Large-scale post-disaster user distributed coverage optimization based on multi-agent reinforcement learning by Wenjun XU, Silei WU, Fengyu WANG, Lan LIN, Guojun LI, Zhi ZHANG

“…In order to quickly restore emergency communication services for large-scale post-disaster users, a distributed intellicise coverage optimization architecture based on multi-agent reinforcement learning (RL) was proposed, which could address the significant differences and dynamics of communication services caused by a large number of access users, and the difficulty of expansion caused by centralized algorithms.Specifically, a distributed k-sums clustering algorithm considering service differences of users was designed in the network characterization layer, which could make each unmanned aerial vehicle base station (UAV-BS) adjust the local networking natively and simply, and obtain states of cluster center for multi-agent RL.In the trajectory control layer, multi-agent soft actor critic (MASAC) with distributed-training-distributed-execution structure was designed for UAV-BS to control trajectory as intelligent nodes.Furthermore, ensemble learning and curriculum learning were integrated to improve the stability and convergence speed of training process.The simulation results show that the proposed distributed k-sums algorithm is superior to the k-means in terms of average load efficiency and clustering balance, and MASAC based trajectory control algorithm can effectively reduce communication interruptions and improve the spectrum efficiency, which outperforms the existing RL algorithms.…”
Get full text

Highway Travel Time Prediction Using Sparse Tensor Completion Tactics and K-Nearest Neighbor Pattern Matching Method by Jiandong Zhao, Yuan Gao, Jinjin Tang, Lingxi Zhu, Jiaqi Ma

“…Aiming at the data sparseness problem, virtual sensor nodes are set up between real RTMS nodes, and the two-dimensional linear interpolation and piecewise method are applied to estimate the average travel time between two nodes. …”
Get full text

Processing browning and increasing multiplication and rotting of Peach rootstock cv. Garnem by using antioxidants and plant growth regulators in vitro by Sarah Ali Muhammad Al-Hayany, Ayad Assi Obaid Obaid

“…L-1 to the cultivated nodes gave the highest number of leaves. The best rooting percent was obtained when plantlets were s treated with 2 mg. …”
Get full text

Clustering with dynamic route adjustment and fuzzy based update cycle in wireless sensor networks with sink mobility by Asif Khan, Mohammad Amjad

“…Next Update cycle is decided with the help of residual energy of nodes, average sensed data rate and the distance of cluster head from the mobile sink. …”
Get full text

Thermal regime in the troposphere, stratosphere, and mesosphere of the northern polar zone in 1979–2019 by Yu.P. Perevedentsev, K.M. Shantalinsky, N.V. Ismagilov, V.V. Guryanov, A.A. Nikolaev, T.R. Auhadeev

“…N) during 1979–2019 were studied using the data on average monthly air temperatures at the nodes of a regular geographical grid of 1? …”
Get full text

Energy Consumption Analysis of Iot-Manet Based Systems with Anova Assessment for Tackling Covid-19 by Ashu Gautam, Rashima Mahajan, Sherin Zafar

“…In this investigation work, most appropriate routing protocols to knob DDoS attacks are simulated using NS-2 and assessed in terms of average energy consumption in the state of changing speed connections among devices called mesh nodes. …”
Get full text

Facial Expression Recognition with Multi-perspective Feature Fusion Under Deep Residual Convolution by GUAN Xiaorui, GAO Lu, SONG Wenbo, LIN Kezheng

“…Then use the RReLu activation function to replace the original ReLu to avoid deactivation of some nodes when the gradient is large.Finally, the global average pooling layer is used instead of the fully connected layer to achieve dimensionality reduction, and the generated feature vector is sent to Softmax for classification Experiments show that the proposed method produces excellent results on the CK+ and RaFD datasets, which can effectively improve the accuracy of facial expression recognition.…”
Get full text

Influence Maximization in Social Networks Using Discrete Manta-Ray Foraging Optimization Algorithm and Combination of Centrality Criteria by Zaynab Azizpour, Saeid Taghavi Afshord, Bagher Zarei, Mohammad Ali Jabraeil Jamali, Shahin Akbarpour

“…Influence Maximization (IM) is a fundamental problem in social network analysis that seeks to identify a small set of highly influential nodes that can maximize the spread of information. …”
Get full text

Increasing the Genetic Diversity of Patchouli through Gamma Ray Irradiation by Halida Adistya Putri, Lidia Novita Putri, Aline Sisi Handini

“…Data were analyzed using the average value with standard deviation and boxplot. …”
Get full text

Video-assisted thoracoscopic surgery lobectomy – early experience by Robert Šimon, Marián Kudláč, Jozef Belák

“…Introduction: Video-assisted thoracoscopic surgery (VATS) lobectomy is the anatomical resection of a whole lobe ofthe lung followed by removal of the lymph nodes from the mediastinum using a thoracoscope and an access incision(small thoracotomy ≤ 5 cm) without using the rib spreader. …”
Get full text

Study on the Macroeconomic Model Based on the Genetic Algorithm by Sun Shihao, Yixin Zhou, Yong Wang, Wu Wang

