链路逐跳时延与突发状态数据

《链路逐跳时延与突发状态数据集》：数据基于带内网络遥测机制对网络状态进行采集。数据预处理由研究团队使用Python编程语言完成，结合Pandas、NumPy等工具对原始遥测数据进行结构化、归一化及异常值识别。数据用于验证基于神经网络的网络状态智慧预测机制的有效性，评估该机制在网络突发流量预测、资源利用率监测等方面的性能，为多域融合的移动性控制机制和确定性达成技术提供数据支持。测试模拟了网络在不同负载和流量情况下的运行场景，包括正常流量场景、突发流量场景等，以检验模型在不同场景下的预测准确性和响应及时性。每类场景在统一网络拓扑结构与软件平台下进行多轮实验采样，确保数据具有代表性与可对比性。测试过程中，仿真平台统一调用数据采集、预处理、模型训练与预测等关键组件。本数据集的数据类型为：文件，数据量为：40KB，共享方式为：完全共享，不设置保护期。

Link-by-link Delay and Burst State Dataset
This dataset is collected through in-band network telemetry (INT) for network state monitoring. Data preprocessing was performed by the research team using the Python programming language, with tools including Pandas and NumPy to structure, normalize, and detect outliers from the raw telemetry data.
This dataset is employed to validate the efficacy of neural network-based intelligent network state prediction mechanisms, assess their performance in scenarios such as network burst traffic prediction and resource utilization monitoring, and offer data support for multi-domain integrated mobility control mechanisms and technologies for realizing deterministic network performance.
The test simulates network operating scenarios under varying loads and traffic patterns, including normal traffic and burst traffic scenarios, to examine the prediction accuracy and response timeliness of the model across diverse scenarios. Each scenario undergoes multiple rounds of experimental sampling under a unified network topology and software platform, ensuring the dataset is representative and comparable. During the test, the simulation platform uniformly invokes core components including data collection, preprocessing, model training and prediction.
The data type of this dataset is file, with a total size of 40 KB, and it is fully shared with no protection period set.