چكيده به لاتين
Nowadays, the use of Global Satellite Navigation Systems (GNSS) is expanding in various applications on land, air, sea and space. GNSS receivers use two types of techniques for satellite signal tracking: one based on Scalar Tracking (ST) and the other on Vector Tracking (VT). In the VT method, unlike the ST, satellite information is processed centrally. The VT method has many advantages over the ST method, but its main drawback is its high data processing volume which has made it difficult to apply it to GNSS receivers. In this thesis, by considering different structures of VT methods, we choose a low-computation method and then proposes three methods to reduce the computational burden by examining different parts of the VT method. In the first method, with the satellite orbit modeling, it is possible to reduce the computational burden to a high proportion. The analytical results along with the results of the experiments show that the proposed models for modeling the navigation satellite orbit can reduce more than 90% of the computations compared to the conventional methods of satellite positioning and speed. In the second method, by selecting the optimal combination of satellites, it is possible to reduce the computational burden of other methods by more than 20% in multiplication and more than 56% in addition and subtraction operations. Also, in the third method to reduce the computational burden of Kalman filter, three techniques are proposed which can improve over 27% in multiplication calculations and more than 55% in addition and subtraction calculations if using the first two techniques. If the third technique is used together with the first and second techniques, it can improve up to 90% in multiplication and addition calculations. Overall, by applying the three methods proposed in this thesis, the computational burden of the VT method has been greatly reduced and will be easier to implement on GNSS receivers.
