时间序列预测的可重构计算研究
|
摘要
时间序列预测作为一种重要的数据分析方法一直受到研究者的广泛关注，然而在很多实际应用中，尤其在越来越多的嵌入式计算资源有限条件下，预测算法的高计算复杂度和庞大的硬件资源消耗成为制约现实应用的瓶颈。作为当前计算模式发展主流方向之一的可重构计算，凭借其定制并行计算的高性能和重构计算的高灵活性，为此问题提供了一种可行的解决方案。因此，将时间序列预测方法与可重构计算相融合，为有限计算资源条件下时间序列预测的高效计算方法研究提供一种新颖的解决思路和基础性的研究框架，具有明显的理论意义和实用价值。
目前，从算法层面开展的时间序列预测研究中，尚缺乏对可重构计算适应性的考虑。同时，高复杂度算法的高效计算需求与可重构计算平台中有限的硬件资源也存在直接的矛盾。鉴于此，本文通过比较分析，选择性能优越的最小二乘支持向量机(Least Squares Support Vector Machine, LS-SVM)时间序列预测算法作为本文研究工作的算法基础，从算法的可重构计算适应性改进、资源占用与计算效率平衡的可重构计算方法、以及有限可重构计算资源上的任务调度三个方面，开展时间序列预测的可重构计算方法研究，论文的主要研究工作包括以下内容：
（1）针对LS-SVM时间序列预测中存在的算法复杂度高，并且其中的常规线性方程组求解方法不适合可重构计算的问题，提出一种基于聚类的LS-SVM局部建模方法。该方法首先通过低复杂度K-means算法及合理的聚类数确定方法，缩减建模过程样本数量。然后采用改进Cholseky分解方法，实现计算量小且计算稳定性好的线性方程组求解算法。实验结果证明，在可控精度损失条件下，该方法较LS-SVM算法具有更高的训练和预测效率，为后续可重构计算研究提供了合适的算法基础。
（2）在LS-SVM算法进行可重构计算适应性改进基础上，针对LS-SVM可重构计算实现中硬件资源占用与计算效率的矛盾问题，提出一种基于部分动态自重构系统的LS-SVM训练过程可重构计算实现方法。该方法在构建部分动态自重构计算系统体系结构基础上，采用时域复用、空域并行的计算结构，实现硬件资源占用与计算效率的平衡。实验结果表明，本文方法可明显提高硬件资源利用率，并具有良好的计算效率。
（3）为进一步满足复杂时间序列预测任务的高效计算需求，开展部分动态重构系统静态任务调度方法研究。针对其中存在的重构区划分方法及反碎片技术理想化、配置预取策略被忽视以及最优调度方法计算效率低等问题，在充分研究调度工作机制和最优调度方法的基础上，提出一种基于改进异构最早完成时间表调度方法的启发式算法。该方法在重构区静态划分策略下，以提高利用率和灵活性为原则确定重构区规模，并改进最低水平线法，减少片上布局过程的碎片产生；进而通过积极的配置预取和任务插入，减少配置器和重构区空闲时间。实验结果表明，该调度算法与最优调度算法相比，具有良好的调度性能，同时可以获得明显的运行效率提升。
Time series forcasting as a practically useful data analysis method has drawntremendous attentions in scientific research. However, complicated algorithmarchitecture and huge hardware resource occupation have greatly limitedapplications of time series forecasting. Together with increasing applications oftime series forecasting in general embedded systems, efficient forcasting methodsconsuming only restricted hardware resources have become more and morenecessary. As one of the mainstreams in the development of computing,reconfigurable computing(RC) provides a feasible solution to this problem withcustomized high performance parallel computing and flexible reconfiguration.Therefore, the fusion of time series forecasting and RC can provide a novelsolution and basic research framework to the high performance computingresearch of time series forecasting consuming only limited computing resource.So this fusion research has both theoretical and practical significance.
Although a number of research works have been proposed to improve thecomputing efficiency of time series forecasting algorithms, applications of RChave surprisingly been ignored in most of these works. Meanwhile, there is anobvious conflict between the efficient computing requirement of complicatedtime series forcasting algorithms and limited RC hardware resources. Given allthese, with careful comparison and analysis, we choose least squares supportvector machine(LS-SVM) as the target algorithms in this paper. Furthermore, wefocus on the RC research of time series forecasting from the perspective of RCapplicability improvement of LS-SVM, RC method with balanced resourceoccupation and computing efficiency, and task scheduling in limited hardwareresources. We here in mainly describe the following work.
(1)In LS-SVM based time series forecasting, the high computing complexityand improper linear equations solving method influence its applicability ofreconfigurable computing. To solve this problem, we propose a local modelingmethod of LS-SVM based on clustering strategy. Firstly, we adopt K-meansclustering algorithm with reasonable cluster number to get the local modelingsamples and reduces space complexity. Then, modified Cholesky factorization isused to get analytical, stable and low complex solutions of linear equations.Experiments show that CLS-SVM could obtain better training and predictionefficiency compared to LS-SVM with acceptable precision drop. Moreover, it can reduce the space complexity and get analytical, stable and low complex solutions.All these property make it suitable for reconfigurable computing.
(2)To solve the conflict of hardware resources occupation and computingefficiency in the reconfigurable computing of LS-SVM, we propose an LS-SVMreconfigurable computing method based on partial dynamic self-reconfigurablesystem. First we present partial dynamical self-reconfigurable systemarchitecture. Then we propose a time multiplexing and spacial parallel computingstructure to balance the hardware resource occupation and computing efficiency.Experiment results show that when compared to non-time multiplexingarchitecture, our method has high hardware utilization while achieving highcomputing efficiency. This makes the embedded application of LS-SVM withrestricted hardware resources possible.
(3)To realize efficient time series forecasting with RC, we explore the taskscheduling in FPGA based partial dynamic RC system. To solve the existingproblems including improper assumptions in reconfigurable partition(RP)divisionmethods and anti-fragments technology, ignorance of configuration prefetch andlow efficiency in optimized scheduling method, with careful study of schedulingmechanism and optimal scheduling method, we propose a heuristic based ontypical heterogeneous multi-core system list scheduling method. With staticdivision strategy of RP, RP sizes are decided under the principle of improvingflexibility and resources utilization. Then chip layout is realized with modifiedminimum horizontal method to reduce fragments. Finally, by active configurationprefetch and task insertion under the limit of serial configuration, we improve thescheduling performance. Simulation experiments and real world application toCLS-SVM algorithm all show that, our heuristic method has a good schedulingperformance and generality. Meanwhile, our heuristic also achieves an obviousefficiency improvement compared to optimal scheduling algorithm.
目前，从算法层面开展的时间序列预测研究中，尚缺乏对可重构计算适应性的考虑。同时，高复杂度算法的高效计算需求与可重构计算平台中有限的硬件资源也存在直接的矛盾。鉴于此，本文通过比较分析，选择性能优越的最小二乘支持向量机(Least Squares Support Vector Machine, LS-SVM)时间序列预测算法作为本文研究工作的算法基础，从算法的可重构计算适应性改进、资源占用与计算效率平衡的可重构计算方法、以及有限可重构计算资源上的任务调度三个方面，开展时间序列预测的可重构计算方法研究，论文的主要研究工作包括以下内容：
（1）针对LS-SVM时间序列预测中存在的算法复杂度高，并且其中的常规线性方程组求解方法不适合可重构计算的问题，提出一种基于聚类的LS-SVM局部建模方法。该方法首先通过低复杂度K-means算法及合理的聚类数确定方法，缩减建模过程样本数量。然后采用改进Cholseky分解方法，实现计算量小且计算稳定性好的线性方程组求解算法。实验结果证明，在可控精度损失条件下，该方法较LS-SVM算法具有更高的训练和预测效率，为后续可重构计算研究提供了合适的算法基础。
（2）在LS-SVM算法进行可重构计算适应性改进基础上，针对LS-SVM可重构计算实现中硬件资源占用与计算效率的矛盾问题，提出一种基于部分动态自重构系统的LS-SVM训练过程可重构计算实现方法。该方法在构建部分动态自重构计算系统体系结构基础上，采用时域复用、空域并行的计算结构，实现硬件资源占用与计算效率的平衡。实验结果表明，本文方法可明显提高硬件资源利用率，并具有良好的计算效率。
（3）为进一步满足复杂时间序列预测任务的高效计算需求，开展部分动态重构系统静态任务调度方法研究。针对其中存在的重构区划分方法及反碎片技术理想化、配置预取策略被忽视以及最优调度方法计算效率低等问题，在充分研究调度工作机制和最优调度方法的基础上，提出一种基于改进异构最早完成时间表调度方法的启发式算法。该方法在重构区静态划分策略下，以提高利用率和灵活性为原则确定重构区规模，并改进最低水平线法，减少片上布局过程的碎片产生；进而通过积极的配置预取和任务插入，减少配置器和重构区空闲时间。实验结果表明，该调度算法与最优调度算法相比，具有良好的调度性能，同时可以获得明显的运行效率提升。
Time series forcasting as a practically useful data analysis method has drawntremendous attentions in scientific research. However, complicated algorithmarchitecture and huge hardware resource occupation have greatly limitedapplications of time series forecasting. Together with increasing applications oftime series forecasting in general embedded systems, efficient forcasting methodsconsuming only restricted hardware resources have become more and morenecessary. As one of the mainstreams in the development of computing,reconfigurable computing(RC) provides a feasible solution to this problem withcustomized high performance parallel computing and flexible reconfiguration.Therefore, the fusion of time series forecasting and RC can provide a novelsolution and basic research framework to the high performance computingresearch of time series forecasting consuming only limited computing resource.So this fusion research has both theoretical and practical significance.
Although a number of research works have been proposed to improve thecomputing efficiency of time series forecasting algorithms, applications of RChave surprisingly been ignored in most of these works. Meanwhile, there is anobvious conflict between the efficient computing requirement of complicatedtime series forcasting algorithms and limited RC hardware resources. Given allthese, with careful comparison and analysis, we choose least squares supportvector machine(LS-SVM) as the target algorithms in this paper. Furthermore, wefocus on the RC research of time series forecasting from the perspective of RCapplicability improvement of LS-SVM, RC method with balanced resourceoccupation and computing efficiency, and task scheduling in limited hardwareresources. We here in mainly describe the following work.
(1)In LS-SVM based time series forecasting, the high computing complexityand improper linear equations solving method influence its applicability ofreconfigurable computing. To solve this problem, we propose a local modelingmethod of LS-SVM based on clustering strategy. Firstly, we adopt K-meansclustering algorithm with reasonable cluster number to get the local modelingsamples and reduces space complexity. Then, modified Cholesky factorization isused to get analytical, stable and low complex solutions of linear equations.Experiments show that CLS-SVM could obtain better training and predictionefficiency compared to LS-SVM with acceptable precision drop. Moreover, it can reduce the space complexity and get analytical, stable and low complex solutions.All these property make it suitable for reconfigurable computing.
