MME algorithms and cyclic stationary feature detection algorithm. Simulation results show that the performance of these two kinds of joint spectrum sensing algorithms is superior to both individual performance. At the same time, its performance is better than the performance of the simple MME-cyclic stationary feature joint spectrum sensing algorithm.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130907830","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603190
Baorong He, Dekuan Xu
Song titles is a special form of language expression with modernity and popularity: they are short in form and concise in meaning and can reflect the ideology and values of an era. In this paper, we built a co-occurrence network with the titles of approximately six thousand Chinese popular songs. We make an all-round research about it from the perspective of complex networks and explain such characteristics as small world effect, scale-free, hierarchy, betweenness centrality and assortiveness and so on. This paper reveals the unique nature of the co-occurrence network of the titles of popular songs and broadens the scope of language network studies.
{"title":"An exploration on the word co-occurrence network of Chinese popular song titles","authors":"Baorong He, Dekuan Xu","doi":"10.1109/FSKD.2016.7603190","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603190","url":null,"abstract":"Song titles is a special form of language expression with modernity and popularity: they are short in form and concise in meaning and can reflect the ideology and values of an era. In this paper, we built a co-occurrence network with the titles of approximately six thousand Chinese popular songs. We make an all-round research about it from the perspective of complex networks and explain such characteristics as small world effect, scale-free, hierarchy, betweenness centrality and assortiveness and so on. This paper reveals the unique nature of the co-occurrence network of the titles of popular songs and broadens the scope of language network studies.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131125437","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603376
Dapu Li, Wei Ai, Yu Ye, Jie Liang
In this article, an efficient parallel algorithm for a hybrid CPU-GPU platform is proposed to enable large-scale molecular dynamics (MD) simulations of the metal solidification process. The results, implemented the parallel algorithm program on the hybrid CPU-GPU platform shows better performance than the program based on previous algorithms running on the CPU cluster platform. By contrast, the total execution time of the new program has been obviously decreased. Particularly, because of the use of the modified load balancing method, the neighbor list update time is approximately zero. The parallel program based on the CUDA+OpenMP model shows a factor of 6 16-core calculation speedups compared to the parallel program based on the MPI+OpenMP model, and the optimal computational efficiency is achieved in the simulation system including 10,000,000 aluminum atoms. Finally, the good consistency between them verifies the correctness of the algorithm efficiently, by comparison of the theoretical results and experimental results.
{"title":"A efficient algorithm for molecular dynamics simulation on hybrid CPU-GPU computing platforms","authors":"Dapu Li, Wei Ai, Yu Ye, Jie Liang","doi":"10.1109/FSKD.2016.7603376","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603376","url":null,"abstract":"In this article, an efficient parallel algorithm for a hybrid CPU-GPU platform is proposed to enable large-scale molecular dynamics (MD) simulations of the metal solidification process. The results, implemented the parallel algorithm program on the hybrid CPU-GPU platform shows better performance than the program based on previous algorithms running on the CPU cluster platform. By contrast, the total execution time of the new program has been obviously decreased. Particularly, because of the use of the modified load balancing method, the neighbor list update time is approximately zero. The parallel program based on the CUDA+OpenMP model shows a factor of 6 16-core calculation speedups compared to the parallel program based on the MPI+OpenMP model, and the optimal computational efficiency is achieved in the simulation system including 10,000,000 aluminum atoms. Finally, the good consistency between them verifies the correctness of the algorithm efficiently, by comparison of the theoretical results and experimental results.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"2257 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130225228","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603285
Huijuan Zhang, R. Sun
Because of rare earth futures stock variability and uncertainty of the market, many investors hope to be able to predict the price of rare earth futures on the stock market in the future. The neural network does do better than others in short-term forecasting, and there is no need to establish a complex nonlinear mathematical model and relationship. Based on these advantages, this paper uses the neural network based on genetic algorithm to predict the closing price of rare earth stock by analyzing the historical data of rare earth stock. In the genetic algorithm, the parameters such as crossover rate, mutation rate, iterations and population size are analyzed. Based on the parameter analysis results, a hybrid machine learning model which is suitable for the prediction of rare earth stock is established, which provides a reference for the investors.
