Due to the fading characteristics of wireless channels and the burstiness of data traffic,how to deal with congestion in Ad-hoc networks with effective algorithms is still open and challenging.In this paper,we focus o...Due to the fading characteristics of wireless channels and the burstiness of data traffic,how to deal with congestion in Ad-hoc networks with effective algorithms is still open and challenging.In this paper,we focus on enabling congestion control to minimize network transmission delays through flexible power control.To effectively solve the congestion problem,we propose a distributed cross-layer scheduling algorithm,which is empowered by graph-based multi-agent deep reinforcement learning.The transmit power is adaptively adjusted in real-time by our algorithm based only on local information(i.e.,channel state information and queue length)and local communication(i.e.,information exchanged with neighbors).Moreover,the training complexity of the algorithm is low due to the regional cooperation based on the graph attention network.In the evaluation,we show that our algorithm can reduce the transmission delay of data flow under severe signal interference and drastically changing channel states,and demonstrate the adaptability and stability in different topologies.The method is general and can be extended to various types of topologies.展开更多
Several new demands have been put forward for the application of the Beijing continuous GNSS observations due to some particular reasons, such as the limited coverage of the observation network, the different construc...Several new demands have been put forward for the application of the Beijing continuous GNSS observations due to some particular reasons, such as the limited coverage of the observation network, the different construction and management criterion executed by different units and the intense interference resulting from human activity. In this paper, necessary processing of data is carried out, including more accurate calculation, corrections to the replacement, outliers and relocation of equipment, and elimination of linear trends in the E-component for every station. The E-components of the 16 available stations showed a lower sawtooth wave anomaly （slowly westward propagating） before the 2011 Tohoku Mw9. 0 earthquake, a coseismic step rebound （rapid eastward propagating） and a post-seismic D-shaped recovery. These steps constituted a complete earthquake process which was rarely seen before in the GNSS observations and provides a good example for further study. Moreover, the rapid eastward propagating during the earthquake is not influenced by the size of the given normal values, which may play a significant role in earthquake forecasting and early warning.展开更多
基金supported by Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government(MSIT) (No.RS-2022-00155885, Artificial Intelligence Convergence Innovation Human Resources Development (Hanyang University ERICA))supported by the National Natural Science Foundation of China under Grant No. 61971264the National Natural Science Foundation of China/Research Grants Council Collaborative Research Scheme under Grant No. 62261160390
文摘Due to the fading characteristics of wireless channels and the burstiness of data traffic,how to deal with congestion in Ad-hoc networks with effective algorithms is still open and challenging.In this paper,we focus on enabling congestion control to minimize network transmission delays through flexible power control.To effectively solve the congestion problem,we propose a distributed cross-layer scheduling algorithm,which is empowered by graph-based multi-agent deep reinforcement learning.The transmit power is adaptively adjusted in real-time by our algorithm based only on local information(i.e.,channel state information and queue length)and local communication(i.e.,information exchanged with neighbors).Moreover,the training complexity of the algorithm is low due to the regional cooperation based on the graph attention network.In the evaluation,we show that our algorithm can reduce the transmission delay of data flow under severe signal interference and drastically changing channel states,and demonstrate the adaptability and stability in different topologies.The method is general and can be extended to various types of topologies.
基金funded by Research Foundation for Veteran Experts of China Earthquake Administration(201346)the Natural Science Foundation of Beijing Municipality(8041001,8092012)
文摘Several new demands have been put forward for the application of the Beijing continuous GNSS observations due to some particular reasons, such as the limited coverage of the observation network, the different construction and management criterion executed by different units and the intense interference resulting from human activity. In this paper, necessary processing of data is carried out, including more accurate calculation, corrections to the replacement, outliers and relocation of equipment, and elimination of linear trends in the E-component for every station. The E-components of the 16 available stations showed a lower sawtooth wave anomaly （slowly westward propagating） before the 2011 Tohoku Mw9. 0 earthquake, a coseismic step rebound （rapid eastward propagating） and a post-seismic D-shaped recovery. These steps constituted a complete earthquake process which was rarely seen before in the GNSS observations and provides a good example for further study. Moreover, the rapid eastward propagating during the earthquake is not influenced by the size of the given normal values, which may play a significant role in earthquake forecasting and early warning.