(2)To solve the conflict of hardware resources occupation and computingefficiency in the reconfigurable computing of LS-SVM, we propose an LS-SVMreconfigurable computing method based on partial dynamic self-reconfigurablesystem. First we present partial dynamical self-reconfigurable systemarchitecture. Then we propose a time multiplexing and spacial parallel computingstructure to balance the hardware resource occupation and computing efficiency.Experiment results show that when compared to non-time multiplexingarchitecture, our method has high hardware utilization while achieving highcomputing efficiency. This makes the embedded application of LS-SVM withrestricted hardware resources possible.
(3)To realize efficient time series forecasting with RC, we explore the taskscheduling in FPGA based partial dynamic RC system. To solve the existingproblems including improper assumptions in reconfigurable partition(RP)divisionmethods and anti-fragments technology, ignorance of configuration prefetch andlow efficiency in optimized scheduling method, with careful study of schedulingmechanism and optimal scheduling method, we propose a heuristic based ontypical heterogeneous multi-core system list scheduling method. With staticdivision strategy of RP, RP sizes are decided under the principle of improvingflexibility and resources utilization. Then chip layout is realized with modifiedminimum horizontal method to reduce fragments. Finally, by active configurationprefetch and task insertion under the limit of serial configuration, we improve thescheduling performance. Simulation experiments and real world application toCLS-SVM algorithm all show that, our heuristic method has a good schedulingperformance and generality. Meanwhile, our heuristic also achieves an obviousefficiency improvement compared to optimal scheduling algorithm.
引文
[1]何书元.应用时间序列分析[M].北京:北京大学出版社,2005:1-2.
[2]韩敏.混沌时间序列预测理论与方法[M].北京:中国水利水电出版社，2007:155-185.
[3] Zhu Yuelong, Qian Jihong, Fan Qingsong, et al. Study on Hydrology TimeSeries Prediction Based on Wavelet-neural Networks[C].In Proceedings ofthe8th IEEE/ACIS International Conference on Computer and InformationScience,2009.411-415.
[4] Guo Z. X., Wong W. K., Li M.. Sparsely connected neural network-basedtime series forecasting[J]. Information Science,2012,19:54-71.
[5] NASA. Space Technology Roadmaps for the Fourteen TechnologyAreas/Technology Area Strategic Roadmaps and BreakdownStructure[EB/OL].2011. http://www.nasa.gov/offices/oct/home/roadmaps/index.html.
[6]张宝珍.预测与健康管理技术的发展及应用[J].测控技术,2008,23(2):5-7.
[7]彭宇,刘大同,彭喜元.故障预测与健康管理技术综述[J].电子测量与仪器学报,2010,4(1):1-8.
[8] Michael G. Pecht. Prognostics and Health Management of Electronics[M].John Wiley&Sons. Inc., Hoboken, New Jersey,2008:3-20.
[9] Mohanty S, Chattopadhyay A, Wei J, et al. Unsupervised Time-SeriesDamage State Estimation of Complex Structure Using UltrasoundBroadband Based Active Sensing[J]. Structural Durability&HealthMonitoring Journal,2010,130(1):101-124.
[10]许丽佳,王厚军,龙兵.一种状态检测与健康评估方法及其在模拟电路中的应用[J].计算机辅助设计与图形学学报,2008,20(12):1150-1556.
[11] Gang Niu, Bo-Suk Yang. Dempster-Shafer Regression for Multi-Step-ahead Time Series Prediction towards Data-driven Machinery Prognosis[J].Mechanical Systems and Signal Processing,2009,23(1):740-751.
[12] T. Makimoto. The rising wave of Field Programmability[C]. In Proceedingsof the FPL2000,2000:1-6.
[13] Monteneqro S, Roser H P, Huber F. BOSS: software and FPGAmiddleware for the‖flying laptop‖microsatellite[C]. In Proceedings ofDASIA,2005:401-405.
[14]罗赛.可重构计算系统体系结构与实现[D].合肥.中国科学技术大学,2006:16-19.
[15] Srihari Cadambi, Igor Durdanovic, Venkata Jakkula, et al.A MassivelyParallel FPGA-based Coprocessorfor Support Vector Machines[C]. InProceedings of the17th IEEE Symposium on Field Programmable CustomComputing Machines(FCCM'09),2009:115-122.
[16] Science Staff. Challenges and Opportunities[J]. Science Magazine,2011,331(6018):692-693.
[17] Serdar Uckun, Kai Goebel, Peter J F Lucas. Standardizing ResearchMethods for Prognostics[C]. In Proceedings of the International Conferenceon Prognostic and Health Management,2008:1-10.
[18] George E P Box, Gwilym M Jenkins, Gregory C Reinsel. Time SeiresAnalysis: Forecasting and Control[M]. Third Edition. New Jersey: JohnWiley&Sons Inc,1994:2-45.
[19] Jianqing Fan, Qiwei Yao. Nonlinear Time Series: Nonparametric andparametric methods[M]. New York: Springer-Verlag,2003:12-123.
[20] Engle R F. Autoregressive conditional heteroscedasticity with estimates ofthe variance of U.K. inflation[J]. Econometrica1,1982,50:987-1008
[21] Bollerslev T. Generalised autoregressive conditional heteroscedasticity[J].Journal of Econometrics,1986,31:307-327
[22] Tong H, Lim K S. Threshold autoregression, limit cycles and cyclicaldata[J]. Journal of Royal Statistical Society,1980, B42:245-292
[23] William W S Wei.Time Series Analysis: Univariate and MultivariateMethods[M]. Second Edition.London: Pearson EducationInc,2006:68-85,366-380.
[24] Zhang G, Patuwo B, Hu M. Forecasting with Artificial Neural Networks:The State of the Art[J]. International Journal of Forecasting,1998,14(1):35-62.
[25] Werbos P. Back propagation Through Time: What It Does and How to DoIt[J]. Proc of IEEE,1990,78(10):1550-1560.
[26]张军峰,胡寿松.基于一种新型聚类算法的RBF神经网络混沌时间序列预测[J].电子学报,2007,56(2):713-719.
[27] Giles C L, Lawre S, Tsoi A C. Nosiy Time Series Prediction UsingRecurrent Neural Networks and Grammatical Inference[J]. MachineLearning,2001,44(1):335-356.
[28] Herbert Jaeger.A Tutorial on Training Recurrent NeuralNetworks, CoveringBPPT, RTRL, EKF and the "Echo State Network" Approach[EB/OL].2003.http://www.faculty.jacobs-university.de/hjaeger/pubs.html.
[29] H Jaeger, H Haas. Harnessing Nonlinearity: Predicting Chaotic Systemsand Saving Energy in Wireless Communication[J]. Science,2004,10(2):78-80.
[30] Lean Yu, Shouyang Wang, Kin Keung Lai. A Neural-network-basedNonlinear Metamodeling Approach to Financial Time Series Forecasting[J].Applied Soft Computing,2009,9(2):563-574.
[31] Shu-liang Liu, Zhi-qiang Hu,Xiu-kai Chi. Power Load Forecasting Basedon Neural Network and Time Series[C]. In Proceedings of theInternationalConference on Wireless Communications, Networking and MobileComputing,2009:1-5.
[32] Girish K Jha, Parimala Thulasiraman, Ruppa K Thulasiram. PSO BasedNeural Network for Time Series Forecasting[C]. In Proceedings of theInternational Joint Conference on Neural Networks,2009:1422-1427.
[33] Xindi Cai, Nian Zhang, Ganesh K. Venayagamoorthy, Donald C. Wunsch II.Time series prediction with recurrent neural networks trained by a hybridPSO-EA algorithm[J]. Neuro Computing,2007,70(13-15):2342-2353.
[34] Dongqing Zhang, XuanxiNing, Xueni Liu, et al. Nonlinear Time SeriesForecasting with Dynamic RBF Neural Networks[C]. In Proceedings of the7thWorld Congress on Intelligent Control and Automation,2008:6988-6993.
[35] Hans Georg Seedig, Ralph Grothmann, Thomas A Runkler. Forecasting ofClustered Time Series withRecurrent Neural Networks and A FuzzyClustering Scheme[C]. In Proceedings of the International Joint Conferenceon Neural Networks,2009:2846-2853.
[36] Lili Qu, Yan Chen, Zhenfeng Liu. Time Series Forecasting Model withError Correction byStructure Adaptive RBF Neural Network[C].InProceedings of the6th World Congress on Intelligent ControlandAutomation,2006:6831-6835.
[37] Vapnik V. The Nature of Statistical Learning Theory[M]. Second Edition.New York: Springer-Verlag,1999:1-42.
[38] VN Vapnik.统计学习理论的本质[M].张学工,译.北京:清华大学出版社,2000:1-67.
[39]邓乃扬,田英杰.数据挖掘中的新方法:支持向量机[M].北京:科学出版社,2004:255-259.
[40] Johannes Mager, Ulrich Paasche, Bernhard Sick. Forecasting FinancialTime Series with Support Vector Machines Based on Dynamic Kernels[C].In Proceedings of the IEEE Conference on Soft Computing in IndustrialApplications,2008:252-257.
[41] Nicholas I Sapankevych, Ravi Sankar. Time Series Prediction UsingSupport Vector Machines: A Survey[J]. IEEE Computational IntelligenceMagazine,2009,4(2):24-38.
[42] Suykens J, Vandewalle J. Least Square Support Vector MachineClassifiers[J]. Neural Processing Letters,1999,9(3):293-300.
[43] Tony Van Gestel, Johan A KSuykens, Dirk-Emma Baestaens, et al.Financial Time Series Prediction Using Least SquaresSupport VectorMachines Within the EvidenceFramework[J]. IEEE Transactions on NeuralNetwork,2001,12(4):809-821.
[44] Yang Li. Short-Term Load Forecasting Based on LS-SVMOptimized byBCC Algorithm[C]. In Proceedings of the15th International ConferenceonIntelligent System Applications to Power Systems,2009:1-5.
[45] Zhang Fuwei, Chen Wei. Application of Grey LS-SVM in Mid and LongTermPower Load Forecasting[C]. In Proceedings of the InternationalConference onManagement and Service Science,2009:1-4.
[46] Yunhong Gao, Yibo Li. Fault Prediction Model Based on Phase SpaceReconstruction and Least SquaresSupport Vector Machine[C]. InProceedings of the9th International Conference on Hybrid IntelligentSystems,2009:464-467.
[47] Zhenrui Peng, Xijuan Yang, Wenzhe Qi. LS-SVM Based Soft Sensoring[C].In Proceedings of the7th World Congress on Intelligent Control andAutomation,2008:4320-4324.
[48]张英,苏宏业,褚健.基于模糊最小二乘支持向量机的软测量建模[J].控制与决策,2005,20(6):621-624.
