Application of reinforcement learning to wireless sensor networks: models and algorithms

Kok Lim Alvin Yau, Hock Guan Goh, David Chieng, Kae Hsiang Kwong

Research output: Journal PublicationArticlepeer-review

40 Citations (Scopus)


Wireless sensor network (WSN) consists of a large number of sensors and sink nodes which are used to monitor events or environmental parameters, such as movement, temperature, humidity, etc. Reinforcement learning (RL) has been applied in a wide range of schemes in WSNs, such as cooperative communication, routing and rate control, so that the sensors and sink nodes are able to observe and carry out optimal actions on their respective operating environment for network and application performance enhancements. This article provides an extensive review on the application of RL to WSNs. This covers many components and features of RL, such as state, action and reward. This article presents how most schemes in WSNs have been approached using the traditional and enhanced RL models and algorithms. It also presents performance enhancements brought about by the RL algorithms, and open issues associated with the application of RL in WSNs. This article aims to establish a foundation in order to spark new research interests in this area. Our discussion has been presented in a tutorial manner so that it is comprehensive and applicable to readers outside the specialty of both RL and WSNs.

Original languageEnglish
Pages (from-to)1045-1075
Number of pages31
JournalComputing (Vienna/New York)
Issue number11
Publication statusPublished - 24 Nov 2015
Externally publishedYes


  • Artificial intelligence
  • Context awareness
  • Q-learning
  • Reinforcement learning
  • Wireless sensor networks

ASJC Scopus subject areas

  • Software
  • Computational Mathematics
  • Theoretical Computer Science
  • Numerical Analysis
  • Computer Science Applications
  • Computational Theory and Mathematics


Dive into the research topics of 'Application of reinforcement learning to wireless sensor networks: models and algorithms'. Together they form a unique fingerprint.

Cite this