基于D~*思想的ASON动态均衡恢复策略研究
|
摘要
随着Internet的高速发展,全球数据业务呈爆炸式增长。数据业务动态、突发等特性对传统的光传送网(OTN,Optical Transmission Network)提出了更高的要求。自动交换光网络(ASON,Automatically Switched Optical Network)作为一项新的智能光网络技术的出现,赋予了光网络前所未有的灵活性和可扩展性,已经成为下一代网络(NGN,Next Generation Network)的重要发展方向。其中,对多种恢复机制的支持是ASON的一个重要特点。ASON上承载的多种业务要求网络具有快速、智能和多样化的故障恢复能力,同时能够合理有效地分配网络资源。而目前基于传统光传送网设计的一系列恢复机制远远不能满足ASON对生存性的要求。因此,对ASON恢复问题的研究已成为ASON研究的一个重点之一。
D~*算法(D-star algorithm)是一种动态环境下的启发式(Heuristic)搜索算法,应用于环境为未知、部分已知或不断变化的状态空间搜索中。具有智能、高效、鲁棒性、易于实现分布式等优点。博弈论是研究具有斗争或竞争性质现象的理论和方法。论文在研究探讨了ASON及其恢复相关问题的基础上,运用D~*算法,并结合混合策略博弈理论,设计了一种动态均衡的ASON恢复策略,通过路由和波长分配(RWA,Routing and Wavelength Assignment)为受损业务提供恢复通道。目的是在快速恢复故障的同时,合理利用网络资源。从而提高网络的故障恢复率,增强网络的生存性。
论文围绕ASON的动态恢复问题展开论述,主要完成的工作和取得的成果如下:
(1)分析归纳了ASON控制平面的结构、功能,ASON路由体系结构、路由方式、消息分发,ASON的生存性及恢复策略以及ASON动态路由和波长分配(RWA,Route and Wavelength Assignment)技术等与ASON恢复问题相关的背景知识和影响因素。(2)研究了D~*算法和混合策略博弈的相关理论,提出了一种基于D~*思想的ASON动态均衡恢复策略。将网络中的路径代价与当前状态下的波长资源占用情况综合考虑,运用混合博弈原理求解波长占用与路径代价之间的纳什均衡(NashEquilibration),并以此动态建立D~*算法的估计函数,使恢复策略能够在动态通道恢复过程中尽量保证网络资源的合理分配。相对于传统的最短路径恢复策略和简单的动态恢复策略,该策略能够取得更高的恢复率和更低的网络资源占用率。
(3)根据恢复策略的思路,详细分析并设计实现了均衡D~*算法的原理,包括建立数据结构和数学模型,以及流程设计和性能分析等。针对仿真实验时发现由于求解纳什均衡非线性方程组而造成算法收敛较慢的不足,设计了一种结合Newton迭代法和最小二乘法求解纳什均衡非线性方程组的方法,仿真实验证明,可以大大加快算法的收敛速度。
(4)采用C#和MATLAB混合编程技术,自行开发了恢复策略的仿真系统,模拟实际网络环境对恢复算法性能进行了测试,并且,用传统的最短路径算法和另行设计的简单D~*算法与之进行了比较。仿真结果表明,该算法在阻塞率、恢复率以及资源占用率上都明显优于相比较的其他算法,且时间性能可以较好满足ASON中恢复时间的要求。
With the rapid development of Internet, the global data traffic growth is exponential. The dynamic, burst characteristics of data traffic present the high requirement to traditional optical transmission network (OTN). As a new technology of intelligent optical network, the appearance of automatically switched optical network (ASON) endue traditional optical transport network with unprecedented flexibility and expansibility. ASON represented the future direction of next generation networks (NGN). Hereinto, supporting many kinds of restoration strategy is an important characteristic of ASON. Multi-traffics carried in ASON demand the fast and multi-schemes restoration ability of network, simultaneity, reasonably and efficiently assign the network resources. However, at present, a series of restoration strategy there are designed for traditional optical transmission network are hardly satisfied with the survivability request of ASON. Therefore, the research on ASON restoration problem became a focus in ASON studies.
