This paper presents a systematic discussion on GPS C/A code cross correlation and its impact on signal acquisition, tracking and Ground-Based Augmentation System (GBAS) performance. Three types of cross correlation effects are investigated: 1) between two satellites that have approximately the same Doppler frequency; 2) between two satellites that have an offset at an integer number of kHz in Doppler frequency; 3) between a C/A code signal and a signal that consists of alternating zeros and ones. The first type of cross correlation has been well studied in the past decade, and its impact is often found similar to that of multipath. There exist some subtle, but important, differences between cross correlation and multipath, which will be discussed in this paper. Cross correlation cannot be treated as a random interference source, since it is inherently constrained by the Doppler frequency difference between the two satellites. Based on this constraint, the cross correlation functions will be analytically modeled in the time domain and in the frequency domain for each of the three types. All three types of cross correlation are potential threats to weak signal acquisition, PseudoRange tracking and carrier phase tracking. There are impacts on both mobile users and ground reference stations. More specifically, the PseudoRange tracking error will likely not be common between a ground reference and a mobile user, which becomes a concern for differential systems like GBAS. The PseudoRange error is not only a function of the Doppler offset and signal strength, but is also dependent on the tracking loop configuration. For example, the relative motion between the satellites and the antenna, the tracking loop bandwidth, coherent integration time and the carrier smoothing time constant all play key roles in the PseudoRange error model. It has been discovered in previous studies that a sufficiently large time constant in carrier smoothing provides effective mitigation for cross correlation errors. Most GPS users are protected against cross correlation in signal acquisition and tracking. In some worst-case scenarios, however, meter-level PseudoRange errors in GBAS may occur. As a result, cross correlation must be carefully monitored for high-accuracy safety of life applications. It can also falsely trigger other GBAS monitors, which may include the low power monitor and the signal deformation monitor. An overview of the implications of cross correlation on GBAS is provided in this paper.
[1]
Frank van Graas,et al.
Remote-Controlled, Continuously Operating GPS Anomalous Event Monitor
,
2006
.
[2]
Frank van Graas,et al.
Operational Considerations for C/A Code Tracking Errors Due to Cross Correlation
,
2005
.
[3]
Curtis A. Shively,et al.
Availability of GAST D GBAS Considering Continuity of Airborne Monitors
,
2010
.
[4]
R. Hatch.
The synergism of GPS code and carrier measurements
,
1982
.
[5]
James J. Spilker,et al.
GPS Signal Structure and Performance Characteristics
,
1978
.
[6]
Gary McGraw,et al.
Determination of C/A Code Self-Interference Using Cross-Correlation Simulations and Receiver Bench Tests
,
2002
.
[7]
Bradford W. Parkinson,et al.
Observed GPS Signal Continuity Interruptions
,
.
[8]
R. Gold,et al.
Optimal binary sequences for spread spectrum multiplexing (Corresp.)
,
1967,
IEEE Trans. Inf. Theory.
[9]
John W. Lavrakas,et al.
GPS Integrity Failure Modes and Effects Analysis
,
2003
.
[10]
M.S. Braasch.
Autocorrelation sidelobe considerations in the characterization of multipath errors
,
1997,
IEEE Transactions on Aerospace and Electronic Systems.