Integration of GPS with Other Sensors and Network Assistance - Understanding GPS: Principles and Applications - page 535

four GPS satellite signals at -147 dBm within 16 seconds with success rate of 95% or
better. Corresponding minimum performance specifications for other cellular tech-
nologies (CDMA, TDMA, GSM, AMPS, or WCDMA) have not yet been established
but are being pursued. The collection of various types of cell phone technologies are
sometimesreferredtoas“MA”becauseofthelasttwolettersofCDMAandTDMA;
thus, each handset type can be referred to as being of an MA type . The standards
body for each cellular MA technology is addressing the custom requirements and
unique capabilities of each type to create its own minimum performance test. For
CDMA handsets, its precise time knowledge (to approximately 100 ns) provides an
advantage to minimizing the total number of code phase/Doppler search cells
because the projection of the 100-ns time error into the search space is less than
one-half of one GPS chip. Adherence to an MA's minimum performance test does
not, in any way, guarantee that the particular handset meets the FCC mandate.
The FCC mandate is not the only driver for high-sensitivity GPS: Use of a cell
phone within a car as a navigation aid to drivers is an emerging application for GPS
embedded in cellular phones, as referenced in Section 9.3. A nonoptimal GPS
antenna design is expected, with losses ranging from 5 to 15 dB relative to conven-
tional, stand-alone GPS antennas; in addition, operation within the car is expected to
contribute an additional 3- to 5-dB attenuation. Thus, embedded handset applica-
tions alone requires that acquisition and tracking thresholds be extended roughly 10
dB relative to those for GPS operating in more conventional open-sky environments.
In Section 5.13, the benefits of extended coherent and noncoherent integration
in reducing signal acquisition thresholds have been addressed. In order to maximize
GPS coverage in an effort to satisfy the FCC mandate, maximizing the use of
extended integration is highly desirable. Ideally, use of the assistance information
will enable coverage of the search space with the number of correlators in the
receiver to permit parallel searching across satellites and stationarity with respect to
the correlator bin location during the extended integration interval. If sufficient
correlators are not available to cover the total uncertainty space in parallel, some
form of sequential processing is required.

9.4.3 Total Uncertainty Search Space
For a given scenario, one can compute the total Doppler-code phase uncertainty
space required to be searched and, consequently, the number of correlators required
to cover the entire search space in parallel. The initial parameters of position, time,
and frequency uncertainty, along with the particular orientation of the satellite con-
stellation at the time, can be used to compute the total uncertainty search space. Fig-
ure 9.41 depicts the two-dimensional search space for each satellite, the x -axis
representing the total Doppler uncertainty, and the y -axis showing the total code
phase uncertainty. For each satellite, the number of Doppler search bins ( N dopp ) and
code phase search bins ( N cp ) is computed.
The number of required correlators N c to cover the search space in parallel is
given by:

M

= ∑ 1

NNN
c

= × (9.40)

dopp cp

i i

i