论文部分内容阅读
根据高精度卫星导航定位和全球电离层活动监测的需要,利用全球370多个GPS基准站的双频相位实测数据,监测全球电离层总电子含量变化和GPS卫星及接收机的DCB。由于数据量大、数据处理时间长,很难实现高精度快速建模,为此我们采用OpenMP并行算法来加快数据处理速度。实验表明,相对于串行处理,并行处理在8核服务器下能加速7倍以上,在48核服务器下能加速超过40倍。将本文的初步建模结果与CODE、JPL等分析中心的结果进行比较,表明用该方法建立的模型是可靠的。其卫星DCB结果相对于CODE发布的结果精度为0.4ns,相对于JPL发布的结果其精度达到0.3 ns。其测站DCB相对于2个分析中心结果的精度均优于2 ns,垂直总电子含量相对于各分析中心的GIM产品的精度都在5.3TECU以内,相对于CODE的结果的精度最高,达4 TECU。
According to the needs of high-precision satellite navigation and global monitoring of ionospheric activities, global changes of ionospheric total electronic content and GPS satellite and receiver DCB are monitored by the dual-frequency phase measurement data of more than 370 GPS stations in the world. Due to the large amount of data and long data processing time, it is difficult to achieve high-precision and rapid modeling. Therefore, we use OpenMP parallel algorithm to speed up data processing. Experiments show that compared to serial processing, parallel processing can be accelerated more than 7 times under 8-core server and more than 40 times under 48-core server. The preliminary modeling results of this paper are compared with the results of CODE, JPL and other analysis centers, which shows that the model established by this method is reliable. The accuracy of its satellite DCB results relative to the CODE release is 0.4 ns, which is 0.3 ns relative to the results published by JPL. The accuracy of the station DCB relative to the results of two analysis centers is better than 2 ns. The accuracy of the vertical total electronic content relative to the GIM products of each analysis center is within 5.3 TECU, and the accuracy relative to the result of CODE is the highest, reaching 4 TECU.