gpr-forum.com

Forum dedicated to Ground Penetrating Radar technology
It is currently Sun Jun 17, 2018 10:18 pm

All times are UTC




Post new topic Reply to topic  [ 42 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next
Author Message
PostPosted: Thu Jan 07, 2010 10:28 am 
Offline

Joined: Wed Apr 29, 2009 5:52 am
Posts: 96
Location: Zagreb, Croatia
Hi GeoAdmin!
Unfortunatelly it is not the one (we are moving from one office to another so I'll probably put all of the documents through my hands in the next week or so and I hope to find it then).
On the other hand it's an excellent paper that covers all of the methods for determining Time zero (without including the system lag offset, but it is a good start to comprehend and test for yourselves which method works the best for your job and equipment) that I heard of till now so thumbs up for this one!
It also covers the material "sucking in" the signal for the ground coupled antenna which is something rarely explained this thoroughly but in a nice straightforward manner.
Thank you

_________________
"If the only tool you have is a hammer, you will see every problem as a nail."


Top
 Profile  
 
PostPosted: Sat Feb 06, 2010 8:11 pm 
Offline

Joined: Tue Apr 28, 2009 8:58 am
Posts: 23
any luck finding that paper?
Robert


Top
 Profile  
 
PostPosted: Wed Jun 16, 2010 1:34 am 
Offline

Joined: Wed Jun 16, 2010 1:18 am
Posts: 11
Hello Everyone,

I am john. I am new to this forum. I have a lot of queries which i nees help with. I don't know if this is the correct thread to post the queries regarding basic GPR doubts. If not please direct me to it.

Most of the GPRs use stepped frequency modulation. I wanted to know can we use linear FM or polyphase codes for this.

Stepped frequency sweep is used, coz low freq. is used for penetration and high freq.for resolution. Please correct me if i am correct till now.

jOhN


Top
 Profile  
 
PostPosted: Wed Jun 16, 2010 6:23 am 
Offline
Site Admin

Joined: Fri Apr 17, 2009 7:52 am
Posts: 379
Location: Boden, Sweden
Hi John,
You can post on this thread or create one of your own with questions specific to what you need to know.
In your statements you are partially correct. Although there are several very successful companies selling
stepped frequency GPRs, the vast majority of commercially available GPRs are impulse GPR though.

In many research projects the stepped frequency method is used because people already have the required
equipment: a sweep generator (or a tracking generator for that matter) and a spectrum analyzer, or simply
a network analyzer. Impulse GPR demands very precise timing generators, low noise A/D converters and
good synchronization. All of this usually not readily available of the shelf and therefore substantial research
is on demand.

You are correct about the fact that lower frequency is required for penetration and higher frequency is needed
for resolution. Please note that I use the terms "lower" and "higher" since all GPRs are basically high frequency
devices working in the VHF and UHF regions of the spectrum with very few exceptions.

Finally, independently on the fact that you use stepped frequency or impulse radar you are always in the quest
of measuring time. This implies that frequency domain will have to be somehow translated into time domain.
Usually this is done by applying the inverse FFT to the frequency data. In many cases this is done quickly and
effectively and the original way of dealing with the antennas becomes irrelevant. That said, I most say that
you will probably hear differently from others using one way or another and saying their way is superior.

Well, welcome to the forum and hope this is the place you were looking for.
Regards,

_________________
Reinaldo Alvarez Cabrera
Geoscanners AB
Sweden
http://www.geoscanners.com


Top
 Profile  
 
PostPosted: Wed Jun 16, 2010 11:26 am 
Offline

Joined: Wed Jun 16, 2010 1:18 am
Posts: 11
Hi GeoAdmin,

Thanx for welcoming into the group. I have some further doubts, i think you can help me with it.

1) As we know the range is dependent of frequency. As the freq. increases the range comes down. In GPR most of the systems being developed now are based on Stepped frequency CW wave. In this as i said earlier high freq. is for resolution and low freq is for penetration. Consider a typical GPR the freq range varies from 200 mhz to around 3 - 4 Ghz. The range we can achieve with 4 Ghz signal is very low in some cms.( using the formula r = (c/2*f)). And the typical range of the GPR will be 3 - 4 feet. So there are 2 target is at a depth of 2ft or so, The high freq. signal will not first of all penetrate to that depth so where does the question of resolution arise. I hope you got the problem.