D~* (D-star) algorithm is a heuristic search algorithm in dynamic environments. It is applied to paths planning in unknown, partially known, and changing state spaces, with advantages such as intelligence, high efficiency, robustness, and easy to actualize distributing, etc. Game theory is the method and architecture to research the phenomena with struggling or competitive characters. Based on researching the relative problem of ASON and restoration in ASON, the article design a dynamic equilibrated restoration strategy in ASON using D* algorithm and uniting game theory of mixed strategy to provide the restoration channel for failed traffic by routing and wavelength assignment. The purpose is to restore fault rapidly, simultaneously, to assign the network resources reasonably, and then to improve the rate of restoration and enhance survivability of network.
In this thesis, the research is focus on restoration in ASON. The main works and innovative results are listed as follow:
(1) Analyzing and summing up the background knowledge and influence factors related problems of restoration of ASON, which include framework and functions of control plane, architecture and modes of route, message distribution, the strategies of survivability and restoration, the techniques of dynamic route and wavelength assignment (RWA) etc. in ASON.
(2) Researching the related theories of D~* algorithm and game theory of mixed strategy, and present a D~* principle-based dynamic equilibrated restoration strategy in ASON. Considering of the path cost and the occupancy of wavelength comprehensively, using game theory of mixed strategy to solve Nash equilibration between occupancy of wavelength and path cost, according this, the evaluation function of D~* algorithm is constructed dynamically, for ensuring a more equitable distribution of network resources in the process of dynamic channel restoration. Compared with traditional restoration strategy based on shortest path and simple dynamic restoration strategy, this algorithm can achieve higher restorability and lower resource occupancy.
(3) According to the idea of the restoration strategy, the article detailedly analyze, design and carry out the principle of equilibrated D~* algorithm, which include constructing data structure and model of mathematic, flow design and capability analysis, etc. Against the disadvantage of slow convergence founded in simulation experiment, due to sloving nonlinear equations of Nash equilibrium, the article design a method which combining Newton iterative method with least square method to solve nonlinear equations of Nash equilibrium. It can be proved to accelerate convergent speed of the algorithm by simulation.
(4) Using C# and MATLAB integrated programming, the article design simulation system of restoration strategy independently, and test the performance of equilibrated D~* algorithm by simulating actual network environment. In addition, we compare equilibrated D~* algorithm with shortest path algorithm and simple D~* algorithm which designed separately. It shows that the blocked rate, restorable rate and rate of resource occupancy of equilibrated D~* algorithm, are all obviously better than other comparative algorithms, and the performance of time of equilibrated D~* algorithm can sufficiently fit the request of restoration time in ASON.
