移动综合应急预警系统的设计与实现
|
摘要
移动通信网络的运行安全与稳定对于国家安全和社会稳定至关重要。为了建立健全通信应急工作机制与国家通信保障,国务院于2006年出台《国家通信保障应急预案》。各通信运营商纷纷建立各自的网络应急预案,提升网络应急预警水平。
目前,应急预警的现状是移动网络优化人员需要使用设备生产商提供的设备监控软件实现优化。该方法采用人工手动方式,对大量字符信息进行监视、人工预警,通过手动逐行指令的方式进行网络参数调节。
为将应急预警过程自动化、智能化,本文设计并实现了移动综合应急预警系统。系统采用C/S架构,C/S之间采用基于UDP的协议进行通信。系统主要由告警采集进程、服务器端进程和客户端进程三部分构成。首先,告警采集进程使用IOCP作为网络IO模型,满足网络并发性高的要求,设计的内存池加高效缓冲方案提高了网络吞吐量；其次,服务器端进程实现预警判决、任务调度、预警通知以及预警实施；再次,客户端进程实现了数据统计、报表生成以及对服务器端进程的配置,包括任务配置、预警门限配置、短信模板编辑、预警指令生成等功能；最后,对网络运行的重要参数CP负荷使用支持向量机回归预测方法进行预测。本系统采用Microsoft Visual Studio 2008作为开发平台,开发语言采用C++和C≠,数据库采用Microsoft SQL Server 2005。
该系统已运行半年时间,可以实时监控100台网元,数据采集的吞吐量不低于854KB/s,告警数据采集周期为5分钟,预警周期为15分钟,CP负荷预测的平均绝对百分误差为1.7912%,系统运行稳定、可靠,达到预期目标。
The safety and stability of mobile communication network the operation is very important for national security and social stability.In order to establish and perfect the national communication security and communications emergency work mechanism, in 2006 the State Council published The National Communication Support Emergency Plan.The telecomm-unication operators are establishing their own network emergency plan to improve emergency warning level.
Currently, the situation of emergency warning is that mobile network optimization staff uses equipment monitoring software provided by equipment manufacturers to realize optimization. The method is using the manual way, monitoring large character information, artificial warning and adjusting network parameters through the manual command line.
To emergency warning process automatically and intelligently, this paper designs and realizes mobile comprehensive emergency and warning system.System using C/S architecture. Server and clients communicate using protocol based on UDP.It mainly divides into alarm collecting process, the server process and the client process. Alarm collection peocess having the highest network simultaneity demand use IOCP as IO model, the paper presents the design of memory pool and efficient buffer solution improve system throughput.The server process realize warning task scheduling, judgment and the early warning notice and warning implementation. The client process majors in data statistics and report generation as well as to the server process configuration, including task configuration,warning threshold configuration, message template editing,warning instructions generation etc.Finally,it forecasts the important parameter of the network operation CP load using support vector machine regression forecast method. The system uses Microsoft Visual Studio 2008 as a development platform, using C++and C# as the development language, using Microsoft SQL Server 2005 as database.
The system has been in operation for 6 months, which can real-time monitoring 100 network nodes. Data acquisition throughput is not less than 854 KB/s, alarm data collection period is 5 minutes and a warning process cycle is 15 minutes. The MAPE of CP load forecasting is 1.7912%.Systen operates stably and reliably, achieving the anticipated goal.
目前,应急预警的现状是移动网络优化人员需要使用设备生产商提供的设备监控软件实现优化。该方法采用人工手动方式,对大量字符信息进行监视、人工预警,通过手动逐行指令的方式进行网络参数调节。
为将应急预警过程自动化、智能化,本文设计并实现了移动综合应急预警系统。系统采用C/S架构,C/S之间采用基于UDP的协议进行通信。系统主要由告警采集进程、服务器端进程和客户端进程三部分构成。首先,告警采集进程使用IOCP作为网络IO模型,满足网络并发性高的要求,设计的内存池加高效缓冲方案提高了网络吞吐量；其次,服务器端进程实现预警判决、任务调度、预警通知以及预警实施；再次,客户端进程实现了数据统计、报表生成以及对服务器端进程的配置,包括任务配置、预警门限配置、短信模板编辑、预警指令生成等功能；最后,对网络运行的重要参数CP负荷使用支持向量机回归预测方法进行预测。本系统采用Microsoft Visual Studio 2008作为开发平台,开发语言采用C++和C≠,数据库采用Microsoft SQL Server 2005。
该系统已运行半年时间,可以实时监控100台网元,数据采集的吞吐量不低于854KB/s,告警数据采集周期为5分钟,预警周期为15分钟,CP负荷预测的平均绝对百分误差为1.7912%,系统运行稳定、可靠,达到预期目标。
The safety and stability of mobile communication network the operation is very important for national security and social stability.In order to establish and perfect the national communication security and communications emergency work mechanism, in 2006 the State Council published The National Communication Support Emergency Plan.The telecomm-unication operators are establishing their own network emergency plan to improve emergency warning level.
