钟差改正数预报效果评估

Prediction effect evaluation of clock correction

针对基于BDS-SMC的远海高精度动态定位中存在的数据中断或通讯延迟现象,从稳定性与精度两方面对GPS、GLONASS、Galileo与BDS-3的钟差改正数进行质量评估,并分析其数据结构特征。结果表明,在中断区间较短且IOD不变的情况下,由于钟差改正数的数量级保持在纳秒级,钟差改正数的预报误差低于钟差数据,故钟差改正数预报可有效应对该现象。针对BDS-SMC通讯成本高昂的问题,本文提出利用钟差改正数预报进行数据压缩,仅利用BDS-SMC传输模型参数可显著降低数据量。以原始数据序列与一次差分序列为训练样本,选取结构简单、参数较少的3种模型进行预报效果的评估。结果表明,由于预报区间较短,差分序列预报模型的误差累积效应不明显,预报精度高于原始序列预报模型。

According to the fact of data interruption or communication delay in high-precision dynamic positioning in the ocean based on BDS-SMC,this paper evaluates the quality of clock correction of GPS,GLONASS,Galileo and BDS-3 from two aspects of stability and accuracy,and analyzes the data structure characteristics.The results show that under the condition of short interruption interval and constant iod,since the order of magnitude of the clock correction remains at ns level,the prediction error of the clock correction is lower than that of the clock bias data,so the clock correction prediction can effectively deal with this phenomenon.According to the high cost of BDS-SMC communication,this paper proposes to use clock correction prediction for data compression,and only using BDS-SMC transmission model parameters can significantly reduce the amount of data.Taking the original data sequence and the primary difference sequence as training samples,three models with simple structure and few parameters are selected to evaluate the prediction effect.The results show that due to the short prediction interval,the error accumulation effect of the differential sequence prediction model is not obvious,and the prediction accuracy is higher than that of the original sequence prediction model.