{"title":"Parameter analysis of hybrid intelligent model for the prediction of rare earth stock futures","authors":"Huijuan Zhang, R. Sun","doi":"10.1109/FSKD.2016.7603285","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603285","url":null,"abstract":"Because of rare earth futures stock variability and uncertainty of the market, many investors hope to be able to predict the price of rare earth futures on the stock market in the future. The neural network does do better than others in short-term forecasting, and there is no need to establish a complex nonlinear mathematical model and relationship. Based on these advantages, this paper uses the neural network based on genetic algorithm to predict the closing price of rare earth stock by analyzing the historical data of rare earth stock. In the genetic algorithm, the parameters such as crossover rate, mutation rate, iterations and population size are analyzed. Based on the parameter analysis results, a hybrid machine learning model which is suitable for the prediction of rare earth stock is established, which provides a reference for the investors.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128740848","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603232
Yu Cao, Xin Niu, Y. Dou
Recently, the automatic object detection in high-resolution remote sensing images has become the key point in the application of remote sensing technology. The traditional methods, such as bag-of-visual-words (BOVW), could perform well in simple scenes, but when it used in complex scenes, the performance drops quickly. This paper we first try to use the current hot deep learning technology: Region-based convolutional neural networks (R-CNN), to detect aircrafts under the complex environments in high-resolution remote sensing images. This method has been proved to be very efficiency when using in object detection in natural images. Here, we tried to introduce this method into the field of the remote sensing. During our experiments, we also compared the impact of different proposal generate methods on the final detection results. And we also proposed some practical tips to accelerate the detection speed. After detection, we proposed to use a novel algorithm which we called box-fusion, to eliminate the redundant and repetitive boxes that covering the same object. As experiments and results shows, the R-CNN method is much more effective and robust than the traditional BOVW method when dealing with aircrafts detection under complex scenes in high-resolution remote sensing images.
{"title":"Region-based convolutional neural networks for object detection in very high resolution remote sensing images","authors":"Yu Cao, Xin Niu, Y. Dou","doi":"10.1109/FSKD.2016.7603232","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603232","url":null,"abstract":"Recently, the automatic object detection in high-resolution remote sensing images has become the key point in the application of remote sensing technology. The traditional methods, such as bag-of-visual-words (BOVW), could perform well in simple scenes, but when it used in complex scenes, the performance drops quickly. This paper we first try to use the current hot deep learning technology: Region-based convolutional neural networks (R-CNN), to detect aircrafts under the complex environments in high-resolution remote sensing images. This method has been proved to be very efficiency when using in object detection in natural images. Here, we tried to introduce this method into the field of the remote sensing. During our experiments, we also compared the impact of different proposal generate methods on the final detection results. And we also proposed some practical tips to accelerate the detection speed. After detection, we proposed to use a novel algorithm which we called box-fusion, to eliminate the redundant and repetitive boxes that covering the same object. As experiments and results shows, the R-CNN method is much more effective and robust than the traditional BOVW method when dealing with aircrafts detection under complex scenes in high-resolution remote sensing images.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121285544","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603490
Tang Liang-rui, Lou Jia, He Yuan
In the electrical power system, there is a wide range of communication services and service needs, and the realization of differentiated services of different types is quite necessary. Thus, based on the combination of characters and requirements of electric power communication business, this paper makes an argument for a routing and wavelength assignment algorithm based on prediction and layered graph model (PLGMRWA). Through the wavelength number prediction mechanism, it realizes the distribution of wavelength resource according to the needs, and uses the layered graph model to wholly solve the problems of route and wavelength assignment in comprehensive considerations of hops and the usage of wavelength. The simulation results indicate that, algorithm PLGMRWA has advantages of blocking capacity and wavelength resource utilization over algorithms GWAS (Grouped wavelength assignment strategy) and DPWA (Dynamic and priority-based wavelength assignment algorithm) for all levels of business in electrical power communication network.
{"title":"Electrical optical network RWA algorithm based on prediction and layered graph model","authors":"Tang Liang-rui, Lou Jia, He Yuan","doi":"10.1109/FSKD.2016.7603490","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603490","url":null,"abstract":"In the electrical power system, there is a wide range of communication services and service needs, and the realization of differentiated services of different types is quite necessary. Thus, based on the combination of characters and requirements of electric power communication business, this paper makes an argument for a routing and wavelength assignment algorithm based on prediction and layered graph model (PLGMRWA). Through the wavelength number prediction mechanism, it realizes the distribution of wavelength resource according to the needs, and uses the layered graph model to wholly solve the problems of route and wavelength assignment in comprehensive considerations of hops and the usage of wavelength. The simulation results indicate that, algorithm PLGMRWA has advantages of blocking capacity and wavelength resource utilization over algorithms GWAS (Grouped wavelength assignment strategy) and DPWA (Dynamic and priority-based wavelength assignment algorithm) for all levels of business in electrical power communication network.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"217 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116527063","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603273
Jing Zhang, Jiang Rong, Qiujun Liao, Jia Wang, Xiaoyun Yang
Based on grey clustering and entropy, this paper proposes a grey clustering classification model to evaluate cloud computing credibility. In the model, entropy is used to resolve clustering weight determination. At last, simulation experiments is given by Matlab to verify to the model, the results show that the model is effective for cloud computing credibility evaluation.