D~*算法(D-star algorithm)是一种动态环境下的启发式(Heuristic)搜索算法,应用于环境为未知、部分已知或不断变化的状态空间搜索中。具有智能、高效、鲁棒性、易于实现分布式等优点。博弈论是研究具有斗争或竞争性质现象的理论和方法。论文在研究探讨了ASON及其恢复相关问题的基础上,运用D~*算法,并结合混合策略博弈理论,设计了一种动态均衡的ASON恢复策略,通过路由和波长分配(RWA,Routing and Wavelength Assignment)为受损业务提供恢复通道。目的是在快速恢复故障的同时,合理利用网络资源。从而提高网络的故障恢复率,增强网络的生存性。
论文围绕ASON的动态恢复问题展开论述,主要完成的工作和取得的成果如下:
(1)分析归纳了ASON控制平面的结构、功能,ASON路由体系结构、路由方式、消息分发,ASON的生存性及恢复策略以及ASON动态路由和波长分配(RWA,Route and Wavelength Assignment)技术等与ASON恢复问题相关的背景知识和影响因素。(2)研究了D~*算法和混合策略博弈的相关理论,提出了一种基于D~*思想的ASON动态均衡恢复策略。将网络中的路径代价与当前状态下的波长资源占用情况综合考虑,运用混合博弈原理求解波长占用与路径代价之间的纳什均衡(NashEquilibration),并以此动态建立D~*算法的估计函数,使恢复策略能够在动态通道恢复过程中尽量保证网络资源的合理分配。相对于传统的最短路径恢复策略和简单的动态恢复策略,该策略能够取得更高的恢复率和更低的网络资源占用率。
(3)根据恢复策略的思路,详细分析并设计实现了均衡D~*算法的原理,包括建立数据结构和数学模型,以及流程设计和性能分析等。针对仿真实验时发现由于求解纳什均衡非线性方程组而造成算法收敛较慢的不足,设计了一种结合Newton迭代法和最小二乘法求解纳什均衡非线性方程组的方法,仿真实验证明,可以大大加快算法的收敛速度。
(4)采用C#和MATLAB混合编程技术,自行开发了恢复策略的仿真系统,模拟实际网络环境对恢复算法性能进行了测试,并且,用传统的最短路径算法和另行设计的简单D~*算法与之进行了比较。仿真结果表明,该算法在阻塞率、恢复率以及资源占用率上都明显优于相比较的其他算法,且时间性能可以较好满足ASON中恢复时间的要求。
With the rapid development of Internet, the global data traffic growth is exponential. The dynamic, burst characteristics of data traffic present the high requirement to traditional optical transmission network (OTN). As a new technology of intelligent optical network, the appearance of automatically switched optical network (ASON) endue traditional optical transport network with unprecedented flexibility and expansibility. ASON represented the future direction of next generation networks (NGN). Hereinto, supporting many kinds of restoration strategy is an important characteristic of ASON. Multi-traffics carried in ASON demand the fast and multi-schemes restoration ability of network, simultaneity, reasonably and efficiently assign the network resources. However, at present, a series of restoration strategy there are designed for traditional optical transmission network are hardly satisfied with the survivability request of ASON. Therefore, the research on ASON restoration problem became a focus in ASON studies.
D~* (D-star) algorithm is a heuristic search algorithm in dynamic environments. It is applied to paths planning in unknown, partially known, and changing state spaces, with advantages such as intelligence, high efficiency, robustness, and easy to actualize distributing, etc. Game theory is the method and architecture to research the phenomena with struggling or competitive characters. Based on researching the relative problem of ASON and restoration in ASON, the article design a dynamic equilibrated restoration strategy in ASON using D* algorithm and uniting game theory of mixed strategy to provide the restoration channel for failed traffic by routing and wavelength assignment. The purpose is to restore fault rapidly, simultaneously, to assign the network resources reasonably, and then to improve the rate of restoration and enhance survivability of network.
In this thesis, the research is focus on restoration in ASON. The main works and innovative results are listed as follow:
(1) Analyzing and summing up the background knowledge and influence factors related problems of restoration of ASON, which include framework and functions of control plane, architecture and modes of route, message distribution, the strategies of survivability and restoration, the techniques of dynamic route and wavelength assignment (RWA) etc. in ASON.
(2) Researching the related theories of D~* algorithm and game theory of mixed strategy, and present a D~* principle-based dynamic equilibrated restoration strategy in ASON. Considering of the path cost and the occupancy of wavelength comprehensively, using game theory of mixed strategy to solve Nash equilibration between occupancy of wavelength and path cost, according this, the evaluation function of D~* algorithm is constructed dynamically, for ensuring a more equitable distribution of network resources in the process of dynamic channel restoration. Compared with traditional restoration strategy based on shortest path and simple dynamic restoration strategy, this algorithm can achieve higher restorability and lower resource occupancy.
(3) According to the idea of the restoration strategy, the article detailedly analyze, design and carry out the principle of equilibrated D~* algorithm, which include constructing data structure and model of mathematic, flow design and capability analysis, etc. Against the disadvantage of slow convergence founded in simulation experiment, due to sloving nonlinear equations of Nash equilibrium, the article design a method which combining Newton iterative method with least square method to solve nonlinear equations of Nash equilibrium. It can be proved to accelerate convergent speed of the algorithm by simulation.
