-
Notifications
You must be signed in to change notification settings - Fork 9
/
TODO
69 lines (50 loc) · 2.41 KB
/
TODO
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
Open Source GPS To Do List
--------------------------
* General improvements
** Get it to work using interrupts from the card (must have newer
version of the card with these interrupts available)
** Allow data message to output every six seconds with full frame
information and keep user side working in sane manner. Likely
involves keeping an old copy so you can see which part has changed
since last sub-frame. (don't look at sub-frame number from status,
that's lame and might have timing problems.)
** WAAS support
- Need previous problem solved at a minimum
- Step 1: Add WAAS PRN codes and search for them
- Step 2: Verify you can track a WAAS satellite
- Step 3: Data pull off changes since it's 5x data rate.
- Step 4: Fix user side for this data change
- Step 5: Implement DO-229c on user side.
** Add a matrix math library (file) so we can do Kalman filters and
other higher level math stuff.
* Linux Specific
** Figure out an elegant solution to number of doppler search bins bug
when you don't have an almanac. Dilemma: The number of search bins
is large if doing cold search in device driver and automatically
reduces when channel is assign. However, userland cold search
assigns channels and messes up the whole idea. Obvious solutions
are to either not do cold search on user side at all (this might
break the DOS version) or allow user side to control number of
search bins. I don't like the second one because it would mean
modifying the interface.
** Write another user side program that uses the device driver to
allow for more testing of interface and show flexibility. Perhaps,
re-use some other open source GPS libraries from other groups to
make these easier.
** Add support for gpsd so we have a moving map via something like gps
drive.
** Allow the user program to set the system time if it has authority
to do so. (root permission?)
** Integrate with other kernel time setting code to allow for
extremely accurate time control. (this could be really useful for
certain applications.)
* DOS Specific
** Get PCI assembly working for DOS and Borland C 2.01
** Get the DOS version to use the RTC instead of timer interrupts
(mostly just for consistantly with Linux version.)
(This next bit for the EMACS editor's benefit.)
Local variables:
mode: outline
paragraph-separate: "[
]*$"
end: