Utilizing Neural Network To Determine Rational Time Of GPS Field Observation

At present,we confirm the time of GPS observation on the ground of the standards of GPS observation. In a certain degree,the technique was restricted in more fields though the standards have had tremendous effect on all kinds of GPS observation. Especially long time for observation will directly lead to reduce work efficiency. In order to raise efficiency in practice,the paper thoroughly studys the time of GPS field observation by utilizing neural network. Main task and conclusion of the paper are as follows:(1)The basic principles of relative positioning by GPS carrier wave are expounded in the paper. The solution to the double difference carrier wave observation equation is detailedly proved. On the basis of he basic principles, the errors from three quarters and some influences on orientation are analyzed.(2)The basic theory of neural network and an algorithm of BP are described in the paper. All the formula for BP are deduced. The number of hidden layers, the number of neurons and choice of transfer functions are discussed.(3)On the basis of analyzing different factors that influence the right time of GPS field observation, determining the time of GPS field observation by utilizing neural network is put forward for the first time in the paper. A model of BP network used for determining the time of GPS field observation is established, and is proved by checking data set. The result indicates that the neural network is effectual. In practice the BP network can not only meet precision expectations but also improve work efficiency remarkably .(4)It is put forward that the neural network needs to be perfected.