Scholarlink

<p>Fast and stable communication in complex aquatic environments is essential for the coordinated operation of underwater robotic clusters. Inspired by the communication mechanisms of weakly electric fish, underwater electrocommunication has emerged as a research hotspot due to its low power consumption, excellent omnidirectionality, and stable transmission. However, when electrical signals are used for communication, the frequency increases and the attenuation is intensified, resulting in shorter transmission distances and different attenuation degrees in different underwater environments. Though lowering the frequency reduces attenuation, it limits the communication rate. Therefore, when the communication distance is fixed, the rational selection of communication frequency can effectively enhance the transmission rate between underwater communication nodes. This study proposes a communication frequency adaptive selection method based on electric field simulation and Joint Distribution Adaptation (JDA) to enable the transmitting electrodes to select the optimal communication frequency by the variations in the underwater environment and communication distance, thereby transmitting the information to the receiving electrodes at the highest rate. Firstly, a three-layer medium simulation model is constructed based on the underwater electric field communication theory. Then, the degree of influence of various factors in different water environments on communication frequency and transmission distance is obtained by analyzing the simulation data. It is finally determined by taking the conductivity, water depth, communication distance, spacing between transmitting electrodes, and deployment depth as the model input parameters. Secondly, an air–seawater–soil model is employed to generate the source domain dataset, while four typical types of marine terrain—plain, concave, convex, and slope—are used to construct the target domain dataset. JDA is applied to transfer knowledge from the source domain to the target domain, thereby enhancing the model’s generalization capability. Finally, the proposed method is verified through multiple evaluation metrics, such as Accuracy (Acc) and Recall (Rec), demonstrating the effectiveness of this method in predicting communication frequencies in different migration scenarios.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Communication Frequency Adaptive Selection Method Based on Electric Field Simulation and Joint Distribution Adaptation

  • 期刊论文

  • 发表日期：

    2025年8月29日

查看全文
  • Peisheng Liu,
  • Tansheng Chen,
  • Jianan Wu,
  • Jinhua Shan

摘要

Fast and stable communication in complex aquatic environments is essential for the coordinated operation of underwater robotic clusters. Inspired by the communication mechanisms of weakly electric fish, underwater electrocommunication has emerged as a research hotspot due to its low power consumption, excellent omnidirectionality, and stable transmission. However, when electrical signals are used for communication, the frequency increases and the attenuation is intensified, resulting in shorter transmission distances and different attenuation degrees in different underwater environments. Though lowering the frequency reduces attenuation, it limits the communication rate. Therefore, when the communication distance is fixed, the rational selection of communication frequency can effectively enhance the transmission rate between underwater communication nodes. This study proposes a communication frequency adaptive selection method based on electric field simulation and Joint Distribution Adaptation (JDA) to enable the transmitting electrodes to select the optimal communication frequency by the variations in the underwater environment and communication distance, thereby transmitting the information to the receiving electrodes at the highest rate. Firstly, a three-layer medium simulation model is constructed based on the underwater electric field communication theory. Then, the degree of influence of various factors in different water environments on communication frequency and transmission distance is obtained by analyzing the simulation data. It is finally determined by taking the conductivity, water depth, communication distance, spacing between transmitting electrodes, and deployment depth as the model input parameters. Secondly, an air–seawater–soil model is employed to generate the source domain dataset, while four typical types of marine terrain—plain, concave, convex, and slope—are used to construct the target domain dataset. JDA is applied to transfer knowledge from the source domain to the target domain, thereby enhancing the model’s generalization capability. Finally, the proposed method is verified through multiple evaluation metrics, such as Accuracy (Acc) and Recall (Rec), demonstrating the effectiveness of this method in predicting communication frequencies in different migration scenarios.