摘要
在分析北斗一号(BD1)定位特点的基础上,通过在西安丰镐东路实测大量数据,观测到BD1定位高程误差较大,难以满足用户需求。针对此类问题,采用BP神经网络的预测结果对BD1定位结果进行修正,以提高定位的精度。确定了神经网络的结构,采用实测数据对网络进行训练,并对神经网络模型进行仿真验证。仿真结果表明,该方法能有效解决BD1定位高程误差较大的问题,具有一定的实用价值。
In analysis of that BD1 positioning character and lots of measure data in fenghao east road in Xi′an area,BD1 has larger height error than people′s needs.So the BP neural network is applied to position forecast to improve the BD1 accuracy of positioning information.Then,the structure of neural network is confirmed,and the network is trained by the measured data,and the network model is also verified by simulation.The result shows that the method can resolve the problem effectively with high positioning accuracy,and it also has practical value.
出处
《电视技术》
北大核心
2010年第S1期165-168,共4页
Video Engineering
基金
陕西省科学技术研究发展计划项目(2007K04-11)
