Intelligent IoT Connectivity: Deep Reinforcement Learning Approach

Minhae Kwon, Juhyeon Lee, Hyunggon Park

Research output: Contribution to journalArticlepeer-review

39 Scopus citations


In this paper, we propose a distributed solution to design a multi-hop ad hoc Internet of Things (IoT) network where mobile IoT devices strategically determine their wireless transmission ranges based on a deep reinforcement learning approach. We consider scenarios where only a limited networking infrastructure is available but a large number of IoT devices are deployed in building a multi-hop ad hoc network to deliver source data to the destination. An IoT device is considered as a decision-making agent that strategically determines its transmission range in a way that maximizes network throughput while minimizing the corresponding transmission power consumption. Each IoT device collects information from its partial observations and learns its environment through a sequence of experiences. Hence, the proposed solution requires only a minimal amount of information from the system. We show that the actions that the IoT devices take from its policy are determined as to activate or inactivate its transmission, i.e., only necessary relay nodes are activated with the maximum transmit power, and nonessential nodes are deactivated to minimize power consumption. Using extensive experiments, we confirm that the proposed solution builds a network with higher network performance than the current state-of-the-art solutions in terms of system goodput and connectivity ratio.

Original languageEnglish
Article number8886442
Pages (from-to)2782-2791
Number of pages10
JournalIEEE Sensors Journal
Issue number5
StatePublished - 1 Mar 2020

Bibliographical note

Publisher Copyright:
© 2001-2012 IEEE.


  • Intelligent IoT connectivity
  • deep reinforcement learning
  • mobile relay networks
  • network formation
  • network topology design
  • wireless ad hoc networks


Dive into the research topics of 'Intelligent IoT Connectivity: Deep Reinforcement Learning Approach'. Together they form a unique fingerprint.

Cite this