ROUTING ALGORITHM OF THE FAMILY OF Q-ROUTING, BASED ON DYNAMIC CHANGES OF LEARNING RATES ACCORDING TO THE ESTIMATES OF THE AVERAGE DELIVERY TIME

Authors

  • Yu.A. Shilova Perm National Research Polytechnic University

DOI:

Keywords:

Q-routing, network, routing, algorithm, learning rate

Abstract

One of the most urgent problems for mobile ad hoc network is the routing problem. A routing algorithm of the family of Q-Routing is developed and called Adaptive Rate Full Echo, and simulation of this algorithm has yielded good results. This algorithm is based on dynamic changes of learning rates according to estimates of the average network delay. The article also provides a brief overview of different routing protocols and algorithms, it provides a more detailed description of the basic algorithm of Q-Routing.

Author Biography

  • Yu.A. Shilova, Perm National Research Polytechnic University
    магистрант группы Телекоммуникации1-13-1м, Пермский национальный исследовательский политехнический университет (ПНИПУ), 614013, г. Пермь, ул. Профессора Поздеева, 7; e-mail: marissaspiritte@mail.ru

References

  1. Boyan J.A., Littman M.L. Packet routing in dynamically changing networks: A reinforcement learning approach // Advances in neural information processing systems. - 1994. - P. 671-671.
  2. Sutton R., Barto A.G. Reinforcement Learning: An Introduction: - Cambridge : MIT Press, 1998. - 328 p.
  3. Vinokurov V.M., Pugovkin A.V. Marsrutizacia v besprovodnyh mobil’nyh Ad hoc-setah // Doklady TUSURa, No 2 (22), c. 1, dekabr’ 2010, s. 288-292.
  4. Desai R., Patil B.P. Reinforcement learning for adaptive network routing //Computing for Sustainable Global Development (INDIACom), 2014 International Conference on. - IEEE, 2014. - P. 815-818.
  5. Silova U.A., Kavalerov M.V. Issledovanie vliania parametra skorosti obucenia na rezul’taty raboty algoritma marsrutizacii Q-routing // Innovacionnye tehnologii: teoria, instrumenty, praktika: Sb. tr. mezdunarodn. konf.  Perm’, PNIPU, 2015.  S. 172179.
  6. Nekrutkin V.V. Modelirovanie raspredelenij. - [Elektron. dannye]. - 90 s.
  7. Choi S., Yeung D.Y. Predictive Q-routing: A memory-based reinforcement learning approach to adaptive traffic control // Advances in Neural Information Processing Systems. - 1996. - T. 8. - P. 945-951.
  8. Kumar S., Miikkulainen R. Dual Reinforcement Q-routing: An On-Line Adaptive routing Algorithm // In Proceedings of the Artificial Neural Networks in Engineering Conference (St. Louis), 1997. - P. 231-238.
  9. Tao N., Baxter J., Weaver L. A multi-agent, policy-gradient approach to network routing // In Proceedings of the 18th Int. Conf. on Machine Learning. - 2001. - P. 553-560.
  10. Subramanian D., Druschel P., Chen J. Ants and reinforcement learning: A case study in routing in dynamic networks // IJCAI (2). - 1997. - P. 832-839.

Downloads

Published

2015-10-14

Issue

No. 2 (2015)

Section

Research: theory and experiment

How to Cite

Shilova, Y. . (2015). ROUTING ALGORITHM OF THE FAMILY OF Q-ROUTING, BASED ON DYNAMIC CHANGES OF LEARNING RATES ACCORDING TO THE ESTIMATES OF THE AVERAGE DELIVERY TIME. Perm Federal Research Centre Journal, 2, 79-93. https://journal.permsc.ru/index.php/pscj/article/view/PSCJ2015n2p13