[49] Wei Shen, Yunyun Zhang, Xiaoyong Ma. Stock Return Forecast with LS-SVM and Particle Swarm Optimization[C]. International Conference onBusiness Intelligence and Financial Engineering,2009:143-147.
[50] Yang Zhang, Yuncai Liu. Missing Traffic Flow Data Prediction UsingLeast Squares SupportVector Machines in Urban Arterial Streets[C]. InProceedings of the IEEE Symposium onComputational Intelligence andData Mining,2009:76-83.
[51] Wang Shaojun, Guo Jia, Liu Qi, Peng Xiyuan. On-Line Traffic Forecastingof Mobile Communication System[C]. International Conference onPervasive Computing Signal Processing and Applications,2010:97-100.
[52]邓聚龙.灰色系统理论教程[M].武汉:华中理工大学出版社,1990:1-38.
[53] Liehr S, Pawelzik K, Kohlmorgen J, et al. Hidden Markov Mixtures ofExperts for Prediction of Non-stationary Dynamics[C]. In Proceedings ofthe1999IEEE Signal Processing Society Workshop,1999:195-204.
[54] SimoS rkk, Aki Vehtari, JoukoLampinen. CATS Benchmark Time SeriesPrediction by Kalman Smoother with Cross-validated Noise Density[J].Neuro Computing,2007,70(13–15):2331-2341.
[55] Xuedong Wu, Zhihuan Song. Online Chaotic time-series prediction with thederivative-free extended Kalman filter[C]. In Proceedings of the7th WorldCongress onIntelligent Control and Automation,2008:2360-2364.
[56] Gordon N J, Salmond D J, Smith A F M. Novel Approach toNonlinear/Non-Gaussian Bayesian State Estimation[J]. Radar and SignalProcessing,1993,140(2):107-113.
[57] Zhang Wei, Zhang Feng. Overview of Data-driven Time Series ForecastingMethods[J]. Journal of Shanxi University of Science and Technology,2010,28(3):22-27.
[58] J Wichard. Forecasting the NN5Time Series with Hybrid Models[J].International Journal of Forecasting,2011,27(3):700-707.
[59] J Wichard, M Ogorzalek. Time Series Prediction with EnsembleModels[C].In Proceedings of the IEEE International Joint Conference onNeural Networks,2004:1625-1630.
[60]窦勇,谢伦国.可重构计算相关问题探讨[J].中国计算机学会通讯,2009,5(11):34-42.
[61] Dehon A, Wawraynek J. Reconfigurable Computing: What, Why, andImplication for Design Automation[C]. In Proceedings of the36th DesignAutomation Conference,1999:610-615.
[62] Compton K, Hauck S. Reconfigurable Computing: A Survey of Systemsand Software[J]. ACM Computing Surveys,2002,34(2):171-210.
[63] Estrin G, BusselB, Turn R, et al. Parallel Processing in A RestructurableComputer System[J]. IEEE Transactions on Electronic Computers,1963,12(5):747-755.
[64] Abbott A Lynn, Athanas Peter M, ChenLuna, et al.Finding Lines andBuilding Pyramids with Splash2[C]. In Proceedings of the IEEE Workshopon FPGAs for Custom Computing Machines,1994.155-163.
[65] Nikolaos Alachiotis, Simon A Berger, Alexandros Stamatakis. AcceleratingPhylogeny-Aware Short DNA Read Alignment with FPGAs[C]. InProceedings of the IEEE International Symposium on Field-ProgrammableCustom Computing Machines(FCCM),2011:226-233.
[66] Kirsch S, Rettig F, Hutter D, et al. An FPGA-Based High-Speed, Low-Latency Processing System for High-Energy Physics[C]. In Proceedings ofthe International Conference on Field Programmable Logic andApplications(FPL),2010:562-567.
[67] Haohuan Fu, Robert G Clapp. Eliminating the Memory Bottleneck: AnFPGA-Based Solution for3D Reverse Time Migration[C]. In Proceedingsof the19th ACM/SIGDA International Symposium on Field ProgrammableGate Arrays(FPGA'11),2011:65-74.
[68] Robin Pottathuparambil, Jack Coyne, Jeffrey Allred,et al. Low-LatencyFPGA Based Financial Data Feed Handler[C]. In Proceedings of the IEEE19th Annual International Symposium on Field-Programmable CustomComputing Machines(FCCM),2011:93-96.
[69] Anil Madhavapeddy, Satnam Singh. Reconfigurable Data Processing forClouds[C]. In Proceedings of the IEEE19th Annual InternationalSymposium on Field-Programmable Custom Computing Machines(FCCM),2011:141-145.
[70] Orlowska-Kowalska T, Kaminski M. FPGA Implementation of theMultilayer Neural Network for the Speed Estimation of the Two-MassDrive System[J]. IEEE Transactions on Industrial Informatics,2011,7(3):436-445.
[71]任小西.基于可重构计算的高可靠星载计算机体系结构研究[D].长沙,湖南大学,2007:7-22.
[72] Scott Hauck, André DeHon. Reconfigurable Computing: the Theory andPractice of FPGA-Based Computation[M]. San Fransisco: MorganKaufmann,2008:475-496.
[73] Callahan T J, Hauser J R, Wawrzynek J. The Garp architecture and Ccompiler[J]. Computer,2000,33(4):62-69.
[74] C Ebeling, D C Cronquist, P Franklin. RaPiD: Reconfigurable PipelinedDatapath[C]. In proceedings of the6th International Workshop on Field-Programmable Logic and Applications,1996:126-135.
[75] Goldstein S, Schmit H, Moe M, et al. PipeRench: A Coprocessor forStreaming Multimedia Acceleration[C]. In Proceedings ofthe26thInternational Symposium on Computer Architecture,1999:28-39.
[76] John D Davis, Charles P Thacker, Chen Chang. BEE3: RevitalizingComputer Architecture Research[EB/OL]. http://research.microsoft.com/projects/BEE3/.
[77] Liu C L, Layland J. Scheduling Algorithm for Multiprogramming in aHard-real-time Environment[J]. Journal of the ACM,1973,20(l):46-61.
[78] Kyprianos Papadimitriou, ApostolosDollas, Scott Hauck. Performance ofPartial Reconfiguration in FPGA Systems: A survey and a Cost Model[J].ACM Transactions on Reconfigurable Technology and Systems,2011,4(4):1-28.
[79]李涛.动态重构系统若干关键问题的研究[D].天津.南开大学,2007:9-11.
[80] Ullmann M, Hubner M, Grimm B, et al. An FPGA Run-time System forDynamical On-demand Reconfiguration[C]. In Proceedings of the18thInternational Parallel and Distributed Processing Symposium,2004:135-142.
[81]王峰.部分重构计算系统的关键技术研究[D].合肥.中国科学技术大学,2007:8-15.
[82] Amos R Omondi, J agath C Rajapakse, Mariusz Bajger. FPGAImplementations of Neural Networks[J]. IEEE Transactions on NeuralNetworks,2007,18(5):21-32.
[83] Amos R, Ormondi, Jagath C, et al. FPGA Implementationsof NeuralNetworks[M]. Berlin: Springer,2006:6-76.
[84] Benjamin Schrauwen, Michiel D’Haene, David Verstraeten, et al. CompactHardware Liquid State Machines on FPGA for Real-time SpeechRecognition[J]. Neural Networks,2008,21(2-3):511-523.
[85] Preethy A P, Radhakrishnan D, Omondi A R. A High PerformanceResidue-number-system Multiply-accumulate Unit[C]. In Proceedings ofthe11th ACM Great Lakes Symposium on VLSI,2001:145-149.
[86] Davide Anguita, Andrea Boni, Sandro Ridella. A Digital Architecture forSupport Vector Machines: Theory, Althorithm, and FPGAImplementation[J]. IEEE Transaction on Neural Networks,2003,14(5):993-1009.
[87] Choi W Y, Ahn D, Pan S B, et al. SVM-Based Speaker Verification Systemfor Match-on-Card and Its Hardware Implementation[J]. ETRI Journal,2006,28(3):320-328.
[88]曹葵康.支持向量机加速方法及应用研究[D].杭州.浙江大学,2010:49-50.
[89] Markos Papadonikolakis, Christos-Savvas Bouganis. A HeterogeneousFPGA Architecture for Support Vector Machine Training[C]. InProceedings of the18th IEEE Annual International Symposium on Field-Programmable Custom Computing Machines(FCCM),2010:211-214.
[90] Faisal MKhan, Mark G Arnold, William M Pottenger. Hardware-BasedSupport Vector Machine Classification in Logarithmic Number Systems[C].In Proceedings of the IEEE International Symposium on Circuits andSystems(ISCAS),2005:5154-5157.
[91] Omar Pina-Ramirez, Raquel Valdes-Cristerna, Oscar Yanez-Suarez. AnFPGA Implementation of Linear Kernel Support Vector Machines[C]. InProceedings of the IEEE International Conference on ReconfigurableComputing and FPGA's(ReConFig'06),2006:1-6.
[92]刘涵.王博.尹嵩.刘丁.一种基于FPGA的新的SVM硬件实现方法[J].仪器仪表学报,2010,31(10):2235-2241.
[93] Depeng Yang, Gregory D Peterson, Husheng Li, et al. An FPGAImplementation for Solving Least Square Problem[C]. In Proceedings ofthe17th IEEE Symposium on Field-Programmable Custom ComputingMachines,2009:303-306.
[94] Guiming Wu, Yong Dou, Gregory D Peterson. Blocking LU Decompositionfor FPGAs[C]. In Proceedings of the18th IEEE Annual InternationalSymposium on Field-Programmable Custom Computing Machines,2010:211-214.
[95] Beretta Ivan, Rana Vincenzo, Atienza David, et al. A Mapping Flow forDynamically Reconfigurable Multi-core System-on-chip Design[J]. IEEETransactions on Computer-Aided Design of Integrated Circuits andSystems,2011,30(8):1211-1224.
[96]殷进勇.可重构系统中实时任务调度算法研究[D].哈尔滨：哈尔滨工程大学,2010:5-10.
[97]刘勇.嵌入式可重构计算系统及其任务调度机制的研究[D].上海:中国科学院上海微系统与信息技术研究所,2006:45-51.
[98] Waheed Ahmed, Muhammad Shafique, Lars Bauer, et al. Run-TimeResource Allocation for Simultaneous Multi-tasking in Multi-coreReconfigurable Processors[C]. In Proceedings of the IEEE19th AnnualInternational Symposium on Field-Programmable Custom ComputingMachines,2011:29-32.
[99] Yang Qu, Juha-Pekka Soininen, Jari Nurmi. Static Scheduling Techniquesfor Dependent Tasks on Dynamically Reconfigurable Devices[J]. Journal ofSystems Architecture,2007,53(11):861-876.
