Fine Doppler frequency estimation in GNSS signal acquisition process

It is known that signal acquisition can be treated as a two-dimensional search and all the possible results are located in the search plane as different cells. In this paper, the theoretical expression of the cross-ambiguity function (CAF) is analyzed, and the property of the approximate expression is exploited, while it is seldom addressed in previous GNSS receiver architectures. Based on the property of the approximate CAF expression, a novel approach is proposed. In this approach, the information in the region of main peak is used to refine the Doppler frequency in the form of a simple interpolation equation. According to the results, in the absence of noise, the frequency estimation error could be theoretically decreased to zero. In the presence of noise, the novel method is equivalent to the least-square solution in terms of accuracy, but it has a significantly lower implementation complexity. Last, an averaging method is proposed to reduce the influence of noise. It shows that averaging operation can decrease the error to (-10Hz, 10Hz) in 90% of the cases when signal-to-noise ratio is C/NO=43dBHz.