Currently, the situation of emergency warning is that mobile network optimization staff uses equipment monitoring software provided by equipment manufacturers to realize optimization. The method is using the manual way, monitoring large character information, artificial warning and adjusting network parameters through the manual command line.
To emergency warning process automatically and intelligently, this paper designs and realizes mobile comprehensive emergency and warning system.System using C/S architecture. Server and clients communicate using protocol based on UDP.It mainly divides into alarm collecting process, the server process and the client process. Alarm collection peocess having the highest network simultaneity demand use IOCP as IO model, the paper presents the design of memory pool and efficient buffer solution improve system throughput.The server process realize warning task scheduling, judgment and the early warning notice and warning implementation. The client process majors in data statistics and report generation as well as to the server process configuration, including task configuration,warning threshold configuration, message template editing,warning instructions generation etc.Finally,it forecasts the important parameter of the network operation CP load using support vector machine regression forecast method. The system uses Microsoft Visual Studio 2008 as a development platform, using C++and C# as the development language, using Microsoft SQL Server 2005 as database.
The system has been in operation for 6 months, which can real-time monitoring 100 network nodes. Data acquisition throughput is not less than 854 KB/s, alarm data collection period is 5 minutes and a warning process cycle is 15 minutes. The MAPE of CP load forecasting is 1.7912%.Systen operates stably and reliably, achieving the anticipated goal.
引文
[1]张威.GSM网络优化——原理与工程(第2版)[M].北京：人民邮电出版社,2010,1.
[2]GSM网络优化算法分析[EB/OL]. http://www.dzsc.com/data/html/2011-9-9/9834 6. html,2011,9.
[3]杨燕,孙震强.浅析3G网络优化方法[J].中国无线电,2006,(10)：21-22.
[4]姚国章,袁敏.构建江苏应急通信保障体系的对策研究[J].南京邮电大学学报,2010,12(1)：22-27.
[5]王玉峰,孟苏英,刘英.应急通信保障情况综合处理平台研究[J].无线电通信技术,2009,35(3)：55-58.
[6]赵蕾.应急通信保障系统的设计与实现[J]电信科学,2011(3)：111-117.
[7]Anthony Jones, Jim Ohlund.杨合庆译.Windows网络编程(第2版)[M].北京：清华大学出版社,2002,10.
[8]李凌.Winsock2网络编程实用教程[M].北京：清华大学出版社,2003,11.
[9]苏羽,王媛媛.Visual C++网络游戏建模与实现[M].北京：北京科海电子出版社,2003,6.
[10]陈和平,周静宁,顾晋广等.IOCP机制与网络代理服务器实现方法[J].计算机应用,2003,4(23)：109-110.
[11]Gyu-Baek, Kim. A Method of Generating Massive Virtual Clients and Model-based Performance Test[C]. In:Proceeding of the Fifth International Conference on Quality Software,2005,250-254.
[12]Mark Russinovich. Inside IO Completion Port. [EB/OL]. http://technet.microsoft. com/en-us/sysinternals/bb963891.aspx,2006,10,1.
[13]飞狼,李春萌,杨涵.SQL Server2005数据库管理与应用指南[M].北京：人民邮电出版社,2007,8.
[14]Christian Nagel,Bill Evjen, Jay Glynn.李敏波译.C#高级编程(第4版)[M].北京：清华大学出版社,2006,10.
[15]Nello Cristianini, John Shawe-Taylor.李国正等译.支持向量机导论[M].北京：电子工业出版社,2006,5.
[16]李海生.支持向量机回归算法与应用研究[D].广州：华南理工大学,2005,5.
[17]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)：33-41
[18]汤华丽.SVM中两类常用分类方法的关系研究[D].重庆：重庆大学,2005,5.
