Optimizing energy efficiency of LoRaWAN-based wireless underground sensor networks: A multi-agent reinforcement learning approach

被引:12
作者
Zhao, Guozheng [1 ]
Lin, Kaiqiang [1 ]
Chapman, David [2 ]
Metje, Nicole [2 ]
Hao, Tong [1 ]
机构
[1] Tongji Univ, Coll Surveying & Geoinformat, 1239 Siping Rd, Shanghai 200092, Peoples R China
[2] Univ Birmingham, Dept Civil Engn, Birmingham B15 2TT, England
基金
中国国家自然科学基金;
关键词
Energy efficiency; Multi-agent reinforcement learning; LoRaWAN-based WUSNs; Dynamic underground environment; Adaptive data rate;
D O I
10.1016/j.iot.2023.100776
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Extended battery lifetime is always desirable for wireless underground sensor networks (WUSNs). The recent integration of LoRaWAN-grade massive machine-type communications (MTC) technology and WUSNs, namely LoRaWAN-based WUSNs, provides a promisingly energy -efficient solution for underground monitoring. However, due to the limited battery energy, massive data collisions and dynamic underground environment, the energy efficiency of LoRaWAN-based WUSNs still possesses a significant challenge. To propose a liable solution, we advocate reinforcement learning (RL) for managing the transmission configuration of underground sensors. In this paper, we firstly develop the multi-agent RL (MARL) algorithm to improve the network energy efficiency, which considers the link quality, energy consumption and packet collisions between packets. Secondly, a reward mechanism is proposed, which is used to define independent state and action for every node to improve the adaptability of the proposed algorithm to dynamic underground environment. Furthermore, through the simulations in different underground environment and at different network scales, our results highlight that the proposed MARL algorithm can quickly optimize the network energy efficiency and far exceed the traditional adaptive data rate (ADR) mechanism. Finally, our proposed algorithm is successfully demonstrated to be able to efficiently adapt to the dynamically changing underground environment. This work provides insights into the energy efficiency optimization, and will lay the foundation for future realistic deployments of LoRaWAN-based WUSNs.
引用
收藏
页数:18
相关论文
共 50 条
  • [31] A multi-agent reinforcement learning based approach for intelligent traffic signal control
    Benhamza, Karima
    Seridi, Hamid
    Agguini, Meriem
    Bentagine, Amel
    EVOLVING SYSTEMS, 2024, 15 (06) : 2383 - 2397
  • [32] Energy Management Simulation with Multi-Agent Reinforcement Learning: An Approach to Achieve Reliability and Resilience
    Deshpande, Kapil
    Moehl, Philipp
    Haemmerle, Alexander
    Weichhart, Georg
    Zoerrer, Helmut
    Pichler, Andreas
    ENERGIES, 2022, 15 (19)
  • [33] Genetic Agent-Based Framework for Energy Efficiency in Wireless Sensor Networks
    Lee, Jangsu
    Kim, Sungchun
    IEICE TRANSACTIONS ON COMMUNICATIONS, 2011, E94B (06) : 1736 - 1739
  • [34] Reinforcement learning based energy efficient protocol for wireless multimedia sensor networks
    Joshi, Upasna
    Kumar, Rajiv
    MULTIMEDIA TOOLS AND APPLICATIONS, 2022, 81 (02) : 2827 - 2840
  • [35] Multi-agent Reinforcement Learning and Its Application to Wireless Network Communication
    Pochaba, Sabrina
    Dorfinger, Peter
    Herlich, Matthias
    Kwitt, Roland
    Hirlaender, Simon
    COMBINING, MODELLING AND ANALYZING IMPRECISION, RANDOMNESS AND DEPENDENCE, SMPS 2024, 2024, 1458 : 363 - 370
  • [36] Reinforcement learning based energy efficient protocol for wireless multimedia sensor networks
    Upasna Joshi
    Rajiv Kumar
    Multimedia Tools and Applications, 2022, 81 : 2827 - 2840
  • [37] Towards Multi-agent Reinforcement Learning for Wireless Network Protocol Synthesis
    Dutta, Hrishikesh
    Biswas, Subir
    2021 INTERNATIONAL CONFERENCE ON COMMUNICATION SYSTEMS & NETWORKS (COMSNETS), 2021, : 614 - 622
  • [38] A Reinforcement Learning Approach to Energy Efficiency and QoS in 5G Wireless Networks
    Wang, Ying
    Dai, Xiangming
    Wang, Jason Min
    Bensaou, Brahim
    IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, 2019, 37 (06) : 1413 - 1423
  • [39] Reinforcement learning–enabled efficient data gathering in underground wireless sensor networks
    Deng Zhao
    Zhangbing Zhou
    Shangguang Wang
    Bo Liu
    Walid Gaaloul
    Personal and Ubiquitous Computing, 2023, 27 : 581 - 598
  • [40] A Multi-agent Reinforcement Learning-Based Transmission Protocol for Underwater Acoustic Networks
    Gao, Yu
    Bi, Zhicheng
    Wang, Chaofeng
    17TH ACM INTERNATIONAL CONFERENCE ON UNDERWATER NETWORKS & SYSTEMS, WUWNET 2023, 2024,