[100] Christoph Steiger, Herbert Walder, Marco Platzner. Operating Systems forReconfigurable Embedded Platforms: Online Scheduling of Real-TimeTasks[J]. IEEE Transactions on Computers,2004,53(11):1393-1407.
[101] Banerjee S, Bozorgzadeh E, Dutt N. Parlgran: Parallelism GranularitySelection for Scheduling Task Chains on Dynamically ReconfigurableArchitectures[C]. In Proceedings of the Asia and South Pacific DesignAutomation Conference (ASP-DAC),2006:491-496.
[102] Compton K, Li Z, Cooley J, et al. Configuration Relocation andDefragmentation for Run-time Reconfigurable Computing[J]. IEEETransactions on VLSI Systems,2002,10(3):209-220.
[103] Li Z, Hauck S. Configuration Prefetching Techniquesfor PartialReconfigurable Coprocessor with Relocation and Defragmentation[C]. InProceedings of the ACM/SIGDA Tenth International Symposium on Field-programmable Gate Arrays,2002:187-195.
[104] Redaelli F, Santambrogio MD, Ogrenci Memik S. An ILP Formulation forthe Task Graph Scheduling Problem Tailored to Bi-dimensionalReconfigurable Architectures[C]. In Proceedings of the InternationalConference on Reconfigurable Computing and FPGAs(ReConFig'08),2008:97-102.
[105] IkbelBelaid, Fabrice Muller, Maher Benjemaa. Optimal Static Schedulingof Real-Time Dependent Tasks on Reconfigurable Hardware Devices[C]. InProceedings of IEEE International Conference on Communications,Computing and Control Applications (CCCA),2011:1-6.
[106] Zonghua Gu, Mingxuan Yuan, Xiuqiang He. Optimal Static TaskScheduling on Reconfigurable Hardware Devices Using Model-Checking[C]. In Proceedings of the13th IEEE Real Time and EmbeddedTechnology and Applications Symposium(RTAS'07),2007:32-44.
[107] Panainte EM, Bertels K., Vassiliadis S. Compiler-driven FPGA-areaAllocation for Reconfigurable Computing[C]. In Proceeding of the Design,Automation, and Test in Europe(DATE),2006:369-374.
[108] Mingxuan Yuan, ZonghuaGu, Xiuqiang He, et al. Hardware/SoftwarePartitioning and Pipelined Scheduling on Runtime ReconfigurableFPGAs[J]. ACM Transactions on Design Automation of Electronic Systems(TODAES),2010,15(2):13-54.
[109] Wenyin Fu, Katherine Compton. An Execution Environment forReconfigurable Computing[C]. In Proceedings of the13th Annual IEEESymposium on Field-Programmable Custom Computing Machines,2005:149-158.
[110] Banerjee S, Bozorgzadeh E, Dutt N. Physically-aware HW-SW Partitioningfor Reconfigurable Architectures with Partial Dynamic Reconfiguration[C].In Proceedings of the Design Automation Conference,2005:335-340.
[111] Banerjee S, Bozorgzadeh E, Dutt N. Integrating Physical Constraints inHW-SW Partitioning for Architectures with Partial DynamicReconfiguration[J]. IEEE Transactions on Very Large Scale Integration(VLSI) Systems,2006,14(11):1189-1202.
[112]刘彦.异构多核片上系统的任务调度及应用研究[D].长沙:湖南大学,2009:54-63.
[113] Ghiasi S, Sarrafzadeh M. Optimal Reconfiguration SequenceManagement[C]. In Proceedings of the Asia and South Pacific DesignAutomation Conference (ASP-DAC),2003:359-365.
[114] Juan Antonio Clemente, Carlos González, Javier Resano, et al. A HardwareTask-Graph Scheduler for Reconfigurable Multi-tasking Systems[C]. InProceedings of the International Conference on Reconfigurable Computingand FPGAs(ReConFig'08),2008:79-84.
[115] Justin Teller, Füsun zgüner, Robert Ewing. Scheduling Tasks onReconfigurable Hardware with a List Scheduler[C]. In Proceedings of theInternational Symposium on Parallel&Distributed Processing,2009:1-4.
[116] Suykens J, Vandewalle J. Least Square Support Vector MachineClassifiers[J]. Neural Processing Letters,1999,9(3):293-300.
[117] Suykens J A K, L Lukas, P Van Dooren, et al. Least Squares SupportVector Machine Classifiers: a Large Scale Algorithm[C]. In Proceedings ofthe European Conference on Circuit Theory and Design,1999:839-842.
[118] CHU W, ONG C J, KEERTHI S S. An Improved Conjugate GradientScheme to the Solution of Least Squares SVM[J]. IEEE Transaction onNeural Networks,2005,16(2):498-501.
[119] Edgar Osuna, Robert Freund, Federico Girosi. An Improved TrainingAlgorithm for Support VectorMachines[C]. In Proceedings of the IEEEWorkshop on Neural Networks for Signal Processing,1997:276-285.
[120] Platt J. Fast Training of Support Vector Machines Using SequentialMinimal Optimization[M]. Massachusetts: MIT Press,1999:185-208.
[121] Keerthi S S, Shevade S K. SMO Algorithm for Least-Squares SVMFormulations[J]. Neural Computation,2003,15(2):487-507.
[122] W. Chu, C. J. Ong, S. S. Keerthy. An improved conjugate gradient methodscheme to the solution of least squares SVM[J]. IEEE Transaction onNeural Networks,2005,16(2):498–501.
[123]张浩然,汪晓东.回归最小二乘支持向量机的增量和在线式学习算法[J].计算机学报,2006,29(3):400-406.
[124]叶美盈,汪晓东,张浩然.基于在线最小二乘支持向量机回归的混沌时间序列预测[J].物理学报,2005,54(6):2568-2572.
[125]郭振凯,宋召青,毛剑琴.一种改进的在线最小二乘支持向量机回归算法[J].控制与决策,2009,24(1):145-148.
[126]吴春国.广义染色体算法与迭代式最小二乘支持向量机回归算法研究
[D].吉林大学,2006:49-70.
[127] Marcelo Espinoza, Johan A. K. Suykens, Bart De Moor. Load Forecastingusing Fixed-Size Least Squares Support Vector Machines[J].Computational Intelligence and Bioinspired Systems,2005,35(12):1018-1026.
[128] Suykens J A K, Lukas L, Vandewalle J. Sparse Approximation Using LeastSquaresSupportVectorMachine[C]. In Proceedings of the IEEEInternationalSymposium onCircuits and Systems,2000:757-760.
[129] Gavin C Cawley, Nicola L C Talbot. Improved sparse least-squares supportvector machines[J]. Neurocomputing,2002,48:1025-1031
[130] Kruif B J, Vries J A. Pruning error minimization in least squares supportvector machines[J]. IEEE Trans on Neural Networks,2003,14(3):696-702.
[131] Hoegaerts L, Suykens J A K, Vandewalle J, et al. A Comparison of PruningAlgorithms for Sparse Least Squares Support Vector Machines[C]. InProceedings of the International Conference on Neural InformationProcessing,2004:1247-1253.
[132]赵永平,孙健国.关于稀疏最小二乘支持向量机的改进剪枝算法[J].系统工程理论与实践,2009,29(6):166-171.
[133]赵永平,孙健国.基于薄壁管法的稀疏最小二乘支持向量回归机[J].电机与控制学报,2009,13(4):581-585.
[134]周欣然,滕召胜,易钊.构造稀疏最小二乘支持向量机的快速剪枝算法[J].电机与控制学报,2009,13(4):626-630.
[135] Valyon J, Horvath G. A Sparse Least Squares Support Vector MachineClassifier[C]. In Proceedings of the IEEE International Conference onNeural Networks,2004:543-548.
[136]甘良志,孙宗海,孙优贤.稀疏最小二乘支持向量机[J].浙江大学学报(工学版),2007,41(2):245-248.
[137] Licheng Jiao, Liefeng Bo, Ling Wang. Fast Sparse Approximation for LeastSquares Support Vector Machine[J]. IEEE Transactions onNeural Networks,2007,18(3):685-697.
[138]赵永平,孙健国.一种快速稀疏最小二乘支持向量回归机[J].控制与决策,2008,12:1347-1352.
[139]陈爱军,宋执环,李平.基于矢量基学习的最小二乘支持向量机建模[J].控制理论与应用,2007,24(1):1-5.
[140]赵永平,孙健国.基于线性无关度的稀疏最小二乘支持向量回归机[J].中国图象图形学报,2009,14(6):1136-1140.
[141] Zongwu Cai, Ram C. Tiwari. Application of a local linear autoregressivemodel to BOD time series[J]. Environmentrics,2000,11(3):341-350.
[142] Kantz H, Schreiber T. Nonlinear Time Series Analysis[M]. Beijing:Tsinghua University Press,1997:202-219.
[143]杨善林,李永森,胡笑旋,等. K-means算法中的k值优化问题研究.系统工程理论与实践,2006,(2):97-101.
[144]王军.基于局部模型的时间序列预测方法[D].哈尔滨,哈尔滨工业大学,2007:49-57.
[145] Richard O.等著.模式分类[M].李宏东,姚天翔等译.机械工业出版社,2003:432~454.
[146] Sandra Paterlini, Tommaso Minerva. Evolutionary Approaches for ClusterAnalysis[M]. Springer: Soft Computing Applications,2003:167-178.
[147] James C, Bezdek. Some New Indexes of Clustering Validity[J]. IEEE Transon Systems, Man and Cybemetics,1998,28(3):301-315.
[148] Smola, R. Williamson, B. Scholkopf. Generalization Bounds for ConvexCombinations of Kernel Functions. NeuroCOLT2Technical Report series,NC2-TR-1998-020, Royal Holloway College, University of London, UK,1998
[149]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):32-42.
[150] Jason D.Bakos and Krishna K. Nagar. Exploiting Matrix Symmetry toImprove FPGA Accelerated Conjugate Gradient[C]. In Proceeding ofFCCM2009,2009:223-226.
[151]周杰,陈啸洋,赵建勋,窦勇.大规模QR分解的FPGA设计与实现[J].计算机工程与科学,2010,32(10):34-37.
[152] W. Zhang, V. Betz, and J. Rose. Portable and Scalable FPGA-BasedAcceleration of a Direct Linear System Solver[C]. In Proceeding ofFPT2008,2008:17-24.
[153] Guiming Wu, Yong Dou, and Gregory D. Peterson. Blocking LUDecomposition for FPGAs[C]. In Proceeding of FCCM2010,2010:211-214.
[154] Time Series Prediction group. http://www.cis.hut.fi/projects/tsp/?page=Timeseries
[155]彭宇,王建民,彭喜元.基于回声状态网络的时间序列预测方法研究[J].电子学报,2010,38(2):149-150.
