Browse > Article

Network Routing by Traffic Prediction on Time Series Models  

Jung, Sang-Joon (경일대학교 교양학부)
Chung, Youn-Ky (경일대학교 컴퓨터공학부)
Kim, Chong-Gun (경일대학교 전자정보공학부)
Abstract
An increase In traffic has a large Influence on the performance of a total network. Therefore, traffic management has become an important issue of network management. In this paper, we propose a new routing algorithm that attempts to analyze network conditions using time series prediction models and to propose predictive optimal routing decisions. Traffic congestion is assumed when the predicting result is bigger than the permitted bandwidth. By collecting traffic in real network, the predictable model is obtained when it minimizes statistical errors. In order to predict network traffic based on time series models, we assume that models satisfy a stationary assumption. The stationary assumption can be evaluated by using ACF(Auto Correlation Function) and PACF(Partial Auto Correlation Function). We can obtain the result of these two functions when it satisfies the stationary assumption. We modify routing oaths by predicting traffic in order to avoid traffic congestion through experiments. As a result, Predicting traffic and balancing load by modifying paths allows us to avoid path congestion and increase network performance.
Keywords
rime series models; flouting algorithms; Traffic analyses;
Citations & Related Records
Times Cited By KSCI : 2  (Citation Analysis)
연도 인용수 순위
1 G. Michael Schneider, Tamas Nemeth, 'A simulation study of the OSPF-OMP routing algorithm,' Computer Networks, Volume 39, Issue 4, 15 July 2002, pp. 457-468   DOI   ScienceOn
2 M. Littman and J. Boyan, 'A distributed reinforcement learning scheme for network routing,' Technical Report CMU-CS-93-165, School of Computer Science, Carnegie Mellon University, 1993
3 J. A. Boyan and M. L. Littman, 'Packet routing in dynamically changing network: A reinforcement learning approach,' In J. Cowan, G. Tesauro, and J. Alspector, editors, Advances in Neural Information Processing Systems 6. MIT Press, Cambridge, MA, 1994
4 Dong Ho Kim, Keun Ho Ryu and Young So Cho, 'A new routing control technique using active temporal data management,' Journal of Systems and Software, Volume 51, Issue 1, 1 April 2000, pp. 37-48   DOI   ScienceOn
5 정상준, 김동주, 권영헌, 김종근, '네트워크 트래픽 예측을 위한 시계열 모형의 적합성 검증', 한국통신학회논문지 제 29권 제 2B 호, 2004, 2   과학기술학회마을
6 Wilinger, W., Wilson, D., Taqqu, M. 'Self-similar Traffic Modeling for Highspeed Networks,' ConneXions, Nov. 1994
7 이덕기, '예측방법의 이해', SPSS 아카데미, 1999
8 정동빈, 원태연, 'SPSS를 활용한 시계역 자료와 단순화 분석', SPSS 아카데미, 2001
9 최기헌, 이종협, 'SAS/ETS를 이용한 시계열 분석과 그 응용', 자유아카데미, 1994
10 F. C. M. Lau, Guihai Chen, Hao Huang and Li Xie, 'A distance-vector routing protocol for networks with unidirectional links,' Computer Communications, Volume 23, Issue 4, 15 February 2000, pp. 418-424   DOI
11 Ryan Kastner, Elaheh Bozorgzadeh and Majid Sarrafzadeh, 'Predictable routing,' Computer Aided Design, 2000. ICCAD-2000. IEEE/ACM International Conference on, 5-9 Nov. 2000
12 Xun Su and Gustavo de Veciana, 'Predictive routing to enhance QoS for stream-based flows sharing excess bandwidth,' Computer Networks, Volume 42, Issue 1, 15 May 2003, pp. 65-80   DOI   ScienceOn
13 김동일, 김창호, 'Ethernet 트래픽의 장기간 의존성 및 Self-Similiar 트래픽 소스 모델링에 관한 연구', Telecommunication Review, 제 11 권 6호, 2001, 12
14 Yantai Shu, Zhigang Jin, Lianfang Zhang, Lei Wang, 'Traffic Prediction Using FARIME Models.' IEEE International Conference on Communications, pp. 891-895. 1999. 6   DOI
15 홍원택, 안성진, 정진욱, '시계열 분석을 이용한 SNMP MIB-II 기반의 회선 이용률 예측 기법', 한국정보처리학회 논문지 제 6 권 제 9 호, 1999. 9   과학기술학회마을
16 이강원, 김태윤, '효율적인 통신망 설계를 위한 예측 시스템 설계', 한국정보과학회 논문지, 제25권 제1호, pp. 76-82, 1998. 1
17 김희진, 박상근, 최덕재, '네트워크 트래픽 감시를 통한 동적인 트래픽 부하 분산에 관한 연구', 한국통신학회 춘계학술발표논문집, 2002, 5