In this paper, a distributed muting strategy based on simplified topology （DRBST） was proposed for LEO satellite networks. The topology of LEO satellite networks was simplified aiming at minimizing intersatellite li...In this paper, a distributed muting strategy based on simplified topology （DRBST） was proposed for LEO satellite networks. The topology of LEO satellite networks was simplified aiming at minimizing intersatellite links handover number. To optimize the route based on the simplified topology, we considered not only the transmission delay but also the queuing delay and the processing delay, which were analyzed using Markov chain and determined using a novel methodology. The DRBST algorithm was simulated in a LEO satellite networks model built using OPNET. The simulation results demonstrate that the low complexity DRBST algorithm can guarantee end-to-end delay bound. Moreover, the muting protocol cost is much less than traditional algorithms.展开更多
针对低轨道卫星信道质量变化迅速、信道参数“过时”的问题,提出了一种基于注意力机制的卷积神经和双向长短时记忆神经网络(attention-convolutional neural network and bi-directional long-short term memory neural network,AT-CNN-...针对低轨道卫星信道质量变化迅速、信道参数“过时”的问题,提出了一种基于注意力机制的卷积神经和双向长短时记忆神经网络(attention-convolutional neural network and bi-directional long-short term memory neural network,AT-CNN-BiLSTM)融合的信道预测方法。该方法由信号预处理、网络训练和信号预测3部分组成。首先在高斯白噪声条件下模拟室外卫星信号,得到卫星信号的训练集和测试集;然后将训练集输入构建的训练网络进行特征提取;最后将测试数据输入网络进行预测分析。仿真结果表明,在与其他4种人工智能方法的对比中,所提出的混合神经网络能够在较快的收敛速度下达到较高的准确率(91.8%),有效地缓解了低轨道卫星信道参数“过时”的现状,对提升卫星通信质量和节省卫星信道资源有良好的改善作用。展开更多
基金Supported by the National Science Foundation of China （No. 60873219）.
文摘In this paper, a distributed muting strategy based on simplified topology （DRBST） was proposed for LEO satellite networks. The topology of LEO satellite networks was simplified aiming at minimizing intersatellite links handover number. To optimize the route based on the simplified topology, we considered not only the transmission delay but also the queuing delay and the processing delay, which were analyzed using Markov chain and determined using a novel methodology. The DRBST algorithm was simulated in a LEO satellite networks model built using OPNET. The simulation results demonstrate that the low complexity DRBST algorithm can guarantee end-to-end delay bound. Moreover, the muting protocol cost is much less than traditional algorithms.
文摘针对低轨道卫星信道质量变化迅速、信道参数“过时”的问题,提出了一种基于注意力机制的卷积神经和双向长短时记忆神经网络(attention-convolutional neural network and bi-directional long-short term memory neural network,AT-CNN-BiLSTM)融合的信道预测方法。该方法由信号预处理、网络训练和信号预测3部分组成。首先在高斯白噪声条件下模拟室外卫星信号,得到卫星信号的训练集和测试集;然后将训练集输入构建的训练网络进行特征提取;最后将测试数据输入网络进行预测分析。仿真结果表明,在与其他4种人工智能方法的对比中,所提出的混合神经网络能够在较快的收敛速度下达到较高的准确率(91.8%),有效地缓解了低轨道卫星信道参数“过时”的现状,对提升卫星通信质量和节省卫星信道资源有良好的改善作用。