[19]Keng-Pei Lin, Ming-Syan Chen. On the Design and Analysis of the Privacy-Preserving SVM Classier[J]. Knowledge and Data Engineering, IEEE Transactions,2011,23(11):1704-1717.
[20]梁燕.SVM分类器的扩展及其应用研究[D].长沙：湖南大学,2008,10.
[21]Masayuki karasuyama, Ryohei Nakano. Optimizing SVR Hyperparameters via Fast Cross-Validation using AOSVR[C]. In:Proceedings Of International Joint Conference on Neural Networks, Orlando, Florida, USA,2007,1186-1191.
[22]田英杰.支持向量回归机及其应用研究[D].北京：中国农业大学,2005,6.
[23]王震,徐博,解永平,孙加奉.单客户机-多服务器模式下IOCP的应用与研究[J].2011,38(10A)：385-388,405.
[24]侯捷.池内春秋——Memory Pool的设计哲学和无痛设计[J].程序员,2002,(9)：94-97.
[25]吕金锁.探针技术在网络信息监控系统中的应用[D].哈尔滨：哈尔滨工程大学,2007,3.
[26]P. Melas, E. J. Zaluska. Performance of message-passing systems using a zero-copy communication protocol[C]. In:1998 International Conference on Parallel Architectures and Compilation Techniques,1998,264-271.
[27]徐力.一种基于实时环境的零拷贝通信技术的研究[D].武汉：华中科技大学,2006,5.
[28]Xiao Chen. An Improved Method of Zero-Copy Data Transmission in the High Speed Network Environment[C]. In:2009 International Conference on Multimedia Information Networking and Security,2009, (2),281-284.
[29]朱吉.基于ASP. NET的仓库管理信息系统的设计与实现[D].武汉：华中科技大学,2007,6.
[30]Alan Beaulieu.张伟超,林青松译.SQL学习指南[M].北京：人民邮电出版社,2010,6
[31]Kalen Delaney.金成姬,陈绍英译.SQL Server2005技术内幕：查询、调整和优化[M].北京：电子工业出版社.2009,7.
[32]Galindo-Legaria, C. A.; Grabs, T.; Gukal, S.; Herbert, S.; Surna, A.; Wang, S.; Wei Yu; Zabback, P.; Shin Zhang. Optimizing Star Join Queries for Data Warehousing in Microsoft SQL Server [C].In:Data Engineering, ICDE 2008, IEEE 24th International Conference,2008,1190-1199.
[33]Ordonez, C. Optimization of Linear Recursive Queries in SQL[J]. Knowledge and Data Engineering, IEEE Transactions,2010,22(2):264-277.
[34]韩书磊,冯元明.爱立信交换机CP负荷的研究[C].中国通信学会第五届学术年会论文集,2008,1.
[35]雷秀娟,史忠科,周亦鹏.PSO优化算法演变及其融合策略[J].计算机工程与应用,2007,43(7)：90-92.
[36]李鹏超.基于模拟退火算法和支持向量回归的网格资源预测[D].长春：吉林大学,2010,1.
[37]周建新,杨卫东,李擎.求解连续函数优化问题的改进蚁群算法及仿真[J].系统仿真学报,2009,21(6)：1685-1688.
[2]GSM网络优化算法分析[EB/OL]. http://www.dzsc.com/data/html/2011-9-9/9834 6. html,2011,9.
[3]杨燕,孙震强.浅析3G网络优化方法[J].中国无线电,2006,(10)：21-22.
[4]姚国章,袁敏.构建江苏应急通信保障体系的对策研究[J].南京邮电大学学报,2010,12(1)：22-27.
[5]王玉峰,孟苏英,刘英.应急通信保障情况综合处理平台研究[J].无线电通信技术,2009,35(3)：55-58.
[6]赵蕾.应急通信保障系统的设计与实现[J]电信科学,2011(3)：111-117.
[7]Anthony Jones, Jim Ohlund.杨合庆译.Windows网络编程(第2版)[M].北京：清华大学出版社,2002,10.
[8]李凌.Winsock2网络编程实用教程[M].北京：清华大学出版社,2003,11.
[9]苏羽,王媛媛.Visual C++网络游戏建模与实现[M].北京：北京科海电子出版社,2003,6.