(4) Using C# and MATLAB integrated programming, the article design simulation system of restoration strategy independently, and test the performance of equilibrated D~* algorithm by simulating actual network environment. In addition, we compare equilibrated D~* algorithm with shortest path algorithm and simple D~* algorithm which designed separately. It shows that the blocked rate, restorable rate and rate of resource occupancy of equilibrated D~* algorithm, are all obviously better than other comparative algorithms, and the performance of time of equilibrated D~* algorithm can sufficiently fit the request of restoration time in ASON.
引文
[1]中国互联网络发展状况统计报告.中国互联网络信息中心,2007,1
[2]LION project:http://lion.deri.ie/
[3]FASHION project:http://www.eurescom.de/public/projects/p1000-series/P1012/default.asp
[4]3TNet高性能宽带信息网:http://www.3tnet.com.cn/
[5]Dongyun Zhou and Suresh Subramaniam,Survivability in Optical Networks[J],IEEE Network,2000,14(6):16-23
[6]ITU-T G.8080/Y1034.Architecture for the automatically switched optical network(ASON)amendment 1,2003
[7]ITU-T G.7715.Routing Architecture and Requirements for Automatically Switched Optical network,2002
[8]Mokhtar A,Azizoglu M.Adaptive Wavelength Routing in All-Optical Networks[J].IEEE/ACM Transactions on Networking,1998,6(2):197-206
[9]Kit-Man Chan,Yum T.P.,Analysis of least congested path routing in WDM lightwave networks[C].INFOCOM'94.Networking for Global communications,13th Proceedings IEEE,1994:962-969.
[10]Lee KC,Li VOK.A wavelength rerouting algorithm in wide-area all-optical networks[J].JOURNAL OF LIGHTWAVE TECHNOLOGY,June 1996,14(6):1218-1229
[11]Chien Chen,Subrata Banerjee.A new model for optimal routing and wavelength assignment in wavelength division multiplexed optical networks[C].San Francisco:INFOCOM '96.Fifteenth Annual Joint Conference of the IEEE Computer Societies.Networking the Next Generation.Proceedings IEEE 1996:164-171
[12]夏俊,喻敬海,吴志坚.波分复用全光网络路由和波长分配算法[J].通信学报,2000,21(8):85-89
[13]肖诗源,金鑫,刘贤德.共享波长转换器全光网的路由和波长分配算法[J].华中科技大学学报(自然科学版),2004,32(6):42-44
[14]Zhu Na,Sun Haljin,Zhou Naifu.Ant colony optimization for dynamic RWA in WDM networks With partial wavelength[J].PHOTONIC NETWORK COMMUNICATIONS,2006,11(2):229 - 236
[15]段亚伟,朱娜,李正茂.基于遗传算法的RWA问题的研究[J].计算机工程与应用,2005,41(23):132-134
[16]Stentz,A.Optimal and efficient path planning for partially-known environments[C].San Diego:In Proceedings of the IEEE International Conference on Robotics and Automation,1994:3310-3317
[17]Stentz A.The focused D* algorithm for real-time replanning[C].Montreal:In Proceedings of the International Joint Conference on Artificall Intelligence,1995:1231-1239
[18]龚倩.智能光交换网络[M].北京:北京邮电大学出版社,2003
[19]ITU-T G.8080/Y.1034.Architecture for the automatically switched optical network(ASON).2001,11
[20]吴彦文,郑大力,仲肇伟.光网络的生存性技术[M].北京:北京邮电大学出版社,2002
[21]Salah Aidarous,Thomas Plevyak.Communications Network Management into 21st Century:Techniques,Standards,Technologies,and Applications[M].New York,IEEE Press,1994:337-340.
