Sorry if these are answered elsewhere, but just wondering:
1) Are there other cities using the same hardware and if so, are any
of them providing access to the GPS data?
2) Is the API custom by the city, or is it provided by the vendor?
3) Does the vendor have any software or hardware experts we can
access?
4) Urban canyon, while moving, is admittedly very difficult for any
GPS, but it would be interesting to know both the specs of the device,
the position of the antenna, any whether any steps are taken to ensure
good position (e.g. waiting until good lat/long is acquired before bus
getting underway)
5) Was the hardware specced for accuracy e.g. within X metres, or just
for "good enough to announce next stop"?
5a) Is it using pure GPS data to determine next stop or is it using
any acceleration, expected route etc.?
6) What technology is it using to transmit the GPS data? (GPS does not
inherently have the ability to "transmit" a position, they must be
sending it over a cell network or local radio network or something).
I'm curious because there can be quite a wide range of capability for
GPS.
My (woefully out-of-date) GPS logger spreadsheet shows some of the
range of possibilities
http://spreadsheets.google.com/pub?key=pXRBadSQxMUKN3TXwc3zOeA
For example a -158dBm, 32 channel MTK GPS can give substantially
better TTFF (time to first fix position) and handle reflection data
better than older models, but a great GPS chip won't do you any good
if the antenna is substantially blocked. There are also tuning steps
to take for example a GPS with settings adjusted for car use tends not
to work as well when you're moving at pedestrian speeds, and vice
versa.
Basically I'm trying to get a sense of where the constraints are: to
what extent are we constrained by the vendor hardware, to what extent
by the vendor software, to what extent by the ability to transmit the
location from the bus, and to what extent by the city's software,
servers and network bandwidth?