2) Can you tell me the signal processing portion for downrange processing. I am not able to simulate it in MATLAB.

This is what i am doing. I am generating a stepped freq cw wave.
Simulating the return signal from the target by delaying it.
Mixing both transmitted and received signal and taking the IFFT of the mixed output. This processs should ideally give me the range profile. But i cam not gettingit properly. Can you tell me what could be the flaw.


Regards,

jOhN.


Top
 Profile  
 
PostPosted: Wed Jun 16, 2010 5:19 pm 
Offline
Site Admin

Joined: Fri Apr 17, 2009 7:52 am
Posts: 379
Location: Boden, Sweden
Hi John,
I don't think I understood your question properly, but this is what I think:
1. The question of resolution is always related to the same frequency not to different ones like you mention. The real spectrum of
GPR might contain multiple frequencies, but because of imperfections in the antenna, restrictions of the bandwidth and more the
spectrum is never flat. What I'm trying to say is that although the 10dB BW can go from say 200 and up to 800MHz it doesn't mean
it is flat all over the pass band and therefore taking about resolutions in the deepest areas based on the higher content of the
spectrum has little if not zero relevance. The opposite holds true as well and analyzing the resolution near the surface taking into
consideration the lower frequency components might seem odd to say the least. A more or less consensus is to take the center
frequency and do your estimates from there. This of course is an estimate and give ground to errors, see this:
http://www.geoscanners.com/pdf/antres.pdf
This is not complete and to some extent very controversial. I good friend of mine, which in my opinion is one of the leading experts
in interpreting GPR data in the present, doesn't agree with me on this. However, I do believe in test and prove and what holds for
"most practical purposes" it is good enough for me, after all I'm just an engineer.
2. I didn't get your question very well, but I think you are having trouble with MATLAB. Your thought flow seems logical and sound,
so in principle it should work although you might look into the delaying part of it.
May I suggest a superb set of tools for seismic/electromagnetic geophysical surveys? I guess I may, so please see this:
http://www.crewes.org/ResearchLinks/FreeSoftware/
this is actually not free if you are not a student, but you can get some very good points and everything is done in MATLAB so it
should be a familiar environment.
Let us know how it is progressing, all this is always very interesting.
My best regards,

_________________
Reinaldo Alvarez Cabrera
Geoscanners AB
Sweden
http://www.geoscanners.com


Top
 Profile  
 
PostPosted: Thu Jun 17, 2010 1:34 pm 
Offline

Joined: Wed Jun 16, 2010 1:18 am
Posts: 11
HI Reinaldo,

Thanx for the prompt reply.I am studying GPR right now and to be frank was intimidated with the theory as to how it is possible. I guess this forum will help me get through with it.

The pdf u gave me mentions that for a carrier freq. of 1300 Mhz the resolution we get is 116mm. This is calculated using the formula (r = c/2f) i guess. If this is the resolution formula, what is the formula relating the range and frequency.

And i think the radar was impulse type as it is mentioned carrier freq., actually i have one more query, in an impulse radar is there any concept of carrier as we transmit a short pulse. And I understand that the carrier mentioned will be the one cycle of the sinewave which will be equal to the pulse width. Please correct me if i am wrong. And if possible elaborate on carrier freq. and bandwidth of an impulse radar.

And if you are saying resolution is dependent on single frequency, how will we calculate the resolution from SFCW waveform.


And I'll check the delay thing as u suggested and get back to you.

Best Regards,

jOhN.


Top
 Profile  
 
PostPosted: Fri Jun 18, 2010 5:26 am 
Offline
Site Admin

Joined: Fri Apr 17, 2009 7:52 am
Posts: 379
Location: Boden, Sweden
Hi John,
Half a wavelength of the center frequency is a good estimate of vertical resolution, so you got this right.
As said before, there are very good experts that can "see" closer than that, but most people can't and therefore
a conservative approach is the way to go.