[22]张杰,徐云斌,宋鸿升,桂烜,顾畹仪.自动交换光网络ASON.北京:人民邮电出版社,2004
[23]天洪亮.宽带网络生存性研究[D].浙江大学博士学位论文,2000
[24]D.Banerjee,B,Mukherjee,A Practical Approach for Routing and Wavelength Assignment in Large Wavelength-Routed Optical Networks[J].IEEE Journal on Selected Areas in communications,1996,14(5):903-908
[25]I.chlamtac,A.Ganz,G.Karmi,Lightpath Communications:An Approach to High Bandwidth optical WANs[J].IEEE Transactions on communications,1992,40(2):1171-1182
[26]R.Ramaswami,K.N.Sivarajan,Optimal Routing and wavelength Assignment in All-Optical Networks[C].Proceedings of IEEE INFOCOM'94,1994:110-119
[27]JingYi He,S.H.Gary Chan,Danny H.K.Tsang,Routing and Wavelength Assignment for WDM Multicast Networks[C].GLOBECOM'01.IEEE,2001:1536-1540
[28]Dijkstra E.A Note on Two Problems in Cormeetion with Graphs[C].Numeric Mathematic,1959,1:269-271
[29]Jong-Seon Kim,Lee D.C.Dynamic routing and wavelength assignment algorithms for multifiber WDM networks with many wavelengths[C].2nd European Conference on Universal Multiservice Networks,2002:180-186
[30]Robert W.Floyd.Algorithm 97:shortest path[J].Communications of the ACM,1962,5(6):345
[31]Yongbing Zhang,Taira K.,Takagi H,etc.An efficient heuristic for routing and wavelength assignment in optical WDM networks[C].IEEE International Conference on communications,ICC 2002.5,2002:2734-2739
[32]徐世中.WDM光传送网路由和波长分配算法研究[D].电子科技大学博士学位论文,2000
[33]顾畹仪.光传送网[M].北京:机械工业出版社,2003
[34]Hart P.,Nilsson N.,Raphael B.A Formal Basis for the Heuristic Determination of Minimum Cost Paths[C].IEEE Trans.Syst.Science and Cybernetics 1968,SSC-4(2):100-107
[35]Hart P.,Nilsson N.,Raphael B.Correction to "A Formal Basis for the Heuristic Determination of Minimum Cost Paths"[C].SIGART Newsletter,Dee.1972,No.37:29-29
[36]Nils J.Nilsson.人工智能[M].郑扣根,庄越挺译.北京:机械工业出版社,2000
[37]姚国庆.博弈论[M].天津:南开大学出版社,2003
[38]Robert Gibbons.博弈论基础[M].高峰译.北京:中国社会科学出版社,1999
[39]Nash J.Equilibrium Points in n-Person Games[C].Proceeding of the National Academy of Sciences 36:48-49
[40]Lee David,Libman Lavy,Orda Ariel.Path protection and blocking probability minimization in optical networks[C].Hong Kong:IEEE INFOCOM 2004,23rd Annual Joint Conference of the IEEE Computer and Communications Societies,2004:142-153
[41]Georgakopoulos George F.,Kavvadias Dimitris J.,Sioutis Leonidas G.Nash Equilibria in All-Optical Networks[C].Hong Kong:1st International Workshop on Internet and Network Economics,WINE 2005:1033-1045
[42]唐雄燕,左鹏.智能光网络-技术与应用实践[M].北京:电子工业出版社,2005
[43]RFC3209-RSVP-TE:Extensions to RSVP for LSP Tunnels.Network Working Group,2001
[44]严蔚敏,吴伟民.数据结构(C语言版)[M].北京:清华大学出版社,2002
[45]张可村,赵英良.数值计算的算法与分析[M].北京:科学出版社,2003
