Adaptive Routing for Sensor Networks using Reinforcement Learning
Abstract
Efficient and robust routing is central to wireless sensor networks (WSN) that feature energy-constrained nodes, unreliable links, and frequent topology change. While most existing routing techniques are designed to reduce routing cost by optimizing one goal, e.g., routing path length, load balance, re-transmission rate, etc, in real scenarios however, these factors affect the routing performance in a complex way, leading to the need of a more sophisticated scheme that makes correct trade-offs. In this paper, we present a novel routing scheme, AdaR that adaptively learns an optimal routing strategy, depending on multiple optimization goals. We base our approach on a least squares reinforcement learning technique, which is both data efficient, and insensitive against initial setting, thus ideal for the context of ad-hoc sensor networks. Experimental results suggest a significant performance gain over a naive Q-learning based implementation.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
AdaR has higher computational complexity and communication overhead than Q-learning.
“this comes at the price of higher computational complexity, and communication overhead due to the growing size of the packets at each hop and the broadcasting of Q-values”
From A comprehensive survey on machine learning for networking: Evolution, applications and research opportunities (2018) · cites this paper as [461]
AdaR is more sensitive to link failures and node mobility.
“which also makes it more sensitive to link failures and node mobility.”
From A comprehensive survey on machine learning for networking: Evolution, applications and research opportunities (2018) · cites this paper as [461]
Q-learning took 40 episodes to reach comparable success rates to AdaR.
“whereas, it took 40 episodes for Q-learning to reach comparable success rates.”
From A comprehensive survey on machine learning for networking: Evolution, applications and research opportunities (2018) · cites this paper as [461]
AdaR reaches a routing success rate of ∼95% with low deviation before the 5th learning episode.
“a routing success rate of ∼ 95% with a low deviation was reached even before the 5th learning episode”
From A comprehensive survey on machine learning for networking: Evolution, applications and research opportunities (2018) · cites this paper as [461]
Q-learning is more sensitive to the initial choice of Q-values than AdaR.
“the authors show that Q-learning is more sensitive to the initial choice of Q-values than AdaR.”
From A comprehensive survey on machine learning for networking: Evolution, applications and research opportunities (2018) · cites this paper as [461]
AdaR converges faster than Q-learning.
“Results show that the algorithm converges faster than Q-learning”
From A comprehensive survey on machine learning for networking: Evolution, applications and research opportunities (2018) · cites this paper as [461]
Cites
Reinforcement Learning: An Introduction
Sutton, R. S., & Barto, A. (2005) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.2004.842673 · Cited by 25,263 · see citing works
Akyildiz, I. F., Su, W., Sankarasubramaniam, Y., & Cayirci, E. (2002) · IEEE Communications Magazine · doi:10.1109/mcom.2002.1024422 · Cited by 13,620 · see citing works
Watkins, C. J. C. H., & Dayan, P. (1992) · Machine Learning · doi:10.1007/bf00992698 · Cited by 9,083 · see citing works
Karp, B., & Kung, H. T. (2000) · Proceedings of the 28th Annual International Conference on Mobile Computing And Networking · doi:10.1145/345910.345953 · Cited by 7,077 · see citing works
Intanagonwiwat, C., Govindan, R., & Estrin, D. L. (2000) · ACM Annual International Conference on Mobile Computing and Networking (MobiCom) · doi:10.1145/345910.345920 · Cited by 5,391 · see citing works
A review of current routing protocols for ad hoc mobile wireless networks
Royer, E. M., & Toh, C. K. (1999) · IEEE Personal Communications · doi:10.1109/98.760423 · Cited by 3,772 · see citing works
Energy aware routing for low energy ad hoc sensor networks
Shah, R. C., & Rabaey, J. M. (2003) · doi:10.1109/wcnc.2002.993520 · Cited by 1,463 · see citing works
Guest Editors' Introduction: Overview of Sensor Networks
Culler, D., Estrin, D. L., & Srivastava, M. B. (2004) · Computer · doi:10.1109/mc.2004.93 · Cited by 1,451 · see citing works
Cruller, D., Estrin, D. L., & Srivastava, M. B. (2004) · Computer · Cited by 1,263 · see citing works
Packet Routing in Dynamically Changing Networks: A Reinforcement Learning Approach
Boyan, J. A., & Littman, M. L. (1994) · Cited by 694 · see citing works
(1992) · Springer eBooks · doi:10.1007/978-1-4615-3618-5 · Cited by 334 · see citing works
Geometric spanner for routing in mobile networks
Gao, J., Guibas, L., Hershberger, J. E., et al. (2001) · ACM International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc) · doi:10.1145/501422.501424 · Cited by 262 · see citing works
Least-Squares Temporal Difference Learning
Boyan, J. A. (1999) · Cited by 250 · see citing works
Model-Based Bayesian ExplorationPreprint
Dearden, R., Friedman, N., & André, D. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1301.6690 · Cited by 226 · see citing works
Geometric spanner for routing in mobile networks
Gao, J., Guibas, L., Hershberger, J. E., et al. (2001) · ACM International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc) · doi:10.1145/501416.501424 · Cited by 159 · see citing works
Model-Free Least-Squares Policy Iteration
Lagoudakis, M. G., & Parr, R. (2001) · Cited by 71 · see citing works
Reinforcement learning for sensing strategies
Kwok, C., & Fox, D. (2005) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2004.1389903 · Cited by 61 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Boutaba, R., Salahuddin, M. A., Limam, N., Ayoubi, S., Shahriar, N., Estrada-Solano, F., & Caicedo, O. M. (2018) · Journal of Internet Services and Applications · doi:10.1186/s13174-018-0087-2 · Cited by 1,036 · see citing works
Hu, T., & Fei, Y. (2010) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2010.28 · Cited by 405 · see citing works
- Authors
- Wang, P., & Wang, T.
- Affiliations
- Zhejiang University (CN)
- University of British Columbia (CA)
- ORCID
- Venue
- The Sixth IEEE International Conference on Computer and Information Technology (CIT'06)
- Publisher
- IEEE
- Published
- 2006-09
- Pages
- 219-219
- OpenAlex
- W2119632647
- Cited by
- 92
- References
- 17
- Citations by year
- 2025: 4 · 2024: 3 · 2023: 4 · 2022: 6 · 2021: 7 · 2020: 6 · 2019: 5 · 2018: 8 · 2017: 2 · 2016: 3 · 2015: 3 · 2014: 5 · 2013: 7 · 2012: 6
- FWCI
- 3.26
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE International Conference on Computer and Information Technology (CIT) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Energy Efficient Wireless Sensor Networks
- Physical Sciences › Computer Science › Computer Networks and Communications › Security in Wireless Sensor Networks
- Physical Sciences › Computer Science › Computer Networks and Communications › Mobile Ad Hoc Networks
- Keywords
- wireless sensor networks, adaptive routing, reinforcement learning, energy-constrained nodes, topology change, Q-learning comparison, ADAR
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