[156] Xilinx. UG702, Partial Reconfiguration User Guide(v12.3)[EB/OL].2010.http://www.xilinx.com/support/documentation/sw_manuals/xilinx12_3/ug702.pdf
[157]潘伟.基于SOPC的部分自重构系统的研究与设计[D].成都.电子科技大学.2009:19-47.
[158] Kaiyu Wang, Zhenan Tang, Yun Zhao, Hualong Li. A Dynamically Self-reconfigurable System Design Based on SSE Instruction Set[J]. ProcediaEngineering,2011,15:1605-1609.
[159] Ming Liu, Wolfgang Kuehn, Zhonghai Lu, Axel Jantsch. Run-time PartialReconfiguration Speed Investigation and Architectural Design SpaceExploration[C]. In Proceedings of the International Conference onFieldProgrammable Logic and Applications(FPL'09),2009:498-502.
[160] Omondi A R, Rajapakse J C. FPGA Implementations of NeuralNetworks[J]. IEEE Transactions on Neural Networks,2007,18(5):21-32.
[161] J. E. Stine, M. J. Schulte. The Symmetric Table Addition Method forAccurate Function Approximation[J]. Journal of VLSI Signal Processing,1999,21(2):167-177.
[162] Esam El-Araby, Ivan Gonzalez, Tarek El-Ghazaw. Exploiting PartialRuntime Reconfiguration for High-Performance ReconfigurableComputing[J]. ACM Transactions on Reconfigurable Technology andSystems,2009,1(4):1-25.
[163] Blythe D. Rise of the Graphics Processor[J].Proceedings of the IEEE,2008,96(5):761-778.
[164] Kostas Theoharoulis, Charalambos Antoniadis, NikolaosBellas, et al.Implementation and Performance Analysis of SEAL Encryptionon FPGA,GPU and Multi-Core Processors[C]. In Proceedings of the IEEEInternational Symposium on Field-Programmable Custom ComputingMachines,2011:65-68.
[165] NVIDIA. NVIDIA CUDA C Programming Guide Version4.2[EB/OL].2012.http://developer.download.nvidia.com/compute/DevZone/docs/html/C/doc/CUDA_C_Programming_Guide.pdf
[166] Roberto Cordone, Francesco Redaelli, Massimo Antonio Redaelli, et al.Partitioning and Scheduling of Task Graphs on Partially DynamicallyReconfigurable FPGAs[J]. IEEE Transactions on Computer-aided Designof Integrated Circuits and Systems,2009,28(5):662-675.
[167] ZonghuaGu, Mingxuan Yuan, Xiuqiang He. Optimal Static TaskScheduling on Reconfigurable Hardware Devices Using Model-Checking[C]. In Proceedings of the13th IEEE Real Time andEmbeddedTechnology and Applications Symposium,2007:32-44.
[168] Xie Guang-jun, Van Shi-qiang, Tang Zi-yue, et al. A PHM System forAEW Radar Based on AOPS-LSSVM Prognostic Algorithmand ExpertKnowledge Database[C]. In Proceedings of the2010Prognostics&SystemHealth Management Conference,2010:1-6.
[169] Francesco Redaelli, Marco D Santambrogio, Donatella Sciuto. TaskScheduling with Configuration Prefetching and Anti-FragmentationTechniques on Dynamically Reconfigurable Systems[C]. In Proceeding of2008Design, Automation and Test in Europe(DATE'08),2008:519-522.
[170] Shiann-RongKuang, Chin-Yang Chen, Ren-Zheng Liao. Partitioning andPipelined Scheduling of Embedded System Using Integer LinearProgramming[C]. In Proceedings of the11th International Conference onParallel and Distributed Systems,2005:37-41.
[171] Haluk Topcuoglu, Salim Hariri, Min-You Wu. Performance-effective andLow-complexity Task Scheduling for Heterogeneous Computing[J]. IEEETransactions on Parallel and Distributed Systems,2002,13(3):260-264.
[172] Minhaj Ahmad Khan. Scheduling for Heterogeneous Systems UsingConstrained Critical Paths[J]. Parallel Computing,2012,38(4-5):175-193.
[173] Mohammad I Daoud, NawwafKharma. Efficient Compile-Time Taskscheduling for Heterogeneous Distributed Computing Systems[C]. InProceedings of the12th International Conference on Parallel andDistributed Systems(ICPADS'06),2006:1-9.
[174] Bansal S, Kumar P, Singh K. An Improved Duplication Strategy forScheduling Precedence Constrained Graphs in Multiprocessor Systems[J].IEEE Transactions on Parallel and Distributed Systems,2003,14(6):533-544.
[175]贾志欣,殷国富,罗阳.二维不规则零件排样问题的遗传算法求解[J].计算机辅助设计与图形学学报,2002,14(5):467-470.
[176]邓冬梅,周来水.矩形件排样的研究进展[J].宇航材料工艺,2006,36(5):16-20.
[177] Andrea Lodi, Silvano Martello, Michele Monaci. Two DimensionalPacking Problems: A Survey[J]. European Journal of Operational Research,2002,141(2):241-252.
[178]马丹.任务间相互依赖的并行作业调度算法研究[D].武汉:华中科技大学,2007:6-8.
[179] Xilinx. Virtex-5Family Overview[EB/OL]. http://www.xilinx.com/support/documentation/data_sheets/ds100.pdf
[180] Xilinx. Virtex-5FPGA configuration user guide[EB/OL].http://www.xilinx.com/support/documentation/user_guides/ug191.pdf
[181] Dick R P, Rhodes D L, Wolf W. TGFF: Task Graphs for Free[C].InProceedings of the6th International Workshop onHardware/SoftwareCodesign(CODES/CASHE'98).1998:97-101.
[182] Scott Hauck. Configuration Prefetch for Single Context ReconfigurableCoprocessors[C]. In Proceedings of the1998ACM/SIGDA sixthinternational symposium on Field programmable gate arrays,1998:1-10.
[183] Guiming Wu, Yong Dou, Junqing Sun, et al. A High Performance andMemory Efficient LU Decomposer on FPGAs[J]. IEEE TransactionsonComputers,2012,61(3):366-378.
[184]Xiaoyong Tang, Kenli Li, Guiping Liao, et al. List Scheduling withDuplication for Heterogeneous Computing Systems[J]. Journal of Paralleland Distributed Computing,2010,70(4):323-329.
[2]韩敏.混沌时间序列预测理论与方法[M].北京:中国水利水电出版社，2007:155-185.
[3] Zhu Yuelong, Qian Jihong, Fan Qingsong, et al. Study on Hydrology TimeSeries Prediction Based on Wavelet-neural Networks[C].In Proceedings ofthe8th IEEE/ACIS International Conference on Computer and InformationScience,2009.411-415.
[4] Guo Z. X., Wong W. K., Li M.. Sparsely connected neural network-basedtime series forecasting[J]. Information Science,2012,19:54-71.
[5] NASA. Space Technology Roadmaps for the Fourteen TechnologyAreas/Technology Area Strategic Roadmaps and BreakdownStructure[EB/OL].2011. http://www.nasa.gov/offices/oct/home/roadmaps/index.html.
[6]张宝珍.预测与健康管理技术的发展及应用[J].测控技术,2008,23(2):5-7.
[7]彭宇,刘大同,彭喜元.故障预测与健康管理技术综述[J].电子测量与仪器学报,2010,4(1):1-8.
[8] Michael G. Pecht. Prognostics and Health Management of Electronics[M].John Wiley&Sons. Inc., Hoboken, New Jersey,2008:3-20.
[9] Mohanty S, Chattopadhyay A, Wei J, et al. Unsupervised Time-SeriesDamage State Estimation of Complex Structure Using UltrasoundBroadband Based Active Sensing[J]. Structural Durability&HealthMonitoring Journal,2010,130(1):101-124.
[10]许丽佳,王厚军,龙兵.一种状态检测与健康评估方法及其在模拟电路中的应用[J].计算机辅助设计与图形学学报,2008,20(12):1150-1556.
[11] Gang Niu, Bo-Suk Yang. Dempster-Shafer Regression for Multi-Step-ahead Time Series Prediction towards Data-driven Machinery Prognosis[J].Mechanical Systems and Signal Processing,2009,23(1):740-751.
[12] T. Makimoto. The rising wave of Field Programmability[C]. In Proceedingsof the FPL2000,2000:1-6.
[13] Monteneqro S, Roser H P, Huber F. BOSS: software and FPGAmiddleware for the‖flying laptop‖microsatellite[C]. In Proceedings ofDASIA,2005:401-405.
[14]罗赛.可重构计算系统体系结构与实现[D].合肥.中国科学技术大学,2006:16-19.
[15] Srihari Cadambi, Igor Durdanovic, Venkata Jakkula, et al.A MassivelyParallel FPGA-based Coprocessorfor Support Vector Machines[C]. InProceedings of the17th IEEE Symposium on Field Programmable CustomComputing Machines(FCCM'09),2009:115-122.
[16] Science Staff. Challenges and Opportunities[J]. Science Magazine,2011,331(6018):692-693.
[17] Serdar Uckun, Kai Goebel, Peter J F Lucas. Standardizing ResearchMethods for Prognostics[C]. In Proceedings of the International Conferenceon Prognostic and Health Management,2008:1-10.
[18] George E P Box, Gwilym M Jenkins, Gregory C Reinsel. Time SeiresAnalysis: Forecasting and Control[M]. Third Edition. New Jersey: JohnWiley&Sons Inc,1994:2-45.
[19] Jianqing Fan, Qiwei Yao. Nonlinear Time Series: Nonparametric andparametric methods[M]. New York: Springer-Verlag,2003:12-123.
[20] Engle R F. Autoregressive conditional heteroscedasticity with estimates ofthe variance of U.K. inflation[J]. Econometrica1,1982,50:987-1008
[21] Bollerslev T. Generalised autoregressive conditional heteroscedasticity[J].Journal of Econometrics,1986,31:307-327
[22] Tong H, Lim K S. Threshold autoregression, limit cycles and cyclicaldata[J]. Journal of Royal Statistical Society,1980, B42:245-292
[23] William W S Wei.Time Series Analysis: Univariate and MultivariateMethods[M]. Second Edition.London: Pearson EducationInc,2006:68-85,366-380.
[24] Zhang G, Patuwo B, Hu M. Forecasting with Artificial Neural Networks:The State of the Art[J]. International Journal of Forecasting,1998,14(1):35-62.
[25] Werbos P. Back propagation Through Time: What It Does and How to DoIt[J]. Proc of IEEE,1990,78(10):1550-1560.
[26]张军峰,胡寿松.基于一种新型聚类算法的RBF神经网络混沌时间序列预测[J].电子学报,2007,56(2):713-719.
[27] Giles C L, Lawre S, Tsoi A C. Nosiy Time Series Prediction UsingRecurrent Neural Networks and Grammatical Inference[J]. MachineLearning,2001,44(1):335-356.