[46]马正飞,殷翔.数学计算方法与软件的工程应用[M].北京:化学工业出版社,2002
[47]张圣勤,MATLAB7.0实用教程[M].北京:机械工业出版社,2006
[48]吕文达.精通C#程序设计[M].北京:清华大学出版社,2004
[49]MATLAB Help Documentation.The MathWorks,Inc.2007
[50]赵士伟,赵明波,陈平.基于COM的MATLAB与C#.NET混合编程的实现与应用[J].山东理工大学学报(自然科学版),2006,20(4):26-29
[51]陈锡生.ASON动态光路建立的阻塞分析[J].中兴通讯技术,2004,6:17-19
[2]LION project:http://lion.deri.ie/
[3]FASHION project:http://www.eurescom.de/public/projects/p1000-series/P1012/default.asp
[4]3TNet高性能宽带信息网:http://www.3tnet.com.cn/
[5]Dongyun Zhou and Suresh Subramaniam,Survivability in Optical Networks[J],IEEE Network,2000,14(6):16-23
[6]ITU-T G.8080/Y1034.Architecture for the automatically switched optical network(ASON)amendment 1,2003
[7]ITU-T G.7715.Routing Architecture and Requirements for Automatically Switched Optical network,2002
[8]Mokhtar A,Azizoglu M.Adaptive Wavelength Routing in All-Optical Networks[J].IEEE/ACM Transactions on Networking,1998,6(2):197-206
[9]Kit-Man Chan,Yum T.P.,Analysis of least congested path routing in WDM lightwave networks[C].INFOCOM'94.Networking for Global communications,13th Proceedings IEEE,1994:962-969.
[10]Lee KC,Li VOK.A wavelength rerouting algorithm in wide-area all-optical networks[J].JOURNAL OF LIGHTWAVE TECHNOLOGY,June 1996,14(6):1218-1229
[11]Chien Chen,Subrata Banerjee.A new model for optimal routing and wavelength assignment in wavelength division multiplexed optical networks[C].San Francisco:INFOCOM '96.Fifteenth Annual Joint Conference of the IEEE Computer Societies.Networking the Next Generation.Proceedings IEEE 1996:164-171
[12]夏俊,喻敬海,吴志坚.波分复用全光网络路由和波长分配算法[J].通信学报,2000,21(8):85-89
[13]肖诗源,金鑫,刘贤德.共享波长转换器全光网的路由和波长分配算法[J].华中科技大学学报(自然科学版),2004,32(6):42-44
[14]Zhu Na,Sun Haljin,Zhou Naifu.Ant colony optimization for dynamic RWA in WDM networks With partial wavelength[J].PHOTONIC NETWORK COMMUNICATIONS,2006,11(2):229 - 236
[15]段亚伟,朱娜,李正茂.基于遗传算法的RWA问题的研究[J].计算机工程与应用,2005,41(23):132-134
[16]Stentz,A.Optimal and efficient path planning for partially-known environments[C].San Diego:In Proceedings of the IEEE International Conference on Robotics and Automation,1994:3310-3317
[17]Stentz A.The focused D* algorithm for real-time replanning[C].Montreal:In Proceedings of the International Joint Conference on Artificall Intelligence,1995:1231-1239
[18]龚倩.智能光交换网络[M].北京:北京邮电大学出版社,2003
[19]ITU-T G.8080/Y.1034.Architecture for the automatically switched optical network(ASON).2001,11
[20]吴彦文,郑大力,仲肇伟.光网络的生存性技术[M].北京:北京邮电大学出版社,2002
[21]Salah Aidarous,Thomas Plevyak.Communications Network Management into 21st Century:Techniques,Standards,Technologies,and Applications[M].New York,IEEE Press,1994:337-340.
