基于GWO-CRF的链路质量预测  

Link quality prediction based on GWO-CRF

作　　者：肖智勇 刘琼[1] XIAO Zhiyong;LIU Qiong(School of Information Science and Engineering,Wuhan University of Science and Technology,Wuhan 430081,China)

机构地区：[1]武汉科技大学信息科学与工程学院,湖北武汉430081

出　　处：《现代电子技术》2022年第14期164-168,共5页Modern Electronics Technique

基　　金：国家自然科学基金青年科学基金项目(61701355)。

摘　　要：针对环境变化导致的链路质量预测结果波动大、预测精度低的问题,文中提出一种基于GWO-CRF的链路质量预测模型。首先,在两种场景下采集链路质量参数并进行预处理,处理后进行数据相关性分析;然后,选择接收信号强度指示和链路质量指示作为预测模型的输入参数,利用条件随机场算法来克服预测结果波动大的问题,从而构建链路质量预测模型。为了提高预测精度,采用灰狼算法对条件随机场算法的参数进行优化,最终得到链路质量的预测结果,并利用两种场景下采集的数据对模型进行有效性验证。结果表明,与支持向量回归、线性回归、指数加权移动平均算法相比,基于GWOCRF算法的预测误差降低2.8%~96.8%。In allusion to the large fluctuations of link quality prediction results and low prediction accuracy caused by environmental changes,a link quality prediction model based on GWO-CRF(gray wolf optimizer-conditional random field)is proposed.The link quality parameters in two scenarios are collected and preprocessed,and then the correlation analysis is performed on the processed data.The received signal strength indicators and link quality indicators are selected as input parameters of the prediction model,and the conditional random field algorithm is used to overcome the fluctuation of the prediction results,so that the link quality prediction model is constructed.In order to improve the prediction accuracy,the grey wolf algorithm is used to optimize the parameters of CRF algorithm to obtain the prediction results of link quality,and the data collected from two scenarios are used to verify the model.The results show that in comparison with the SVR(support vector regression),LR(linear regression)and EWMA(exponentially weighted moving average)algorithm,the prediction error based on GWO-CRF algorithm is reduced by 2.8%~96.8%.

关 键 词：链路质量预测 接收信号强度 条件随机场 灰狼优化算法 模型验证 参数优化 

分 类 号：TN911-34[电子电信—通信与信息系统] TP391.4[电子电信—信息与通信工程]