The range is a function of the depth you need to achieve and therefore depending on the relative dielectric
permittivity of the material you are examining. The deeper the target the longer the range, a bit on this you
can read here: http://www.geoscanners.com/pdf/gprsimnet_tut.pdf

There are however limitations on the range because the material is by no means loss-less and heavy attenuation
will occur. This is when the second parameter enters the equation, the conductivity of the material you are sounding.
The higher the conductivity the higher the attenuation and the less the signal will penetrate reducing your range sometimes
to nothing. See the attached pdf to get an idea. This file was submitted to us by our colleagues in Croatia surveying a lake
full of coal residues mixed with clay. Although the noise level is acceptable down to 600ns, that is no "ringing" is present,
the penetration is basically zero.

Lower frequencies will penetrate deeper, this is a concept commonly grasped without difficulty. A little bit more
subtle approach is that penetration/range is not only a function of the frequency, but factors such as transmitted
power, receiver sensitivity and some others even more obscure also affect the obtainable range. I have, and most probably
many reading this forum have also encounter situations when higher frequency antennas penetrate deeper than lower
frequency antennas in the shallow depth. What's the explanation to this? Well, there are many to my knowledge, but none
of them fit the above statement of lower frequency means deeper penetration.

To your other question, in an impulse radar there is no concept of "carrier" like in radio communications if that's what you meant.
I'm an electrical and electronics engineer by profession and when I came to work with GPR I was very confused at first because
I couldn't understand many things like dipoles with 100% bandwidth, SWR of 3 to 5 being accepted without even blinking and
many other "oddities" you may encounter. So, believe me, I think I know what's your surprise. I put some basic ideas on
bandwidth in this paper: http://www.geoscanners.com/pdf/gprbw.pdf

Finally, I didn't say resolution is dependent on single frequency, I say we use the center frequency for all calculations since the
passband it's not flat and we might be speculating if we use frequencies too close to the end of the bandwidth since those can be
easily masked by more powerful lower frequency components. This is one of the reasons why GPR requires such a large bandwidth.

Best regards,


Attachments:
File comment: Zero penetration in a Croatian lake
zero_penetration.jpg
zero_penetration.jpg [ 33.35 KiB | Viewed 7197 times ]

_________________
Reinaldo Alvarez Cabrera
Geoscanners AB
Sweden
http://www.geoscanners.com
Top
 Profile  
 
PostPosted: Sun Jun 20, 2010 2:27 pm 
Offline

Joined: Wed Jun 16, 2010 1:18 am
Posts: 11
Hi GeoAdmin,

The pdfs were really helpful, thanx for them. You were saying to use the center frequency, but is it not beneficial to use the extreme value which can take care of the worst scenario. Instead of center which we are certainly sure that won't give correct results for half the band.

And I have a doubt regarding antenna beam patterns, as i am not able to grasp it properly still. I think no GPR existing takes care of the near field operation into account. All the GPR's existing consider the antenna still to be working in far field, by playing with the dimensions of the antenna, as the near field and far field discrimination is done on the basis of D(Antenna Dimension) and Wavelength. Do you know any GPR in the market with works in near field conditions. Also it would be helpful if u can share something regarding this near field antenna gain properties. As to how to calibrate the pattern etc. please refer me some literature if possible.

Regarding the down range resolution. I couldn't get the output, i checked the delays as u said. As i mentioned the process earlier, Consider the case of single freq. transmission instead of SFCW( to prove the theory). When the signal is reflected from the target. The information of range is stored in phase. so you mix the signal and low pass filter it and keep only the phase information. How can we translate that to range information mathematically, It is suggested to use IFFT in literature, but i am not able to do it. As the phase change becomes a constant. I don't understand how we get the range profile.

I guess i have so many questions :), so please don't mind.

Thanks for your patience and prompt replies.


Best Regards,

jOhN.


Top
 Profile  
 
PostPosted: Wed Jun 23, 2010 8:30 am 
Offline

Joined: Wed Jun 16, 2010 1:18 am
Posts: 11
Hi GeoAdmin,

Sorry to disturb you,I guess u are busy wid something, But could you find the answer for my query.

It is open to everybody i think, so if anyone else can please help me out, that would also help.


Best Regards,

jOhN


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 42 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group