“…Because the macro data is one-dimensional array data, the essence of the mutation algorithm is to obtain the movement direction of the mutation of data nodes, obtain the distance between the linear programming result and the original data through the least square method, and calculate the average value in the original data, After binary t-correction, it refers to the binary t-correction results of the one-dimensional matrix before the final evaluation output factor and the one-dimensional matrix after the final evaluation output factor. …”
Get full text

Delay-constrained data aggregation scheduling in wireless sensor networks by Cheng Feng, Zhijun Li, Shouxu Jiang, Weipeng Jing

“…This approach is to maximize the number of scheduled nodes in every time slot from deadline to time slot 1. …”
Get full text

Accelerating flood warnings by 10 hours: the power of river network topology in AI-enhanced flood forecasting by Hongjun Wang, Jiyuan Chen, Yinqiang Zheng, Xuan Song

“…Our analysis reveals that the tree-like configuration of river networks leads to over-squashing in GNNs, a problem caused by high resistance distances between nodes. To address this, we propose a novel method that transforms the topological graph into a dense, reachability-based graph, which reduces resistance distances. …”
Get full text

Large-scale post-disaster user distributed coverage optimization based on multi-agent reinforcement learning by Wenjun XU, Silei WU, Fengyu WANG, Lan LIN, Guojun LI, Zhi ZHANG

“…In order to quickly restore emergency communication services for large-scale post-disaster users, a distributed intellicise coverage optimization architecture based on multi-agent reinforcement learning (RL) was proposed, which could address the significant differences and dynamics of communication services caused by a large number of access users, and the difficulty of expansion caused by centralized algorithms.Specifically, a distributed k-sums clustering algorithm considering service differences of users was designed in the network characterization layer, which could make each unmanned aerial vehicle base station (UAV-BS) adjust the local networking natively and simply, and obtain states of cluster center for multi-agent RL.In the trajectory control layer, multi-agent soft actor critic (MASAC) with distributed-training-distributed-execution structure was designed for UAV-BS to control trajectory as intelligent nodes.Furthermore, ensemble learning and curriculum learning were integrated to improve the stability and convergence speed of training process.The simulation results show that the proposed distributed k-sums algorithm is superior to the k-means in terms of average load efficiency and clustering balance, and MASAC based trajectory control algorithm can effectively reduce communication interruptions and improve the spectrum efficiency, which outperforms the existing RL algorithms.…”
Get full text

Author name disambiguation based on heterogeneous graph neural network. by Ge Wang, Zikai Sun, Weiyang Hu, MengHuan Cai

“…As the existing graph heterogeneous neural network can not learn different types of nodes and edge interaction, add multiple attention, design ablation experiments to verify its impact on the network. …”
Get full text

Internet of things data security and privacy protection based on improved federated learning by Wang Gang

“…At the same time, using the decentralized function to encrypt the privacy of the training model, the learning model can provide more secure and reliable services, aiming to solve the problem of large difference in the quality of computing nodes and data privacy leakage in the current FL. …”
Get full text

Dynamic sink movement strategy for expedited query processing in Internet of things-based sensor networks by Dongliang Xu

“…In IoT-driven systems, sensor nodes collect real-time data, initiate end-user or application requests, and forward the gathered data to a cloud server. …”
Get full text

Inter-Satellite Handover Method Based Multi-Objective Optimization in Satellite-Terrestrial Integrated Network by Renpeng LIU, Bo HU, Hequn LI

“…The high-speed motion of low-earth orbit communication satellites results in a highly dynamic network topology, and the spatio-temporal distribution of resources in the satellite-terrestrial integrated network is non-uniform.When multiple users and services switch between satellites, a large number of handover requests are triggered, leading to intensified network resource competition.As a result, the limited satellite resources cannot meet all the handover requests, leading to a significant decrease in handover success rate.In view of the above problem, the multi-objective optimization based satellite handover method was proposed.It introduced the satellite coverage spatio-temporal graph and transforms the dynamic continuous topology into static discrete snapshots, accurately depicted the connections between satellite nodes and users at different times and locations.The multi-objective optimization model was established for satellite handover decisions, and anadaptive accelerated multi-objective evolutionary algorithm(AAMOEA) was proposed to optimized user data rate and network load simultaneously, ensured handover success rate and enhanced network service capability.It built a STIN communication simulation environment and tested the multi user handover performance in a multi satellite overlapping coverage scenario.The results demonstrated that the multi-objective optimization-based satellite handover method achieved an average handover success rate improvement of over 20% compared to benchmark algorithms.…”
Get full text

Optimized Quality of Service for Real-Time Wireless Sensor Networks Using a Partitioning Multipath Routing Approach by Mohammed Zaki Hasan, Tat-Chee Wan

“…The proposed approach significantly improves the network lifetime, while reducing energy consumption and decreasing average end-to-end delays within the sensor network via optimised resource sharing in intermediate nodes compared with existing routing algorithms.…”
Get full text

Improved A* Algorithm Based on Artificial Potential Field Method for Global Path Planning of Unmanned Surface Vessels in Complex Environments by Shuxuan Liu, Zhijia Qian, Sunwei Jiang, Quanming Miao

“…Aiming at the problems of long path planning time, excessive ineffective expansion nodes, and easy collision with obstacles that may occur when using traditional A* algorithm for unmanned surface vessels in complex environments, an improved A* algorithm combined with artificial potential field thinking is proposed for global path planning of unmanned surface vessels. …”
Get full text