Global Positioning System Reference
In-Depth Information
correlator spacing is reduced, but that there is only a slight loss of tracking threshold
as the correlator spacing is reduced (provided that the receiver front end bandwidth
is increased appropriately).
Figure 5.28(c) uses (5.23) with Db
2 to demonstrate the improved DLL track-
ing threshold by reducing the noise bandwidth. Figure 5.28(d) uses (5.23) with Db
=
=
2 to illustrate the improved DLL tracking threshold by increasing the predetection
integration time. The 1-second predetection integration time provides the lowest
code tracking threshold. To support a predetection integration time greater than 20
ms (the navigation message data bit interval), the data wipeoff process must be
implemented for C/A code and the normal mode of P(Y) code. Recall that this tech-
nique uses the GPS receiver's a priori knowledge of the navigation message data bit
stream to remove the 180º data transitions. This data wipeoff technique permits
longer than 20-ms predetection integration times and, if properly implemented, can
achieve nearly 6 dB of additional ( C / N 0 ) dB threshold improvement [see Figure
5.28(d) at threshold crossings]. This is a short-term “desperation” DLL weak signal
hold-on strategy for an externally aided GPS receiver when the carrier is aided open
loop. Data wipeoff also improves the PLL tracking threshold when the carrier loop
is closed-loop aided, but not to the extent that the code loop tracking threshold is
improved. Changes in any part of the SV navigation message data stream by a GPS
control segment upload or autonomously by the SV will cause errors in the data
wipeoff, which, in turn, will cause deterioration in the tracking threshold.
The DLL tracking loop dynamic stress error is determined by:
n
n
dRdt
R
=
(chips)
(5.25)
e
n
ω 0
where d n R / dt n is expressed in chips/s n .
As an example of how the dynamic stress error is computed from (5.25), assume
that the code loop is an unaided third-order C/A code DLL with B n =
2 Hz and D
=
1
chip. From Table 5.6, for a third-order loop
B n /0.7845. If the maximum LOS
jerk stress is 10 g/s, then this is equivalent to d 3 R / dt 3
ω 0 =
98 m/s 3 /293.05 m/chip =
0.3344 chips/s 3 . Substituting these numbers into (5.25) results in a maximum
dynamic stress error of R e =
=
0.02 chip, a 3-sigma effect. Since the 3-sigma threshold
is 1/2-chip, this would indicate that the DLL noise bandwidth is more than adequate
for C/A code. If the receiver was P(Y) code, then R e
0.2 chip, which is still ade-
quate. Note that carrier-aided code techniques removes virtually all the dynamic
stress from the code tracking loop. Therefore, as long as the carrier loop remains
stable, the code loop experiences negligible dynamic stress, and this effect is not
included in the code loop tracking threshold analysis.
=
5.6.4 Modernized GPS M Code Tracking Loop Measurement Errors
The modernized GPS M code uses a BOC s (10,5) modulation technique (see Sections
4.2.3 and 4.5.3). By substituting (4.17) into (5.22), the following approximation for
M code DLL jitter in the presence of thermal noise can be determined [13]:
 
Search WWH ::




Custom Search