| 能量最优下的铱星时序信号分析预测 |
| Analysis and Prediction of Iridium Time Series Signal with Optimal Energy |
| 投稿时间：2020-09-15 修订日期：2020-11-21 |
| DOI： |
| 中文关键词: 铱星 滑动窗口 极端梯度提升 分类预测 能量优化 |
| 英文关键词:iridium star sliding window extreme gradient rise classification prediction energy optimization |
| 基金项目:科技部重大科学仪器专项——海洋物性参数监测仪（2018YFF01014100） |
|
| 摘要点击次数: 233 |
| 全文下载次数: 0 |
| 中文摘要: |
| 铱星作为低轨道，传输速率快，信息损耗率小，大范围应用于海洋研究的通信卫星，其信号质量的可靠性对铱星通信终端的数据发送及能量利用率产生重要的影响，故对铱星信号的监测和分析是重中之重。本文在有效监测铱星信号的基础上使用滑动窗口算法对时序信号质量做总体评估，从而方便对数据进行预处理。并利用极端梯度提升算法（eXtreme Gradient Boosting, XGBoost）的学习预测功能对信号进行分类预测，判断信号是否存在干扰以择优数据发送方案。提高铱星通信终端能量的利用率，减少数据发送的丢失率。结果表明：预测模型AUC（area under the curve of ROC）面积为0.74-0.78；预测精确率为0.76-.082；该预测模型可有效实现铱星通信终端设备的能量优化。 |
| 英文摘要: |
| Iridium satellite, as a low orbit communication satellite with high transmission rate and low information loss rate, is widely used in marine research. The reliability of its signal quality has an important impact on the data transmission and energy utilization rate of Iridium satellite communication terminal, so the monitoring and analysis of Iridium satellite signal is the top priority. In this paper, based on the effective monitoring of iridium signal, the sliding window algorithm is used to evaluate the quality of timing signal, so as to facilitate the data preprocessing. The learning and prediction function of extreme gradient boosting (XGBoost) is used to classify and predict the signals and judge whether there is interference in the signals to select the optimal data transmission scheme. Improve the energy utilization of iridium communication terminal and reduce the loss rate of data transmission. The results show that the area of AUC (area under the curve of ROC) is 0.74-0.78; the prediction accuracy is 0.76-082; the prediction model can effectively realize the energy optimization of iridium communication terminal equipment. |
| 查看/发表评论 下载PDF阅读器 |
