Based on the traditional fuzzy arithmetic and concerned about the influence of carrier-to-noise ratio (C/N0) to positioning accuracy, this paper proposes an improved satellite selection algorithm based on C/N0 and geometric dilution of precision (GDOP). Positioning accuracy is mainly influenced by GDOP and the error of pseudo range measurement, and C/N0 is one of the primer factors influence the latter one. Both the error of pseudo range measurement and GDOP should be concerned in order to decrease the positioning error, instead of separate consideration. Traditional satellite selection algorithms, such as the algorithm of minimum GDOP, the algorithm of max volume of tetrahedron and so on, they just concentrated on the influence of GDOP and ignore the error of pseudo range measurement. This paper chooses the first two satellites with the highest and lowest elevation angle based on the principle of minimum GDOP, and then consider C/N0 and the satellite geometric distribution both when choosing the others. Use fuzzy satellite selection algorithm based on entropy method to weight the two factors. Pick out the satellite combination with smaller GDOP and higher C/N0 in the end. Compare and analyze the two results separately get from the method proposed in this paper and the traditional one, the simulation result shows that the improved satellite selection algorithm that considered GDOP and C/N0 both, compared to the traditional ones, can lead to smaller positioning error and better positioning result.