[10]陈和平,周静宁,顾晋广等.IOCP机制与网络代理服务器实现方法[J].计算机应用,2003,4(23)：109-110.
[11]Gyu-Baek, Kim. A Method of Generating Massive Virtual Clients and Model-based Performance Test[C]. In:Proceeding of the Fifth International Conference on Quality Software,2005,250-254.
[12]Mark Russinovich. Inside IO Completion Port. [EB/OL]. http://technet.microsoft. com/en-us/sysinternals/bb963891.aspx,2006,10,1.
[13]飞狼,李春萌,杨涵.SQL Server2005数据库管理与应用指南[M].北京：人民邮电出版社,2007,8.
[14]Christian Nagel,Bill Evjen, Jay Glynn.李敏波译.C#高级编程(第4版)[M].北京：清华大学出版社,2006,10.
[15]Nello Cristianini, John Shawe-Taylor.李国正等译.支持向量机导论[M].北京：电子工业出版社,2006,5.
[16]李海生.支持向量机回归算法与应用研究[D].广州：华南理工大学,2005,5.
[17]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)：33-41
[18]汤华丽.SVM中两类常用分类方法的关系研究[D].重庆：重庆大学,2005,5.
[19]Keng-Pei Lin, Ming-Syan Chen. On the Design and Analysis of the Privacy-Preserving SVM Classier[J]. Knowledge and Data Engineering, IEEE Transactions,2011,23(11):1704-1717.
[20]梁燕.SVM分类器的扩展及其应用研究[D].长沙：湖南大学,2008,10.
[21]Masayuki karasuyama, Ryohei Nakano. Optimizing SVR Hyperparameters via Fast Cross-Validation using AOSVR[C]. In:Proceedings Of International Joint Conference on Neural Networks, Orlando, Florida, USA,2007,1186-1191.
[22]田英杰.支持向量回归机及其应用研究[D].北京：中国农业大学,2005,6.
[23]王震,徐博,解永平,孙加奉.单客户机-多服务器模式下IOCP的应用与研究[J].2011,38(10A)：385-388,405.
[24]侯捷.池内春秋——Memory Pool的设计哲学和无痛设计[J].程序员,2002,(9)：94-97.
[25]吕金锁.探针技术在网络信息监控系统中的应用[D].哈尔滨：哈尔滨工程大学,2007,3.
[26]P. Melas, E. J. Zaluska. Performance of message-passing systems using a zero-copy communication protocol[C]. In:1998 International Conference on Parallel Architectures and Compilation Techniques,1998,264-271.
[27]徐力.一种基于实时环境的零拷贝通信技术的研究[D].武汉：华中科技大学,2006,5.
[28]Xiao Chen. An Improved Method of Zero-Copy Data Transmission in the High Speed Network Environment[C]. In:2009 International Conference on Multimedia Information Networking and Security,2009, (2),281-284.
[29]朱吉.基于ASP. NET的仓库管理信息系统的设计与实现[D].武汉：华中科技大学,2007,6.
[30]Alan Beaulieu.张伟超,林青松译.SQL学习指南[M].北京：人民邮电出版社,2010,6
[31]Kalen Delaney.金成姬,陈绍英译.SQL Server2005技术内幕：查询、调整和优化[M].北京：电子工业出版社.2009,7.
[32]Galindo-Legaria, C. A.; Grabs, T.; Gukal, S.; Herbert, S.; Surna, A.; Wang, S.; Wei Yu; Zabback, P.; Shin Zhang. Optimizing Star Join Queries for Data Warehousing in Microsoft SQL Server [C].In:Data Engineering, ICDE 2008, IEEE 24th International Conference,2008,1190-1199.
[33]Ordonez, C. Optimization of Linear Recursive Queries in SQL[J]. Knowledge and Data Engineering, IEEE Transactions,2010,22(2):264-277.
[34]韩书磊,冯元明.爱立信交换机CP负荷的研究[C].中国通信学会第五届学术年会论文集,2008,1.
[35]雷秀娟,史忠科,周亦鹏.PSO优化算法演变及其融合策略[J].计算机工程与应用,2007,43(7)：90-92.
[36]李鹏超.基于模拟退火算法和支持向量回归的网格资源预测[D].长春：吉林大学,2010,1.
[37]周建新,杨卫东,李擎.求解连续函数优化问题的改进蚁群算法及仿真[J].系统仿真学报,2009,21(6)：1685-1688.