| |
ZL3AI > APRDIG 10.05.04 08:14l 244 Lines 10398 Bytes #999 (0) @ WW
BID : 3177-ZL3AI
Read: GUEST
Subj: TAPR Digest, Apr 20, 6/17
Path: DB0FHN<DB0FOR<DB0SIF<DB0EA<DB0RES<ON0BEL<IK1ZNW<VE3FJB<ZL2TZE<ZL3VML
Sent: 040510/0647Z @:ZL3VML.#80.NZL.OC #:23695 [Chch-NZ] FBB7.00i $:3177-ZL3AI
From: ZL3AI@ZL3VML.#80.NZL.OC
To : APRDIG@WW
Subject: Re: APRS greater precision
From: "Curt, WE7U" <archer@eskimo.com>
Date: Tue, 20 Apr 2004 09:40:13 -0700 (PDT)
X-Message-Number: 22
On Sun, 18 Apr 2004, Gerry Creager N5JXS wrote:
>But how do you know if your GPS is lying to you? Too many trust the
>hardware without understanding the error budget behind it. Worse, some
>manufacturers add software that tells you how "accurate" your autonomous
>position fix is, without a true scientific basis for that number.
>
>If you assume you can routinely get better than about 6m horizontal
>accuracy with a civil GPS, you're likely not well grounded in spatial
>statistics.
>
>That said, there should, in my opinion, be a better mechanism for
>transmitting and user control of precision than what we have today.
Agreed. Can we talk about relative positions of manually-placed objects,
and how the spec doesn't allow you to place them precisely enough relative
to each other unless you use compressed posits (or OpenTrac)?
For SAR, I have to switch to compressed objects/items, which aren't
supported in most other APRS applications. At the moment that's not much
of a problem for me (not having it displayed properly on other clients),
but for others, it might be. I might have need of compatibility like that
in the future though.
Note that I'm not talking about GPS positions here at all, and not talking
about map accuracy either, so get those out of your head.
--
Curt, WE7U archer at eskimo dot com
Arlington, WA, USA http://www.eskimo.com/~archer
"Lotto: A tax on people who are bad at math." -- unknown
"Windows: Microsoft's tax on computer illiterates." -- WE7U
"The world DOES revolve around me: I picked the coordinate system!"
----------------------------------------------------------------------
Subject: Re: APRS Kenwood Radios
From: Wes Johnston <wes@johnston.net>
Date: Tue, 20 Apr 2004 12:47:48 -0400
X-Message-Number: 23
Bob, I hope you know I don't mean any disrespect by these comments.... I'm
just grounded in practicality.
At 11:56 AM 4/20/2004, Robert Bruninga wrote:
>>It really does seem ashame ham radio has dropped so far
>>that a vendor is holding back progress.
>
>Duh, wrong planet you're on. Kenwood is holding nothing
>back. It is just a radio and a display. And it can display
>almost ALL of the APRS protocol. What is lacking is the
>creativity and initiative of HAMS to make use of it in
>imaginiative ways...
>Name me ONE thing that you think the Kenwood
>Mobile Radio APRS capability is "holding back"
>for the mobile or hand-held user? And I bet I can
>think of a way to do it with a little ingenuity instead
>of a lot bitching...
Everytime someone comes up with a new idea, you and some others jump up and
say "what about the kenwood users? We can't alienate them." (of course I
am paraphrasing). This external processor idea which reformats the data
and stuffs it back into the kenwood is a very interesting idea, but
impractical as stated in my previous email. I know kenwood offers a
firmware update for NMEA 2.0, so what I'm about to say is a moot point but
will serve as an example. Anyone running a GPS with NMEA 2.0 and putting
out GPRMC sentences just alienated all kenwood users who haven't
upgraded. As more and more new things come along, my kenwood will display
?? more and more often.
>Do you want to see 500 stations on the front panel
>of your radio instead of the nearest 40? Duh, it would
>be impossible to make any sense out of the 500
>nearest APRS users spread over half the continent.
>If you do want to see that many, then plug in a laptop.
>They conveniently provide you a serial port for that
>purpose..
Yes, that is very very true.... an external processor which does not use
the kenwood display and has it's own keyboard. Oh, and that laptop (err
external processor) has another advantage over the kenwood... it is
software upgradeable.
Bob you did some amazing stuff with DOS aprs and SATAPRS talking to the
kenwood radio... really neat stuff... and if I must say, that is what whet
my appetite for a kenwood so much before I bought one.
>And it seems that the most vociferous detractors
>and complainers don't even own one... or have
>a clue how they work and what they can do...
I own one and love it. I am very aware of it's short comings though....
(my minivan has some shortcomings too... namely how hard it is to look cool
with a baby seat in the back). I am not about to say _anything_ bad about
kenwood... they have been very good to me, and have done wonders for APRS
with their radios.... anyone remember 5 years ago when every manufacturer
was _trying_ to get on the APRS bandwagon? I remember Icom making a 2m
radio with a TFT TV screen on it... which you could (in theory) hook to a
laptop TV output and see aprs data... yeah right. Kenwood has my
support... but they also have my suggestions for making their product
better. Provide us a better software tools to talk to the TNC and
display. Now what would be cool would be to figure out their display panel
protocol and insert a processor there or replace their display entirely.