[28] Herbert Jaeger.A Tutorial on Training Recurrent NeuralNetworks, CoveringBPPT, RTRL, EKF and the "Echo State Network" Approach[EB/OL].2003.http://www.faculty.jacobs-university.de/hjaeger/pubs.html.
[29] H Jaeger, H Haas. Harnessing Nonlinearity: Predicting Chaotic Systemsand Saving Energy in Wireless Communication[J]. Science,2004,10(2):78-80.
[30] Lean Yu, Shouyang Wang, Kin Keung Lai. A Neural-network-basedNonlinear Metamodeling Approach to Financial Time Series Forecasting[J].Applied Soft Computing,2009,9(2):563-574.
[31] Shu-liang Liu, Zhi-qiang Hu,Xiu-kai Chi. Power Load Forecasting Basedon Neural Network and Time Series[C]. In Proceedings of theInternationalConference on Wireless Communications, Networking and MobileComputing,2009:1-5.
[32] Girish K Jha, Parimala Thulasiraman, Ruppa K Thulasiram. PSO BasedNeural Network for Time Series Forecasting[C]. In Proceedings of theInternational Joint Conference on Neural Networks,2009:1422-1427.
[33] Xindi Cai, Nian Zhang, Ganesh K. Venayagamoorthy, Donald C. Wunsch II.Time series prediction with recurrent neural networks trained by a hybridPSO-EA algorithm[J]. Neuro Computing,2007,70(13-15):2342-2353.
[34] Dongqing Zhang, XuanxiNing, Xueni Liu, et al. Nonlinear Time SeriesForecasting with Dynamic RBF Neural Networks[C]. In Proceedings of the7thWorld Congress on Intelligent Control and Automation,2008:6988-6993.
[35] Hans Georg Seedig, Ralph Grothmann, Thomas A Runkler. Forecasting ofClustered Time Series withRecurrent Neural Networks and A FuzzyClustering Scheme[C]. In Proceedings of the International Joint Conferenceon Neural Networks,2009:2846-2853.
[36] Lili Qu, Yan Chen, Zhenfeng Liu. Time Series Forecasting Model withError Correction byStructure Adaptive RBF Neural Network[C].InProceedings of the6th World Congress on Intelligent ControlandAutomation,2006:6831-6835.
[37] Vapnik V. The Nature of Statistical Learning Theory[M]. Second Edition.New York: Springer-Verlag,1999:1-42.
[38] VN Vapnik.统计学习理论的本质[M].张学工,译.北京:清华大学出版社,2000:1-67.
[39]邓乃扬,田英杰.数据挖掘中的新方法:支持向量机[M].北京:科学出版社,2004:255-259.
[40] Johannes Mager, Ulrich Paasche, Bernhard Sick. Forecasting FinancialTime Series with Support Vector Machines Based on Dynamic Kernels[C].In Proceedings of the IEEE Conference on Soft Computing in IndustrialApplications,2008:252-257.
[41] Nicholas I Sapankevych, Ravi Sankar. Time Series Prediction UsingSupport Vector Machines: A Survey[J]. IEEE Computational IntelligenceMagazine,2009,4(2):24-38.
[42] Suykens J, Vandewalle J. Least Square Support Vector MachineClassifiers[J]. Neural Processing Letters,1999,9(3):293-300.
[43] Tony Van Gestel, Johan A KSuykens, Dirk-Emma Baestaens, et al.Financial Time Series Prediction Using Least SquaresSupport VectorMachines Within the EvidenceFramework[J]. IEEE Transactions on NeuralNetwork,2001,12(4):809-821.
[44] Yang Li. Short-Term Load Forecasting Based on LS-SVMOptimized byBCC Algorithm[C]. In Proceedings of the15th International ConferenceonIntelligent System Applications to Power Systems,2009:1-5.
[45] Zhang Fuwei, Chen Wei. Application of Grey LS-SVM in Mid and LongTermPower Load Forecasting[C]. In Proceedings of the InternationalConference onManagement and Service Science,2009:1-4.
[46] Yunhong Gao, Yibo Li. Fault Prediction Model Based on Phase SpaceReconstruction and Least SquaresSupport Vector Machine[C]. InProceedings of the9th International Conference on Hybrid IntelligentSystems,2009:464-467.
[47] Zhenrui Peng, Xijuan Yang, Wenzhe Qi. LS-SVM Based Soft Sensoring[C].In Proceedings of the7th World Congress on Intelligent Control andAutomation,2008:4320-4324.
[48]张英,苏宏业,褚健.基于模糊最小二乘支持向量机的软测量建模[J].控制与决策,2005,20(6):621-624.
[49] Wei Shen, Yunyun Zhang, Xiaoyong Ma. Stock Return Forecast with LS-SVM and Particle Swarm Optimization[C]. International Conference onBusiness Intelligence and Financial Engineering,2009:143-147.
[50] Yang Zhang, Yuncai Liu. Missing Traffic Flow Data Prediction UsingLeast Squares SupportVector Machines in Urban Arterial Streets[C]. InProceedings of the IEEE Symposium onComputational Intelligence andData Mining,2009:76-83.
[51] Wang Shaojun, Guo Jia, Liu Qi, Peng Xiyuan. On-Line Traffic Forecastingof Mobile Communication System[C]. International Conference onPervasive Computing Signal Processing and Applications,2010:97-100.
[52]邓聚龙.灰色系统理论教程[M].武汉:华中理工大学出版社,1990:1-38.
[53] Liehr S, Pawelzik K, Kohlmorgen J, et al. Hidden Markov Mixtures ofExperts for Prediction of Non-stationary Dynamics[C]. In Proceedings ofthe1999IEEE Signal Processing Society Workshop,1999:195-204.
[54] SimoS rkk, Aki Vehtari, JoukoLampinen. CATS Benchmark Time SeriesPrediction by Kalman Smoother with Cross-validated Noise Density[J].Neuro Computing,2007,70(13–15):2331-2341.
[55] Xuedong Wu, Zhihuan Song. Online Chaotic time-series prediction with thederivative-free extended Kalman filter[C]. In Proceedings of the7th WorldCongress onIntelligent Control and Automation,2008:2360-2364.
[56] Gordon N J, Salmond D J, Smith A F M. Novel Approach toNonlinear/Non-Gaussian Bayesian State Estimation[J]. Radar and SignalProcessing,1993,140(2):107-113.
[57] Zhang Wei, Zhang Feng. Overview of Data-driven Time Series ForecastingMethods[J]. Journal of Shanxi University of Science and Technology,2010,28(3):22-27.
[58] J Wichard. Forecasting the NN5Time Series with Hybrid Models[J].International Journal of Forecasting,2011,27(3):700-707.
[59] J Wichard, M Ogorzalek. Time Series Prediction with EnsembleModels[C].In Proceedings of the IEEE International Joint Conference onNeural Networks,2004:1625-1630.
[60]窦勇,谢伦国.可重构计算相关问题探讨[J].中国计算机学会通讯,2009,5(11):34-42.
[61] Dehon A, Wawraynek J. Reconfigurable Computing: What, Why, andImplication for Design Automation[C]. In Proceedings of the36th DesignAutomation Conference,1999:610-615.
[62] Compton K, Hauck S. Reconfigurable Computing: A Survey of Systemsand Software[J]. ACM Computing Surveys,2002,34(2):171-210.
[63] Estrin G, BusselB, Turn R, et al. Parallel Processing in A RestructurableComputer System[J]. IEEE Transactions on Electronic Computers,1963,12(5):747-755.
[64] Abbott A Lynn, Athanas Peter M, ChenLuna, et al.Finding Lines andBuilding Pyramids with Splash2[C]. In Proceedings of the IEEE Workshopon FPGAs for Custom Computing Machines,1994.155-163.
[65] Nikolaos Alachiotis, Simon A Berger, Alexandros Stamatakis. AcceleratingPhylogeny-Aware Short DNA Read Alignment with FPGAs[C]. InProceedings of the IEEE International Symposium on Field-ProgrammableCustom Computing Machines(FCCM),2011:226-233.
[66] Kirsch S, Rettig F, Hutter D, et al. An FPGA-Based High-Speed, Low-Latency Processing System for High-Energy Physics[C]. In Proceedings ofthe International Conference on Field Programmable Logic andApplications(FPL),2010:562-567.
[67] Haohuan Fu, Robert G Clapp. Eliminating the Memory Bottleneck: AnFPGA-Based Solution for3D Reverse Time Migration[C]. In Proceedingsof the19th ACM/SIGDA International Symposium on Field ProgrammableGate Arrays(FPGA'11),2011:65-74.
[68] Robin Pottathuparambil, Jack Coyne, Jeffrey Allred,et al. Low-LatencyFPGA Based Financial Data Feed Handler[C]. In Proceedings of the IEEE19th Annual International Symposium on Field-Programmable CustomComputing Machines(FCCM),2011:93-96.
[69] Anil Madhavapeddy, Satnam Singh. Reconfigurable Data Processing forClouds[C]. In Proceedings of the IEEE19th Annual InternationalSymposium on Field-Programmable Custom Computing Machines(FCCM),2011:141-145.
[70] Orlowska-Kowalska T, Kaminski M. FPGA Implementation of theMultilayer Neural Network for the Speed Estimation of the Two-MassDrive System[J]. IEEE Transactions on Industrial Informatics,2011,7(3):436-445.
[71]任小西.基于可重构计算的高可靠星载计算机体系结构研究[D].长沙,湖南大学,2007:7-22.
[72] Scott Hauck, André DeHon. Reconfigurable Computing: the Theory andPractice of FPGA-Based Computation[M]. San Fransisco: MorganKaufmann,2008:475-496.
[73] Callahan T J, Hauser J R, Wawrzynek J. The Garp architecture and Ccompiler[J]. Computer,2000,33(4):62-69.
[74] C Ebeling, D C Cronquist, P Franklin. RaPiD: Reconfigurable PipelinedDatapath[C]. In proceedings of the6th International Workshop on Field-Programmable Logic and Applications,1996:126-135.
[75] Goldstein S, Schmit H, Moe M, et al. PipeRench: A Coprocessor forStreaming Multimedia Acceleration[C]. In Proceedings ofthe26thInternational Symposium on Computer Architecture,1999:28-39.
[76] John D Davis, Charles P Thacker, Chen Chang. BEE3: RevitalizingComputer Architecture Research[EB/OL]. http://research.microsoft.com/projects/BEE3/.
[77] Liu C L, Layland J. Scheduling Algorithm for Multiprogramming in aHard-real-time Environment[J]. Journal of the ACM,1973,20(l):46-61.
[78] Kyprianos Papadimitriou, ApostolosDollas, Scott Hauck. Performance ofPartial Reconfiguration in FPGA Systems: A survey and a Cost Model[J].ACM Transactions on Reconfigurable Technology and Systems,2011,4(4):1-28.
[79]李涛.动态重构系统若干关键问题的研究[D].天津.南开大学,2007:9-11.
[80] Ullmann M, Hubner M, Grimm B, et al. An FPGA Run-time System forDynamical On-demand Reconfiguration[C]. In Proceedings of the18thInternational Parallel and Distributed Processing Symposium,2004:135-142.
[81]王峰.部分重构计算系统的关键技术研究[D].合肥.中国科学技术大学,2007:8-15.
[82] Amos R Omondi, J agath C Rajapakse, Mariusz Bajger. FPGAImplementations of Neural Networks[J]. IEEE Transactions on NeuralNetworks,2007,18(5):21-32.
[83] Amos R, Ormondi, Jagath C, et al. FPGA Implementationsof NeuralNetworks[M]. Berlin: Springer,2006:6-76.
[84] Benjamin Schrauwen, Michiel D’Haene, David Verstraeten, et al. CompactHardware Liquid State Machines on FPGA for Real-time SpeechRecognition[J]. Neural Networks,2008,21(2-3):511-523.
[85] Preethy A P, Radhakrishnan D, Omondi A R. A High PerformanceResidue-number-system Multiply-accumulate Unit[C]. In Proceedings ofthe11th ACM Great Lakes Symposium on VLSI,2001:145-149.
[86] Davide Anguita, Andrea Boni, Sandro Ridella. A Digital Architecture forSupport Vector Machines: Theory, Althorithm, and FPGAImplementation[J]. IEEE Transaction on Neural Networks,2003,14(5):993-1009.
[87] Choi W Y, Ahn D, Pan S B, et al. SVM-Based Speaker Verification Systemfor Match-on-Card and Its Hardware Implementation[J]. ETRI Journal,2006,28(3):320-328.
[88]曹葵康.支持向量机加速方法及应用研究[D].杭州.浙江大学,2010:49-50.
[89] Markos Papadonikolakis, Christos-Savvas Bouganis. A HeterogeneousFPGA Architecture for Support Vector Machine Training[C]. InProceedings of the18th IEEE Annual International Symposium on Field-Programmable Custom Computing Machines(FCCM),2010:211-214.
[90] Faisal MKhan, Mark G Arnold, William M Pottenger. Hardware-BasedSupport Vector Machine Classification in Logarithmic Number Systems[C].In Proceedings of the IEEE International Symposium on Circuits andSystems(ISCAS),2005:5154-5157.
[91] Omar Pina-Ramirez, Raquel Valdes-Cristerna, Oscar Yanez-Suarez. AnFPGA Implementation of Linear Kernel Support Vector Machines[C]. InProceedings of the IEEE International Conference on ReconfigurableComputing and FPGA's(ReConFig'06),2006:1-6.
[92]刘涵.王博.尹嵩.刘丁.一种基于FPGA的新的SVM硬件实现方法[J].仪器仪表学报,2010,31(10):2235-2241.
[93] Depeng Yang, Gregory D Peterson, Husheng Li, et al. An FPGAImplementation for Solving Least Square Problem[C]. In Proceedings ofthe17th IEEE Symposium on Field-Programmable Custom ComputingMachines,2009:303-306.
[94] Guiming Wu, Yong Dou, Gregory D Peterson. Blocking LU Decompositionfor FPGAs[C]. In Proceedings of the18th IEEE Annual InternationalSymposium on Field-Programmable Custom Computing Machines,2010:211-214.
[95] Beretta Ivan, Rana Vincenzo, Atienza David, et al. A Mapping Flow forDynamically Reconfigurable Multi-core System-on-chip Design[J]. IEEETransactions on Computer-Aided Design of Integrated Circuits andSystems,2011,30(8):1211-1224.
[96]殷进勇.可重构系统中实时任务调度算法研究[D].哈尔滨：哈尔滨工程大学,2010:5-10.
[97]刘勇.嵌入式可重构计算系统及其任务调度机制的研究[D].上海:中国科学院上海微系统与信息技术研究所,2006:45-51.
[98] Waheed Ahmed, Muhammad Shafique, Lars Bauer, et al. Run-TimeResource Allocation for Simultaneous Multi-tasking in Multi-coreReconfigurable Processors[C]. In Proceedings of the IEEE19th AnnualInternational Symposium on Field-Programmable Custom ComputingMachines,2011:29-32.
[99] Yang Qu, Juha-Pekka Soininen, Jari Nurmi. Static Scheduling Techniquesfor Dependent Tasks on Dynamically Reconfigurable Devices[J]. Journal ofSystems Architecture,2007,53(11):861-876.
[100] Christoph Steiger, Herbert Walder, Marco Platzner. Operating Systems forReconfigurable Embedded Platforms: Online Scheduling of Real-TimeTasks[J]. IEEE Transactions on Computers,2004,53(11):1393-1407.
[101] Banerjee S, Bozorgzadeh E, Dutt N. Parlgran: Parallelism GranularitySelection for Scheduling Task Chains on Dynamically ReconfigurableArchitectures[C]. In Proceedings of the Asia and South Pacific DesignAutomation Conference (ASP-DAC),2006:491-496.
[102] Compton K, Li Z, Cooley J, et al. Configuration Relocation andDefragmentation for Run-time Reconfigurable Computing[J]. IEEETransactions on VLSI Systems,2002,10(3):209-220.
[103] Li Z, Hauck S. Configuration Prefetching Techniquesfor PartialReconfigurable Coprocessor with Relocation and Defragmentation[C]. InProceedings of the ACM/SIGDA Tenth International Symposium on Field-programmable Gate Arrays,2002:187-195.
[104] Redaelli F, Santambrogio MD, Ogrenci Memik S. An ILP Formulation forthe Task Graph Scheduling Problem Tailored to Bi-dimensionalReconfigurable Architectures[C]. In Proceedings of the InternationalConference on Reconfigurable Computing and FPGAs(ReConFig'08),2008:97-102.
[105] IkbelBelaid, Fabrice Muller, Maher Benjemaa. Optimal Static Schedulingof Real-Time Dependent Tasks on Reconfigurable Hardware Devices[C]. InProceedings of IEEE International Conference on Communications,Computing and Control Applications (CCCA),2011:1-6.
[106] Zonghua Gu, Mingxuan Yuan, Xiuqiang He. Optimal Static TaskScheduling on Reconfigurable Hardware Devices Using Model-Checking[C]. In Proceedings of the13th IEEE Real Time and EmbeddedTechnology and Applications Symposium(RTAS'07),2007:32-44.
[107] Panainte EM, Bertels K., Vassiliadis S. Compiler-driven FPGA-areaAllocation for Reconfigurable Computing[C]. In Proceeding of the Design,Automation, and Test in Europe(DATE),2006:369-374.
[108] Mingxuan Yuan, ZonghuaGu, Xiuqiang He, et al. Hardware/SoftwarePartitioning and Pipelined Scheduling on Runtime ReconfigurableFPGAs[J]. ACM Transactions on Design Automation of Electronic Systems(TODAES),2010,15(2):13-54.
[109] Wenyin Fu, Katherine Compton. An Execution Environment forReconfigurable Computing[C]. In Proceedings of the13th Annual IEEESymposium on Field-Programmable Custom Computing Machines,2005:149-158.
[110] Banerjee S, Bozorgzadeh E, Dutt N. Physically-aware HW-SW Partitioningfor Reconfigurable Architectures with Partial Dynamic Reconfiguration[C].In Proceedings of the Design Automation Conference,2005:335-340.
[111] Banerjee S, Bozorgzadeh E, Dutt N. Integrating Physical Constraints inHW-SW Partitioning for Architectures with Partial DynamicReconfiguration[J]. IEEE Transactions on Very Large Scale Integration(VLSI) Systems,2006,14(11):1189-1202.
[112]刘彦.异构多核片上系统的任务调度及应用研究[D].长沙:湖南大学,2009:54-63.
[113] Ghiasi S, Sarrafzadeh M. Optimal Reconfiguration SequenceManagement[C]. In Proceedings of the Asia and South Pacific DesignAutomation Conference (ASP-DAC),2003:359-365.
[114] Juan Antonio Clemente, Carlos González, Javier Resano, et al. A HardwareTask-Graph Scheduler for Reconfigurable Multi-tasking Systems[C]. InProceedings of the International Conference on Reconfigurable Computingand FPGAs(ReConFig'08),2008:79-84.
[115] Justin Teller, Füsun zgüner, Robert Ewing. Scheduling Tasks onReconfigurable Hardware with a List Scheduler[C]. In Proceedings of theInternational Symposium on Parallel&Distributed Processing,2009:1-4.
[116] Suykens J, Vandewalle J. Least Square Support Vector MachineClassifiers[J]. Neural Processing Letters,1999,9(3):293-300.
[117] Suykens J A K, L Lukas, P Van Dooren, et al. Least Squares SupportVector Machine Classifiers: a Large Scale Algorithm[C]. In Proceedings ofthe European Conference on Circuit Theory and Design,1999:839-842.
[118] CHU W, ONG C J, KEERTHI S S. An Improved Conjugate GradientScheme to the Solution of Least Squares SVM[J]. IEEE Transaction onNeural Networks,2005,16(2):498-501.
[119] Edgar Osuna, Robert Freund, Federico Girosi. An Improved TrainingAlgorithm for Support VectorMachines[C]. In Proceedings of the IEEEWorkshop on Neural Networks for Signal Processing,1997:276-285.
[120] Platt J. Fast Training of Support Vector Machines Using SequentialMinimal Optimization[M]. Massachusetts: MIT Press,1999:185-208.
[121] Keerthi S S, Shevade S K. SMO Algorithm for Least-Squares SVMFormulations[J]. Neural Computation,2003,15(2):487-507.
[122] W. Chu, C. J. Ong, S. S. Keerthy. An improved conjugate gradient methodscheme to the solution of least squares SVM[J]. IEEE Transaction onNeural Networks,2005,16(2):498–501.
[123]张浩然,汪晓东.回归最小二乘支持向量机的增量和在线式学习算法[J].计算机学报,2006,29(3):400-406.
[124]叶美盈,汪晓东,张浩然.基于在线最小二乘支持向量机回归的混沌时间序列预测[J].物理学报,2005,54(6):2568-2572.
[125]郭振凯,宋召青,毛剑琴.一种改进的在线最小二乘支持向量机回归算法[J].控制与决策,2009,24(1):145-148.
[126]吴春国.广义染色体算法与迭代式最小二乘支持向量机回归算法研究
[D].吉林大学,2006:49-70.
[127] Marcelo Espinoza, Johan A. K. Suykens, Bart De Moor. Load Forecastingusing Fixed-Size Least Squares Support Vector Machines[J].Computational Intelligence and Bioinspired Systems,2005,35(12):1018-1026.