{"title":"Cloud computing credibility evaluation based on grey clustering and entropy","authors":"Jing Zhang, Jiang Rong, Qiujun Liao, Jia Wang, Xiaoyun Yang","doi":"10.1109/FSKD.2016.7603273","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603273","url":null,"abstract":"Based on grey clustering and entropy, this paper proposes a grey clustering classification model to evaluate cloud computing credibility. In the model, entropy is used to resolve clustering weight determination. At last, simulation experiments is given by Matlab to verify to the model, the results show that the model is effective for cloud computing credibility evaluation.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126665172","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603283
Huanlin Liu, Ling Yu
Moving force is very important for bridge design, structural analysis and structural health monitoring. Some studies on moving force identification (MFI) attract extensive attentions in the past decades. A novel two-step MFI method is proposed based on particle swarm optimization (PSO) and time domain method (TDM) in this study. The new proposed MFI method includes two steps. In the first step, the PSO is used to identify the constant loads without matrix inversion. In the second step, the conventional TDM is employed to estimate the rest time-varying loads where the Tikhonov regularization and general cross validation (GCV) are introduced to improve the MFI accuracy and to select optimal regularization parameters, respectively. A simply supported beam bridge subjected to moving forces is taken as a numerical simulation example to assess the performance of the proposed method. The illustrated results show that the new two-step MFI method can more effectively identify the moving forces compared to the conventional TDM and the improved Tikhonov regularization method, the proposed new method can provide more accurate MFI results on two moving forces under eight combinations of bridge responses.
{"title":"Moving force identification based on particle swarm optimization","authors":"Huanlin Liu, Ling Yu","doi":"10.1109/FSKD.2016.7603283","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603283","url":null,"abstract":"Moving force is very important for bridge design, structural analysis and structural health monitoring. Some studies on moving force identification (MFI) attract extensive attentions in the past decades. A novel two-step MFI method is proposed based on particle swarm optimization (PSO) and time domain method (TDM) in this study. The new proposed MFI method includes two steps. In the first step, the PSO is used to identify the constant loads without matrix inversion. In the second step, the conventional TDM is employed to estimate the rest time-varying loads where the Tikhonov regularization and general cross validation (GCV) are introduced to improve the MFI accuracy and to select optimal regularization parameters, respectively. A simply supported beam bridge subjected to moving forces is taken as a numerical simulation example to assess the performance of the proposed method. The illustrated results show that the new two-step MFI method can more effectively identify the moving forces compared to the conventional TDM and the improved Tikhonov regularization method, the proposed new method can provide more accurate MFI results on two moving forces under eight combinations of bridge responses.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127133439","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603279
Xiaofan Liu, Xinge Liu, Meilan Tang
This paper considers exponential stability of delayed neural networks(NNs). Based on some novel integral inequalities and a modified Lyapunov-Krasovskii functional(LKF), further result on delay-dependent exponential stability is obtained for the considered delayed neural networks in form of linear matrix inequality(LMI). The effectiveness of our result in this paper is also demonstrated by a numerical example.
{"title":"Further results on exponential stability of delayed neural networks","authors":"Xiaofan Liu, Xinge Liu, Meilan Tang","doi":"10.1109/FSKD.2016.7603279","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603279","url":null,"abstract":"This paper considers exponential stability of delayed neural networks(NNs). Based on some novel integral inequalities and a modified Lyapunov-Krasovskii functional(LKF), further result on delay-dependent exponential stability is obtained for the considered delayed neural networks in form of linear matrix inequality(LMI). The effectiveness of our result in this paper is also demonstrated by a numerical example.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127146595","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603295
Haitao Lin, Xiaobin Yang, Xiaopeng Yang
The system of max-product fuzzy relation inequalities(FRI) has been studied when it is consistent. In this paper, we study max-product FRI system when it is inconsistent. Based on some new concepts and theorems of approximate solutions, we provide an algorithm to solve its approximate solutions of the inconsistent FRI system. Also, we give an example to demonstrate the efficiency of the algorithm.
{"title":"Approximate solution to inconsistent system of max-product fuzzy relation inequalities","authors":"Haitao Lin, Xiaobin Yang, Xiaopeng Yang","doi":"10.1109/FSKD.2016.7603295","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603295","url":null,"abstract":"The system of max-product fuzzy relation inequalities(FRI) has been studied when it is consistent. In this paper, we study max-product FRI system when it is inconsistent. Based on some new concepts and theorems of approximate solutions, we provide an algorithm to solve its approximate solutions of the inconsistent FRI system. Also, we give an example to demonstrate the efficiency of the algorithm.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125482737","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：