Hi Marcus
Al last a voice in the wilderness or is that wilder beasts :)
Yes everything is power , be it noise or signal , I suppose a signal
is simply a refined noise or it would not have a bandwidth ..
There will always be a advantage over a data 'decode' , by using
correlation as one is required to assemble assemble data block's .
then by logical routine , recover the original data , for Opera
that's a minimum of 40% of the original signal , say 100 data
block's , where as the pattern match . takes places over 40 % ,
effectively summing the 40% to increase the recovered power ,,, then
needing a single [final] yes or know decision ..
I think ultra stability will enhance the Dynamic system , but as in
the ''Ros House style'' it was developed on simple 'ham' kit ,
the - 5 dB is realised in this configuration .
The gain could exceed 10 db , but the level of false positives would
be 'silly' JR went for -5 dB , which gives a % false equal to or
better than the Opera decoder - unfortunately , the psk-map
displays' all spots , which leads to a certain amount of 'excitement'
the web linked windows app is the ''only'' validated out put from the
Dynamic mode ,
Note ,''tested at op05'' the opera decoder's are linked to band ,
with set speed , the parameters are not the same for all the
speeds , scaling may not give compatible results , Op05 is the
experimental 'flash' decoder , which is giving good results on vhf
with wide spread signals , where as the OP8 and OP32 are configured
in the more stable [ zero spread] lower bands ...
'' both systems reliably produce false detections ''
This is a statement of fact not a comparison, if neither produced a
false out put , then it could be argued , that the minimum -dB
level had not been addressed and the design requirement to ' dig
out of the noise' had not been achieved , other wise what is the
point .. the observed problem being , Decodes = random call
Correlation = valid call ... which provides endless amusement ... :)
.. From observation , the Dynamic system can be seen to be affected
by assumedly band or local conditions , some times nothing , other
times , more than is desirable , which , would tend to suggest its
set at the optimum ?
Both use the same decisions , but having the web linking in place ,
JR added the final 'dynamic' check of the detection 'time' while
vetting the call lists , other wise its a stand alone system
73-Graham
G0NBD
Half steins or schnapps :)
--------------------------------------------------
From: "Markus Vester" <[email protected]>
Sent: Saturday, May 23, 2015 7:23 PM
To: <[email protected]>
Subject: LF: False Decode or Real? - Opera vs opds
Hi Graham,
OPDS makes use of Wolfs some what excellent spectrum software as
DSP
Spectrum Lab is used as the frontend for opds, the only signal
processing done there is a straight high-resolution FFT before
exporting the data. Opds then searches for spectral peaks, and
transforms a 1024 bin channel around each peak back to time domain (0.5
Hz wide, 4x oversampled). For coherent signals, the central carrier is
extracted (including possible drift and fading), and synchroneous
demodulation is performed. Then the resulting real waveform is Fourier
transformed again for a CPU-efficient crosscorrelation to each of the
callsign templates.
both systems reliably produce false detections when subjected
to noise ..
Opds-32 has been running on LF continuously, typically analyzing about
20 QRM peaks every 10 minutes in a 115 Hz wide band. The searchlist has
currently around 50 entries. With a correlation threshold of 15 dB, 28
false detections have occured in 20 weeks since year 2015:
https://dl.dropboxusercontent.com/u/26404526/opds32.txt
In a pdf file distributed with Opera 1.5.7 software, the author EA5HVK
mentions that he tested opds and got 11 false detections in only 4
hours - an observation which is obviously not corroberated by my
statistics.
hence ultra stable TX and RX can give advantage
By using synchroneous demodulation rather than power detection, opds
can detect coherent and stable signals that about 4 dB weaker. Attached
are two plots, showing a side-by-side comparison of detection
probability and SNR output from the Opera 1.5.6 decoder and the
opds2h5c detector.
SpecLab's digimode terminal was used to generated coherent and
perfectly timed Opera signals, and white noise from the test signal
generator was added with variable power density (dB/Hz). The very same
output was analyzed wihin SpecLab to feed opds, and played to the Opera
software usingh VAC. To speed up the experiments, all testing was done
at Op-05 speed (30 seconds), and SNR values were then scaled down by 24
dB to Opera-32. During the test, no false detections were observed in
the output from either program.
In the attached splot uccessrate.png, the solid lines with squares show
detection probability (0 to 100%) against average SNR in 2.5 kHz. The
classic Opera decoder (red) achieved 50 % detections at -40 dB. Opds
correlation (blue) goes down to -49 dB, showing a 4 dB advantage for
these ideal signals. The blue crosses indicate correlation dB output
from opds - note that only hits above the standard 15 dB threshold
(dashed line) were counted as successful detections.
SNR_output.png shows indicated SNR values versus actual SNR from both
programs. Opera 1.5.6 seemed to consistently read 1 dB high, whereas
opds reads approximately 1 or 2 dB low, with a larger scatter. Part of
this negative offset is because I had originally assumed the "dBOp"
scale to be referenced 4 dB (instead of 3 dB) below PEP.
I have not yet tested the dynamic deep search in Opera 1.5.7, but the
claimed -45 dB threshold (ie. 5 dB better than the decoder) seems quite
plausible. I believe opds is also around 4 dB less sensitive for
non-coherent signals, which would then put both programs in same
ballpark.
Now mines a pint or are we onto shorts now ?
Please explain...
All the best,
Markus (DF6NM)
-----Ursprüngliche Mitteilung-----
Von: Graham <[email protected]>
An: rsgb_lf_group <[email protected]>
Verschickt: Sa, 23 Mai 2015 4:48 pm
Betreff: LF: MF 630m: False Decode or Real?
intelligent life in other Galaxy's ? Like the Bar ? 11:14 . Suns
not
over the yard yet .. Tad early Eddie ?
Armature radio V Armature hour , take 2
Opera MF and LF is a BOGOFF mode , Buy one and get one free ,
just
that its free to uses and the first is a data mode the second
mode
is a correlation system , dynamically engaged .. page 70 seems to
miss this rather important fact along with the design solutions
embodied therein , I'm waiting for page 71 ..
The test of a good design is that no one notices , it just works .
In that 24 hour window , Opera LF produced no false data
detections
or false dynamic detections , where as the wspr system regularly
fills the LF map , a simple test of design , preventing false
correlation
detections is more difficult than false data . Opera LF is -40
dB
and -45 dB ... that well cool as J C would (of) said
Now the tacky bit
As Im sure Markus will tell, OPDS and Dynamic share the same
design criteria
To drag low signals out of noise , by pattern matching , OPDS
makes
use of Wolfs some what excellent spectrum software as DSP , whilst
Mr.
Ros uses his own designs . both systems reliably produce false
detections when subjected to noise .. not all the time , just
depends
on the detection monkeys sense of humour on the day.
The design solutions' branch , OPDS makes accurate frequency
measurement and Bandwidth , along with allowing parameters to be
adjusted by the user ,as well as maintaining the look up table
hence ultra stable TX and RX can give advantage .. And as pointed
out
, presents the user with a set of parameters, which may be used
as
validation
Opera Dynamic retains the Plug and Play house style , yes these are
criteria , but are evaluated by the system , load it and it dose
the
rest , Opera data runs as normal , Opera dynamic is engaged
should
the decoder fail , the sever handles the validation and maintains
the
list.
Both systems OPDS and DYNAMIC produce false real hits , Dynamic
take
things one step on, where wspr uses the internet as part of the
DSP
sync . Opera Dynamic uses the internet to pool ' recovered time'
data
and validate the spots by identity and coincidence of time ..
Now mines a pint or are we onto shorts now ?
73-G
...
|