[128] Suykens J A K, Lukas L, Vandewalle J. Sparse Approximation Using LeastSquaresSupportVectorMachine[C]. In Proceedings of the IEEEInternationalSymposium onCircuits and Systems,2000:757-760.
[129] Gavin C Cawley, Nicola L C Talbot. Improved sparse least-squares supportvector machines[J]. Neurocomputing,2002,48:1025-1031
[130] Kruif B J, Vries J A. Pruning error minimization in least squares supportvector machines[J]. IEEE Trans on Neural Networks,2003,14(3):696-702.
[131] Hoegaerts L, Suykens J A K, Vandewalle J, et al. A Comparison of PruningAlgorithms for Sparse Least Squares Support Vector Machines[C]. InProceedings of the International Conference on Neural InformationProcessing,2004:1247-1253.
[132]赵永平,孙健国.关于稀疏最小二乘支持向量机的改进剪枝算法[J].系统工程理论与实践,2009,29(6):166-171.
[133]赵永平,孙健国.基于薄壁管法的稀疏最小二乘支持向量回归机[J].电机与控制学报,2009,13(4):581-585.
[134]周欣然,滕召胜,易钊.构造稀疏最小二乘支持向量机的快速剪枝算法[J].电机与控制学报,2009,13(4):626-630.
[135] Valyon J, Horvath G. A Sparse Least Squares Support Vector MachineClassifier[C]. In Proceedings of the IEEE International Conference onNeural Networks,2004:543-548.
[136]甘良志,孙宗海,孙优贤.稀疏最小二乘支持向量机[J].浙江大学学报(工学版),2007,41(2):245-248.
[137] Licheng Jiao, Liefeng Bo, Ling Wang. Fast Sparse Approximation for LeastSquares Support Vector Machine[J]. IEEE Transactions onNeural Networks,2007,18(3):685-697.
[138]赵永平,孙健国.一种快速稀疏最小二乘支持向量回归机[J].控制与决策,2008,12:1347-1352.
[139]陈爱军,宋执环,李平.基于矢量基学习的最小二乘支持向量机建模[J].控制理论与应用,2007,24(1):1-5.
[140]赵永平,孙健国.基于线性无关度的稀疏最小二乘支持向量回归机[J].中国图象图形学报,2009,14(6):1136-1140.
[141] Zongwu Cai, Ram C. Tiwari. Application of a local linear autoregressivemodel to BOD time series[J]. Environmentrics,2000,11(3):341-350.
[142] Kantz H, Schreiber T. Nonlinear Time Series Analysis[M]. Beijing:Tsinghua University Press,1997:202-219.
[143]杨善林,李永森,胡笑旋,等. K-means算法中的k值优化问题研究.系统工程理论与实践,2006,(2):97-101.
[144]王军.基于局部模型的时间序列预测方法[D].哈尔滨,哈尔滨工业大学,2007:49-57.
[145] Richard O.等著.模式分类[M].李宏东,姚天翔等译.机械工业出版社,2003:432~454.
[146] Sandra Paterlini, Tommaso Minerva. Evolutionary Approaches for ClusterAnalysis[M]. Springer: Soft Computing Applications,2003:167-178.
[147] James C, Bezdek. Some New Indexes of Clustering Validity[J]. IEEE Transon Systems, Man and Cybemetics,1998,28(3):301-315.
[148] Smola, R. Williamson, B. Scholkopf. Generalization Bounds for ConvexCombinations of Kernel Functions. NeuroCOLT2Technical Report series,NC2-TR-1998-020, Royal Holloway College, University of London, UK,1998
[149]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):32-42.
[150] Jason D.Bakos and Krishna K. Nagar. Exploiting Matrix Symmetry toImprove FPGA Accelerated Conjugate Gradient[C]. In Proceeding ofFCCM2009,2009:223-226.
[151]周杰,陈啸洋,赵建勋,窦勇.大规模QR分解的FPGA设计与实现[J].计算机工程与科学,2010,32(10):34-37.
[152] W. Zhang, V. Betz, and J. Rose. Portable and Scalable FPGA-BasedAcceleration of a Direct Linear System Solver[C]. In Proceeding ofFPT2008,2008:17-24.
[153] Guiming Wu, Yong Dou, and Gregory D. Peterson. Blocking LUDecomposition for FPGAs[C]. In Proceeding of FCCM2010,2010:211-214.
[154] Time Series Prediction group. http://www.cis.hut.fi/projects/tsp/?page=Timeseries
[155]彭宇,王建民,彭喜元.基于回声状态网络的时间序列预测方法研究[J].电子学报,2010,38(2):149-150.
[156] Xilinx. UG702, Partial Reconfiguration User Guide(v12.3)[EB/OL].2010.http://www.xilinx.com/support/documentation/sw_manuals/xilinx12_3/ug702.pdf
[157]潘伟.基于SOPC的部分自重构系统的研究与设计[D].成都.电子科技大学.2009:19-47.
[158] Kaiyu Wang, Zhenan Tang, Yun Zhao, Hualong Li. A Dynamically Self-reconfigurable System Design Based on SSE Instruction Set[J]. ProcediaEngineering,2011,15:1605-1609.
[159] Ming Liu, Wolfgang Kuehn, Zhonghai Lu, Axel Jantsch. Run-time PartialReconfiguration Speed Investigation and Architectural Design SpaceExploration[C]. In Proceedings of the International Conference onFieldProgrammable Logic and Applications(FPL'09),2009:498-502.
[160] Omondi A R, Rajapakse J C. FPGA Implementations of NeuralNetworks[J]. IEEE Transactions on Neural Networks,2007,18(5):21-32.
[161] J. E. Stine, M. J. Schulte. The Symmetric Table Addition Method forAccurate Function Approximation[J]. Journal of VLSI Signal Processing,1999,21(2):167-177.
[162] Esam El-Araby, Ivan Gonzalez, Tarek El-Ghazaw. Exploiting PartialRuntime Reconfiguration for High-Performance ReconfigurableComputing[J]. ACM Transactions on Reconfigurable Technology andSystems,2009,1(4):1-25.
[163] Blythe D. Rise of the Graphics Processor[J].Proceedings of the IEEE,2008,96(5):761-778.
[164] Kostas Theoharoulis, Charalambos Antoniadis, NikolaosBellas, et al.Implementation and Performance Analysis of SEAL Encryptionon FPGA,GPU and Multi-Core Processors[C]. In Proceedings of the IEEEInternational Symposium on Field-Programmable Custom ComputingMachines,2011:65-68.
[165] NVIDIA. NVIDIA CUDA C Programming Guide Version4.2[EB/OL].2012.http://developer.download.nvidia.com/compute/DevZone/docs/html/C/doc/CUDA_C_Programming_Guide.pdf
[166] Roberto Cordone, Francesco Redaelli, Massimo Antonio Redaelli, et al.Partitioning and Scheduling of Task Graphs on Partially DynamicallyReconfigurable FPGAs[J]. IEEE Transactions on Computer-aided Designof Integrated Circuits and Systems,2009,28(5):662-675.
[167] ZonghuaGu, Mingxuan Yuan, Xiuqiang He. Optimal Static TaskScheduling on Reconfigurable Hardware Devices Using Model-Checking[C]. In Proceedings of the13th IEEE Real Time andEmbeddedTechnology and Applications Symposium,2007:32-44.
[168] Xie Guang-jun, Van Shi-qiang, Tang Zi-yue, et al. A PHM System forAEW Radar Based on AOPS-LSSVM Prognostic Algorithmand ExpertKnowledge Database[C]. In Proceedings of the2010Prognostics&SystemHealth Management Conference,2010:1-6.
[169] Francesco Redaelli, Marco D Santambrogio, Donatella Sciuto. TaskScheduling with Configuration Prefetching and Anti-FragmentationTechniques on Dynamically Reconfigurable Systems[C]. In Proceeding of2008Design, Automation and Test in Europe(DATE'08),2008:519-522.
[170] Shiann-RongKuang, Chin-Yang Chen, Ren-Zheng Liao. Partitioning andPipelined Scheduling of Embedded System Using Integer LinearProgramming[C]. In Proceedings of the11th International Conference onParallel and Distributed Systems,2005:37-41.
[171] Haluk Topcuoglu, Salim Hariri, Min-You Wu. Performance-effective andLow-complexity Task Scheduling for Heterogeneous Computing[J]. IEEETransactions on Parallel and Distributed Systems,2002,13(3):260-264.
[172] Minhaj Ahmad Khan. Scheduling for Heterogeneous Systems UsingConstrained Critical Paths[J]. Parallel Computing,2012,38(4-5):175-193.
[173] Mohammad I Daoud, NawwafKharma. Efficient Compile-Time Taskscheduling for Heterogeneous Distributed Computing Systems[C]. InProceedings of the12th International Conference on Parallel andDistributed Systems(ICPADS'06),2006:1-9.
[174] Bansal S, Kumar P, Singh K. An Improved Duplication Strategy forScheduling Precedence Constrained Graphs in Multiprocessor Systems[J].IEEE Transactions on Parallel and Distributed Systems,2003,14(6):533-544.
[175]贾志欣,殷国富,罗阳.二维不规则零件排样问题的遗传算法求解[J].计算机辅助设计与图形学学报,2002,14(5):467-470.
[176]邓冬梅,周来水.矩形件排样的研究进展[J].宇航材料工艺,2006,36(5):16-20.
[177] Andrea Lodi, Silvano Martello, Michele Monaci. Two DimensionalPacking Problems: A Survey[J]. European Journal of Operational Research,2002,141(2):241-252.
[178]马丹.任务间相互依赖的并行作业调度算法研究[D].武汉:华中科技大学,2007:6-8.
[179] Xilinx. Virtex-5Family Overview[EB/OL]. http://www.xilinx.com/support/documentation/data_sheets/ds100.pdf
[180] Xilinx. Virtex-5FPGA configuration user guide[EB/OL].http://www.xilinx.com/support/documentation/user_guides/ug191.pdf
[181] Dick R P, Rhodes D L, Wolf W. TGFF: Task Graphs for Free[C].InProceedings of the6th International Workshop onHardware/SoftwareCodesign(CODES/CASHE'98).1998:97-101.
[182] Scott Hauck. Configuration Prefetch for Single Context ReconfigurableCoprocessors[C]. In Proceedings of the1998ACM/SIGDA sixthinternational symposium on Field programmable gate arrays,1998:1-10.
[183] Guiming Wu, Yong Dou, Junqing Sun, et al. A High Performance andMemory Efficient LU Decomposer on FPGAs[J]. IEEE TransactionsonComputers,2012,61(3):366-378.
[184]Xiaoyong Tang, Kenli Li, Guiping Liao, et al. List Scheduling withDuplication for Heterogeneous Computing Systems[J]. Journal of Paralleland Distributed Computing,2010,70(4):323-329.