(meteorobs) [radiometeoren] Re: Question about airplane and automatic counting Spectrum Lab

Sat Sep 14 11:38:04 EDT 2013

Hello James

As I stated in my previous post, I asked for "Ideas." I was hoping someone
would post a long duration Audio File somewhere for me to download
containing Aircraft and meteors, although I suppose I could mix 2 files
together that have both. I could create code for SpecLAB only if I had
sample audio recordings since there does not seem to be any beacons in my
region. I am very good at coding but I need files to work on. 

From: meteorobs-bounces at meteorobs.org
[mailto:meteorobs-bounces at meteorobs.org] On Behalf Of James Beauchamp
Sent: Friday, September 13, 2013 11:13 PM
To: Meteor science and meteor observing; ernotp at gmail.com
Cc: radiometeoren at vvs.be; cm2esp at gmail.com
Subject: Re: (meteorobs) [radiometeoren] Re: Question about airplane and
automatic counting Spectrum Lab

Hi guys, 

As someone who has been running a receiver with speclab for a few years now,
but lacked the time to do the programming, I would suggest the following
recommendation...

Since the doppler solution will be a bistatic one, aircraft reflections will
slowly change compared to those of meteor hits.  If you commonly get "head
echo" hits prior to persistent echos from overdense events (sometimes you
will get no persistence in small events, so it is the head echo only), you
can use FFT bins from previous time slices to discriminate targets with long
time constants from the short ones.  Aircraft will slowly change in
frequency over time compared to meteors that suddenly appear with HIGH
doppler rate head echos.

It is a bit complicated on the code side, to which I am not experienced
enough to help with recommended methods, but if i were a coding "god", i
would average the bins in the time domain.

Aircraft will exhibit slow frequency change from a high abscissa to a lower
one, but will be slowly changing compared to meteor reflections.

If, due to the lower base frequency, you get ONLY long-term reflections (no
head echos) from the meteors, another discriminator could be the
instantaneous appearance of an echo over several seconds.  Again, the code
could use time-based binning of previous FFT's to identify sudden
appearances of consistent-doppler signals.  This will be a bit more noisy,
with a higher number of false targets from aircraft (due to fading as the
reflections appear), but might be better value than nothing at all.

Just a couple of thoughts...

James

  _____  

From: "maarten.vanleenhove at telenet.be" <maarten.vanleenhove at telenet.be>
To: ernotp at gmail.com 
Cc: radiometeoren at vvs.be; cm2esp at gmail.com; Meteor science and meteor
observing <meteorobs at meteorobs.org> 
Sent: Thursday, September 12, 2013 4:21 AM
Subject: Re: (meteorobs) [radiometeoren] Re: Question about airplane and
automatic counting Spectrum Lab

If someone would find a solution for this, I 'd like to hear about it!

kind regards,

Maarten

  _____  

Van: "ernottp" <ernotp at gmail.com>
Aan: cm2esp at gmail.com
Cc: "Meteor science and meteor observing" <meteorobs at meteorobs.org>,
radiometeoren at vvs.be
Verzonden: Donderdag 12 september 2013 10:58:45
Onderwerp: [radiometeoren] Re:  Question about airplane and automatic
counting Spectrum Lab

Hi,
Maarten Vanleenhove wrote a script for Spectrum Lab in which he it tries to
avoid false detections due to aircrafts.
His url is:
http://www.supernovae.be/meteorcounting
However, errors occur.

Following him: "When two aircrafts reflections merge, the script will, due
to the signal width, assume a meteor was picked up. When large meteors
appear, sometimes the meteors will be double counted"

I  do not use this script because we detect too many intersections of
aircrafts (beacon at Dourbes in Belgium 49.97Mz)

Pierre Ernotte

2013/9/11 Raydel Abreu (CM2ESP) <cm2esp at gmail.com>

Hello all,

Recently, by September 6th I was able to migrate my radio forward scatter
system from 187 MHz TV to 61 MHz TV. Obviously as the frequency is lower a
very high increment in rates is present.

With the previous system a manual counting was performed so false positives
by airplane traces were ignored. As my new system is peaking 150 meteor
traces by hour as an average manual counting is impossible.

A modified script for SpectrumLab based on the meteor_test distributed on
the software installation package was written to track the weak but sporadic
present direct signal carrier and it works very great.

However, at least one or two times a day some airplanes gets into the
working pass-band and produces a lot of false positives; producing false
"hotspots" which difficult to identified as airplanes produced or real
meteors outburst like the one recently detected by some members.

Any advice or tips??? Or a recommended better script to deal with this
"airplane outburst". The happening of this problem is very seldom, but still
affecting data quality. See efficiency and the "hotspots" in the attached
picture.

Thanks and best wishes,

Raydel, CM2ESP

Havana, Cuba.

_______________________________________________
meteorobs mailing list
meteorobs at meteorobs.org
http://lists.meteorobs.org/mailman/listinfo/meteorobs

  _____  

No virus found in this message.
Checked by AVG - www.avg.com
Version: 2013.0.3408 / Virus Database: 3222/6665 - Release Date: 09/14/13

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.meteorobs.org/pipermail/meteorobs/attachments/20130914/69c6924b/attachment-0001.html 