The optimizations of FFT algorithm in GPS software receiver

In a GPS system, the process of acquisition costs most time of the whole, which is due to the large number of required operations in the correlators. Common methods usually use FFT or PFA(Prime Factor Algorithm)to calculate the correlation, choosing float as the data type of the operation, which wil...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Xiangzhen Yao, Xinxian Qin, Shaolong Cui, Jinyun Fang
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In a GPS system, the process of acquisition costs most time of the whole, which is due to the large number of required operations in the correlators. Common methods usually use FFT or PFA(Prime Factor Algorithm)to calculate the correlation, choosing float as the data type of the operation, which will obviously increase computational complexity. In consideration of the specific storage mode of GPS data, we propose a novel algorithm using 2-bit to store the GPS data, and using 2-bit to store the quantization of the local carrier frequency, then, the floating-point calculations has been replaced by bit-wise calculation, and the twiddle factors has been pre-stored using 16-bit 2's complement integer. In this method, FFT will be performed only with integer arithmetic. Experiments show the significant performance and efficiency improvements of the proposed algorithm.
ISSN:2161-024X
DOI:10.1109/GeoInformatics.2011.5981131