Satellite network traffic is spatially and temporally dependent. Traditional prediction models are unable to extract the spatial temporal characteristics of effectively. In this paper, a model combined with graph convolutional neural (GCN) gated recursive unit (GRU) proposed. On basis using GCN learn topology satellite traffic, data used as input GRU time change rule node attributes, then finally make through full connection layer. The simulation experiment shows that compared single model,...

Network traffic forecasting is essential for efficient network management and planning. Accurate long-term models are also proactive control of upcoming congestion events. Due to the complex spatial-temporal dependencies between flows, traditional time series often unable fully extract characteristics flows. To address this issue, we propose a novel dual-channel based graph convolutional (DC-STGCN) model. The proposed model consists two temporal components that characterize daily weekly...

Network traffic prediction is an important tool for the management and control of IoT, timely accurate models play a crucial role in improving IoT service quality. The degree burstiness intelligent network high, which creates problems prediction. To address problem faced by traditional statistical models, cannot effectively extract features when dealing with inadequate sample data, addition to poor interpretability deep this paper proposes model (fusion prior knowledge network) that...

In order to efficiently transmit spatial data with diversified service types, this paper proposes an Advanced Intelligent Framing Algorithm (AIFG) for Orbiting System (AOS), based on the virtual channel multiplexing AOS technology, aiming meet differentiated quality of (QoS) requirements. To enable timely transmission delay-sensitive services, optimal time threshold framing such services is modeled as a Markov decision process (MDP). The algorithm utilizes Proximal Policy Optimization (PPO),...

Timely and accurate network traffic prediction is a necessary means to realize intelligent management control. However, this work still challenging considering the complex temporal spatial dependence between traffic. In terms of dimension, links connect different nodes, flowing through nodes has specific correlation. not only at adjacent time points correlated, but also importance distant necessarily less than nearest point. paper, we propose novel method based on joint attention GCN-GRU...

The rapid development of information technology has promoted the transformation traditional networks into intelligent networks. Huge data traffic is generated by various types services in networks, which can easily lead to network congestion, system instability, and other problems. These problems may incur great requirements pose challenges for queue management algorithms. Most active (AQM) algorithms judge congestion level based only on size average length while ignoring variations. This...

The current Internet of Things (IoT) network structure is evolving from small-scale distributed systems to a large-scale hierarchical collaboration between backbone and access networks. In this context, the dynamic changes in node connections surge service demands, coupled with sluggish fault detection speeds, significantly shorten effective transmission time. To address issue, paper proposes an inverse simulated annealing for enhanced OSPF route convergence IoT networks (OSPF-ICSA)....

Simultaneous use of heterogeneous radio access technologies to increase the performance real-time, reliability and capacity is an inherent feature satellite-5G integrated network (Sat5G). However, there still a l... | Find, read cite all research you need on Tech Science Press

With the rapid growth of encrypted network traffic, identification to it becomes a hot topic in information security. Since existing methods have difficulties identifying application which traffic belongs to, new scheme is proposed this paper. The has two levels. In first level, entropy and estimation Monte Carlo π value as features are used identify by C4.5 decision tree. second types distinguished from selected above. First, variational automatic encoder extract layer features, combined...

In order to efficiently transmit spatial data with diversified service types, based on the virtual channel multiplexing technology of Advanced Orbiting Systems ( AOS ), this paper proposes an Intelligent Frame Algorithm in Systems( AIFG ) for differentiated QoS requirements. To expedite transmission delay-sensitive services, optimal time threshold framing such services is modeled as a Markov Decision Process. The algorithm utilizes Proximal Policy Optimization (PPO), achieving solution...

ECG can accurately reflect the activity process of human heart. Through monitoring, users detect various types cardiac diseases as early possible and take effective countermeasures in time. However, problems such noise interference algorithm defects seriously affect accuracy detection. In this paper, aiming at eliminating signal superimposed signal, an improved filter is designed by combining second-order shaped IIR low-pass high-pass Butterworth filter, a better filtering effect achieved....

Abstract In the integrative intelligent network, burst and heterogeneous characteristics of aggregation node traffic lead to increased queuing delay frame loss rate business flow. It is difficult meet quality service (QoS) requirements differentiated response above problems, a mixed queue scheduling based on demand (MQSD) proposed for integrated network heaven earth. Different algorithms are used schedule different flows, which distinguishes needs business. The weight flow calculated using...

Abstract Aiming at the problems of deterministic transmission delay-sensitive service flows and mixed different in integrative intelligent network, which can easily lead to network congestion collapse excessive delay, a scheduling mechanism named SATAS based on information age is proposed. The aggregation node are classified into other types flows. Based idea time division multiplexing, flow buffer queue scheduled according gated list. Further, combined with information, multi-objective...

Buffer sizing for switching and routing devices is of significance guaranteeing the Quality Service (QoS) critical services on Internet Things (IoT), continuously evolving scheduling mechanisms complex traffic characteristics pose new challenges traditional method static buffer based rule-of-thumb. In this paper, scope extended from a basic system under homogeneous arrival input to an integrated heterogeneous which more ubiquitous. context, Voices, videos other data in IoT are categorized...

Trying to overcome the obstacles, such as low throughput and high SER or BER in traditional satellite systems, PNC was implied Satellite System established this paper with efficiency reliability. The performances of AF-PNC, DF-PNC DNF-PNC were investigated simulations. With simulations results, an adaptive strategy-Adaptive proposed according real channel situations. simulation showed that Adaptive had better adaptability lower than single method. Our work reveals are worthy consideration...

In wireless sensor networks (WSNs), base stations (BS) or sink nodes (SNs) play a crucial role in receiving, processing, and transmitting data collected from sensors. Therefore, their traffic prediction performance significantly impacts the efficiency of transmission. While deep learning (DL) models have been widely utilized WSNs due to impressive feature extraction capabilities, they still face challenges related sample size dependency interpretability. Motivated by these challenges, this...

There are many types of spatial data link layer protocols and different frame structures. To achieve effective framing resolution the protocol, we must first solve problem identifying synchronization words. In this paper ,a word recognition algorithm for protocol based on N-ECLAT is proposed. The combines structure characteristics vertical mining Eclat algorithm. By introducing parameter N, can be accurately identified, which lays foundation identification CCSDS protocol. Simulation results...

For the spatial information network link load imbalance, task of differentiated service quality requirements is difficult to meet. This paper proposes a level agreement, which includes and priority avoidance strategies required by task, provides basis for resource allocation requirements. On this basis, space station designed perform edge calculation on flow that about enter side, identify mark level, provide technical support task-driven network-wide scheduling. The simulation results show...

Considering the problem that SDN controller can not fully resolve virtual network function during Software Defined Network (SDN) and Function Virtualization (NVF) collaborative deployment in satellite network, a SDN/NFV coordination based on three-layer is designed. The framework extends proposes extension method (SDN-FE) LightGBM algorithm. SDN-FE has extended functions of original three aspects. One to establish classification model for intelligent forwarding data packets. second learn...