Has anyone heard that there will be a new d700 at Dayton? I got a sneaking
feeling, but nothing to back it up.
Wes
----------------------------------------------------------------------
Subject: Re: APRS greater precision
From: "Curt, WE7U" <archer@eskimo.com>
Date: Tue, 20 Apr 2004 10:00:53 -0700 (PDT)
X-Message-Number: 24
On Mon, 19 Apr 2004, Robert Bruninga wrote:
>Ignoring positions on APRS is not something I would encourage.
Unfortunately that's what you get with compressed station positions, and
very definitely with compressed objects/items, with many of the apps out
there.
I sure would like to see support for items, and compressed
items/objects/station positions put into most of the apps. Those are all
in the spec.
--
Curt, WE7U archer at eskimo dot com
Arlington, WA, USA http://www.eskimo.com/~archer
"Lotto: A tax on people who are bad at math." -- unknown
"Windows: Microsoft's tax on computer illiterates." -- WE7U
"The world DOES revolve around me: I picked the coordinate system!"
----------------------------------------------------------------------
Subject: Re: [ui-view] Ambiguity?
From: Steve Dimse <k4hg@tapr.org>
Date: Tue, 20 Apr 2004 13:08:39 -0400
X-Message-Number: 25
On 4/20/04 at 9:10 AM Curt, WE7U <archer@eskimo.com> sent:
>Some of the other APRS clients draw them as circles, which is just
>plain wrong. Work the numbers yourself.
That is true if the position is ambigous as a result of deliberate
obfuscation, for example simply dropping the last digits of the position.
However, if it is due to uncertainty, then the result is closer to a circle
than a square.
Unless there is something to link the error in latitude with the error in
longitude, the probablility of a position having error x in both longitude
and latitude (which is the corner of your box) is less than the probability
of having error x in one and no error on the other (the top, left, right,
and bottom of the circle). If you are doing positions manually, say from
taking bearings on landmarks, then the errors are not linked in the same
direction...yes, a bad lat makes it more likely there is a bad lon, but the
error is not necessarily of the same magnitude or in the same direction,
and therefore the error at the corner point is lower than the error at the
top.
At first glance, one might be tempted to say that latitude and longitude
errors are linked in GPS positions, and therefore the box is the correct
shape. Imagine a three satellite fix at the horizon (turning this into a
2-D problem), and then introduce a timing error in one of the satellites.
If the satellite is directly north of the user, the error will only be in
latitude, the satellite needs to be at a 45 degree angle to cause equal
errors...and of course, in that case, the error will indeed be in the same
direction. But...
For the same error in timing, the lat and lon are less affected than at the
90 degree points, to be specific, sin(pi/4) or .707. To get a position
error that corresponds to the corner point, you need to have a timing error
1.4 times greater. So, unless there is some reason that you have larger
errors when a satellite bears 45 degrees than 90 degrees, the error figure
will be a circle.
So, which is correct depends entirely on where the ambiguity comes
from...maybe we should add another field to the ambiguous position
reports...perhaps an equation that describes the shape of the error
function ;-)
Steve K4HG
----------------------------------------------------------------------
Subject: Re: APRS greater precision
From: "Curt, WE7U" <archer@eskimo.com>
Date: Tue, 20 Apr 2004 10:10:39 -0700 (PDT)
X-Message-Number: 26
On Mon, 19 Apr 2004, Robert Bruninga wrote:
>>>>Jeff King <jeff@aerodata.net> 4/19/04 3:22:16 PM >>>
>>
>>I've a mapping application can't display a Base91
>>compressed posit, what would you call that?
>>You call it tomato I call it tomatoe? If it doesn't display it,
>>it effectively ignores it.
>
>That's why I have resisted and still discourage anyone from
>using the compressed format.
That's unfortunate, as I have real-world applications where I need those
extra digits. I use it in Xastir quite a bit. So far it doesn't bother me
greatly that the Windows/Mac users can't see them. If it bothers them, they
can run Xastir on those platforms too.
>Just as you quoted in my next sentence:
>
>>>Ignoring positions on APRS is not something
>>>I would encourage.
>
>and that is why I do not encourage the use of the
>compressed position report on APRS... and why
>Iam working to come up with a workable alternative
>that adds precision but without causing any existing
>application not to still be able to use the position.
Why not encourage the developers to implement Base-91 instead of developing
yet another format for posits? It's in the spec you know...
--
Curt, WE7U archer at eskimo dot com
Arlington, WA, USA http://www.eskimo.com/~archer
"Lotto: A tax on people who are bad at math." -- unknown
"Windows: Microsoft's tax on computer illiterates." -- WE7U
"The world DOES revolve around me: I picked the coordinate system!"
----------------------------------------------------------------------
Read previous mail | Read next mail
| |
