时间敏感网络中多级循环队列与转发调度方法和系统数据集

时间敏感网络中多级循环队列与转发调度方法和系统数据集，该项数据由北京邮电大学提供，在北京邮电大学采集数据，数据采集的时间为2021年5月与2022年12月。该项数据集包括专利“时间敏感网络中多级循环队列与转发调度方法和系统”中关键支撑数据。数据采集的内容主要包括时间敏感网络中多级循环队列与转发调度方法和系统在测试实验时产生的数据。数据具体包括多级循环队列与转发调度机制数据和基于幂方分解的数据包分割算法实验数据。多级循环队列与转发调度机制数据包括多级循环队列与转发调度机制和循环队列与转发调度机制在不同的流需求总数和不同的流需求比例下调度成功率差异。本实验基于Python3.8定义网络需求和调度模型，由random_regular_graph生成设置具有16个节点的随机正则图，链路传输速率均为1000Mbps，各链路带宽时延均等。参考IEC/IEEE 60802工业自动化网络标准中描述的流量特征生成单播周期流，数据帧大小取MTU，周期组取值为12us整数倍。仿真获取在不同的流需求总数和不同的流需求比例下调度成功率。基于幂方分解的数据包分割算法实验数据包括数据包不分割方案、定长分割方案、基于幂方分解的分割方案的仿真数据。本实验假设数据流服从泊松分布，数据包大小服从64字节到64K字节的均匀分布。首先使用MATLAB程序生成输入数据包到达时间与长度信息，然后调用分割算法获得信元大小与产生时间，接着调用Buddy System仿真程序分配内存空间，获得内存分配状态。最后使用MATLAB程序分析实验结果，计算得到内存利用效率，数据量66KB。

Dataset for Multi-level Cyclic Queue and Forwarding Scheduling Method and System in Time-Sensitive Networks (TSN)
This dataset is provided and collected by Beijing University of Posts and Telecommunications (BUPT), with data acquisition conducted in May 2021 and December 2022. It includes the key supporting data from the patent titled "Multi-level Cyclic Queue and Forwarding Scheduling Method and System in Time-Sensitive Networks". The collected data mainly covers the data generated during the test experiments of the multi-level cyclic queue and forwarding scheduling method and system in time-sensitive networks, specifically including two categories: data of the multi-level cyclic queue and forwarding scheduling mechanism, and experimental data of the packet segmentation algorithm based on power decomposition.

The data of the multi-level cyclic queue and forwarding scheduling mechanism covers the differences in scheduling success rates between the multi-level cyclic queue and forwarding scheduling mechanism and the conventional cyclic queue and forwarding scheduling mechanism under different total flow demands and different flow demand ratios. This experiment defines the network demand and scheduling model based on Python 3.8, uses the `random_regular_graph` function to generate a random regular graph with 16 nodes, sets the link transmission rate to 1000 Mbps for all links, and ensures equal bandwidth and delay for each link. Unicast periodic flows are generated referring to the traffic characteristics described in the IEC/IEEE 60802 industrial automation network standard, where the data frame size is set to MTU, and the period group takes an integer multiple of 12 μs. The scheduling success rates under different total flow demands and different flow demand ratios are obtained through simulation.

The experimental data of the packet segmentation algorithm based on power decomposition includes simulation data of three schemes: packet non-segmentation, fixed-length segmentation, and segmentation based on power decomposition. This experiment assumes that the data flow follows a Poisson distribution, and the packet size follows a uniform distribution ranging from 64 bytes to 64 KB. First, the MATLAB program is used to generate the arrival time and length information of input packets, then the segmentation algorithm is called to obtain the cell size and generation time, followed by calling the Buddy System simulation program to allocate memory space and obtain the memory allocation status. Finally, the MATLAB program is used to analyze the experimental results and calculate the memory utilization efficiency, with a total data volume of 66 KB.

搜集汇总

背景与挑战

背景概述

该数据集聚焦于时间敏感网络（TSN）中的多级循环队列与转发调度方法，由北京邮电大学于2021年5月和2022年12月采集，包含专利“时间敏感网络中多级循环队列与转发调度方法和系统”的关键实验支撑数据。数据集主要包括两部分：一是多级循环队列与转发调度机制在不同流需求条件下的调度成功率仿真数据，基于Python3.8构建网络模型；二是基于幂方分解的数据包分割算法仿真数据，用于比较不同分割方案的内存利用效率，实验通过MATLAB和仿真程序完成。数据集规模较小（75.19KB），以xlsx和docx格式提供，适用于TSN调度优化和算法性能分析研究。