[22]张杰,徐云斌,宋鸿升,桂烜,顾畹仪.自动交换光网络ASON.北京:人民邮电出版社,2004
[23]天洪亮.宽带网络生存性研究[D].浙江大学博士学位论文,2000
[24]D.Banerjee,B,Mukherjee,A Practical Approach for Routing and Wavelength Assignment in Large Wavelength-Routed Optical Networks[J].IEEE Journal on Selected Areas in communications,1996,14(5):903-908
[25]I.chlamtac,A.Ganz,G.Karmi,Lightpath Communications:An Approach to High Bandwidth optical WANs[J].IEEE Transactions on communications,1992,40(2):1171-1182
[26]R.Ramaswami,K.N.Sivarajan,Optimal Routing and wavelength Assignment in All-Optical Networks[C].Proceedings of IEEE INFOCOM'94,1994:110-119
[27]JingYi He,S.H.Gary Chan,Danny H.K.Tsang,Routing and Wavelength Assignment for WDM Multicast Networks[C].GLOBECOM'01.IEEE,2001:1536-1540
[28]Dijkstra E.A Note on Two Problems in Cormeetion with Graphs[C].Numeric Mathematic,1959,1:269-271
[29]Jong-Seon Kim,Lee D.C.Dynamic routing and wavelength assignment algorithms for multifiber WDM networks with many wavelengths[C].2nd European Conference on Universal Multiservice Networks,2002:180-186
[30]Robert W.Floyd.Algorithm 97:shortest path[J].Communications of the ACM,1962,5(6):345
[31]Yongbing Zhang,Taira K.,Takagi H,etc.An efficient heuristic for routing and wavelength assignment in optical WDM networks[C].IEEE International Conference on communications,ICC 2002.5,2002:2734-2739
[32]徐世中.WDM光传送网路由和波长分配算法研究[D].电子科技大学博士学位论文,2000
[33]顾畹仪.光传送网[M].北京:机械工业出版社,2003
[34]Hart P.,Nilsson N.,Raphael B.A Formal Basis for the Heuristic Determination of Minimum Cost Paths[C].IEEE Trans.Syst.Science and Cybernetics 1968,SSC-4(2):100-107
[35]Hart P.,Nilsson N.,Raphael B.Correction to "A Formal Basis for the Heuristic Determination of Minimum Cost Paths"[C].SIGART Newsletter,Dee.1972,No.37:29-29
[36]Nils J.Nilsson.人工智能[M].郑扣根,庄越挺译.北京:机械工业出版社,2000
[37]姚国庆.博弈论[M].天津:南开大学出版社,2003
[38]Robert Gibbons.博弈论基础[M].高峰译.北京:中国社会科学出版社,1999
[39]Nash J.Equilibrium Points in n-Person Games[C].Proceeding of the National Academy of Sciences 36:48-49
[40]Lee David,Libman Lavy,Orda Ariel.Path protection and blocking probability minimization in optical networks[C].Hong Kong:IEEE INFOCOM 2004,23rd Annual Joint Conference of the IEEE Computer and Communications Societies,2004:142-153
[41]Georgakopoulos George F.,Kavvadias Dimitris J.,Sioutis Leonidas G.Nash Equilibria in All-Optical Networks[C].Hong Kong:1st International Workshop on Internet and Network Economics,WINE 2005:1033-1045
[42]唐雄燕,左鹏.智能光网络-技术与应用实践[M].北京:电子工业出版社,2005
[43]RFC3209-RSVP-TE:Extensions to RSVP for LSP Tunnels.Network Working Group,2001
[44]严蔚敏,吴伟民.数据结构(C语言版)[M].北京:清华大学出版社,2002
[45]张可村,赵英良.数值计算的算法与分析[M].北京:科学出版社,2003
[46]马正飞,殷翔.数学计算方法与软件的工程应用[M].北京:化学工业出版社,2002
[47]张圣勤,MATLAB7.0实用教程[M].北京:机械工业出版社,2006
[48]吕文达.精通C#程序设计[M].北京:清华大学出版社,2004
[49]MATLAB Help Documentation.The MathWorks,Inc.2007
[50]赵士伟,赵明波,陈平.基于COM的MATLAB与C#.NET混合编程的实现与应用[J].山东理工大学学报(自然科学版),2006,20(4):26-29
[51]陈锡生.ASON动态光路建立的阻塞分析[J].中兴通讯技术,2004,6:17-19