In the optical fiber time transfer system, fiber optic time transfer is limited by the pulse signal time delay measurement precision, and cannot benefit from a higher precision of the carrier phase information. Its transfer precision compared to the frequency transfer exists in a larger gap transfer. This paper proposes a time delay measurement method based on carrier phase and pseudo-code ranging for optical time transfer. The time signal is modulated with pseudo-random code and carrier at the transmitter, and the time delay is measured at the receiver by the methods of pseudo-code ranging and carrier phase measurement. The time transfer is achieved by eliminating the transmission link delay through a two-way method. The first-order difference value of the carrier phase measurement and the pseudo-code ranging measurement are used as the observation quantities, and they are fused through a Kalman filtering method to finally obtain the high-precision time difference measurement. We validate the theory on the common-clock experimental platform over a 50 km fiber link, The time transfer stabilities of the systems are $5.2254\times {10}^{-14}$/s and $2.146\times {10}^{-16}$/${10}^4$ s (modified Allan deviation), $3.0169\times {10}^{-14}$/s and $1.2392\times {10}^{-12}$/${10}^4$ s (time deviation). The standard deviation of the time transfer system after fusion can reach 2.4255 ps.

In the positioning process of GPS, the linear least squares algorithm and Kalman filtering algorithm are widely used but still have shortcomings. Application of extreme learning machine in this area is proposed in this paper, which breaks through the limitations of the traditional method of positioning based on mathematical models. Two simulation experiments of ELM in GPS positioning process are presented in this paper while the latter is a supplement to the former. Each one contains three phases, including simulation data generation, network training and network prediction, each of which is considered carefully. The feasibility of extreme learning machine is verified through experimental simulation. A more accurate positioning result can be obtained.