PDA

View Full Version : [Question] Distance to ServingCell in WCDMA?



hoangtuna
2011-12-25, 06:38 PM
Hi everyone,

I have a question need your help.

In WCDMA, is there any parameter like "Timing Advance" of GSM. The parameter help to approximate the distance from UE to ServingCell.
"Round-Trip-Timing"?.How to calculate?

Thanks for your help.

Hoangtuna.

esoestemp
2011-12-26, 03:39 AM
Yes - there is such parameter.
Its name is Propagation delay:
The duration of one chip (3.84Mcps) correspond to approximately 78 meters in propagation distance. If the delay estimation operates on the accuracy of samples/chip then the achievable maximum accuracy is approximately 20 meters. There are other inaccuracies that will cause degradation to the positioning but 20 meters can be considered as best possible positioning performance. UMTS specifies that it will provide location information for mobiles to an accuracy of 50m. With GPS assistance, maybe even 10 meter accuracy is possible

s.sunilpal123
2011-12-27, 02:52 PM
Yes - there is such parameter.
Its name is Propagation delay:
The duration of one chip (3.84Mcps) correspond to approximately 78 meters in propagation distance. If the delay estimation operates on the accuracy of samples/chip then the achievable maximum accuracy is approximately 20 meters. There are other inaccuracies that will cause degradation to the positioning but 20 meters can be considered as best possible positioning performance. UMTS specifies that it will provide location information for mobiles to an accuracy of 50m. With GPS assistance, maybe even 10 meter accuracy is possible

How did u calculate 20 m from 78 m?

esoestemp
2011-12-28, 05:44 AM
This is a quote from other resources(I am not discovering hot water) and in my opinion here the point is overall characteristic of this method for mobile positioning.

dholama
2011-12-29, 06:28 PM
Yes,

It is called as Propogation Delay.



Hi everyone,

I have a question need your help.

In WCDMA, is there any parameter like "Timing Advance" of GSM. The parameter help to approximate the distance from UE to ServingCell.
"Round-Trip-Timing"?.How to calculate?

Thanks for your help.

Hoangtuna.

dholama
2011-12-29, 06:29 PM
Nomally 1 % corresppongs to 350 mtrs

Processor
2011-12-30, 04:43 AM
This is a quote from other resources(I am not discovering hot water) and in my opinion here the point is overall characteristic of this method for mobile positioning.

It is more useful if you offer an advise or solution based on what you know or have tried or you well know is proven....just my 2cents

esoestemp
2011-12-30, 07:18 AM
You mean to tell the gay that I discovered what is Propagation delay and the formalization in 3GPP is mine work.
It is better to think more of that you posted. OK - because you are writing in this topic, can you explain for us what is your practical experience, explanation and philosophy for this sophisticated thing called Propagation delay - I am going to rep you!

wolverine
2011-12-30, 07:49 AM
I guess the big difference between the two, is that in GSM the timing advance could be read from the UE but in UMTS the propagation delay is something the NodeB calculates. So you can only see it if you have access to the nodeB.

hoangtuna
2011-12-30, 09:33 AM
I guess the big difference between the two, is that in GSM the timing advance could be read from the UE but in UMTS the propagation delay is something the NodeB calculates. So you can only see it if you have access to the nodeB.

That's right. In GSM, with an TEMS logfile, you can know the distance, but with WCDMA logfile, can we do it?

josogun2
2011-12-30, 11:12 AM
Even without TA and propagation delay in tems one can still know the distance to serving cell with the use of ruler on the map.
Some analysis tool like actix we auto generate the distance to serving cell too.


That's right. In GSM, with an TEMS logfile, you can know the distance, but with WCDMA logfile, can we do it?

yafawi79
2012-01-02, 08:27 PM
how about cell radius parameter?

kentck86
2012-02-13, 09:37 PM
TA and propagation delay are just estimation distance from serving cell to the UE, but how do we pinpoint the exact direction of the exact Lat Long for the UE if we only have statistic data and without drivetest data?

patek1968
2012-02-14, 10:36 AM
if you want to know the exact long lat of the ue, drivetest must be done using tems or nemo..i think its impossible just only statistic data.

kentck86
2012-02-14, 11:09 AM
Hi,

I think it is possible to get the lat/long for the UE if we have the TA/Propagation Delay data, Serving Cellid,RNC,LAC,MCC by using triangulation calculation.

dabreulima
2012-02-29, 04:50 AM
Dear All,

Look this tool: http://www.finetopix.com/showthread.php?27365-Tool-for-analysis-coverage-on-Google-Earth&highlight=propagation+delay

T (http://www.finetopix.com/showthread.php?27365-Tool-for-analysis-coverage-on-Google-Earth&highlight=propagation+delay)ool for analysis coverage on google earth based on Timing advance and propagation delay. Just for Huawei, but you can adapt for another vendor.

25064

Kindly,

Processor
2012-03-01, 12:45 AM
Can this be seen from a Tems log?..how can i tell this?

firstmaxim
2012-03-01, 03:15 AM
Hi everyone,

I have a question need your help.

In WCDMA, is there any parameter like "Timing Advance" of GSM. The parameter help to approximate the distance from UE to ServingCell.
"Round-Trip-Timing"?.How to calculate?


Thanks for your help.

Hoangtuna.

1 bit duration = 1/270 kbps =>3.7 micro second => 1.11 km (= 3.7 microsecond x c)
Since there are 63 steps => 1.11 x 63 = 70 km
Considering round trip => 35 km

linken318
2013-02-20, 02:17 PM
Its name is Propagation delay , we can't see it from drive test but in RNC trace .
to-nodeB's Signaling :[NBAP-RL-SETUP-REQ] contain thisparameter
1 propagationDelay =3 chips
1 chip =0.26us
0.26us × c =78 meters
1 propagationDelay=78×3=234 meters

hshai
2013-11-11, 03:08 AM
Hi bro,
Do you know the method to calculate lat/lon base on Propagation delay and other things? Please share the idea!

Thanks!

mirc84
2013-11-21, 04:21 AM
Hi bro,
Do you know the method to calculate lat/lon base on Propagation delay and other things? Please share the idea!

Thanks!

Hi,
Propagation delay (or ta in gsm) gives you a distance, nothing more. You should at least know the position of the site, the azimuth of the antenna which is serving you and optionally the horizontal opening (i forgot the exact name of the antenna param) of the antenna. I. This way is it possible to find out the area of your user. (Just search xx meters(pro delay or ta) in the antenna direction).
in huawei you may also find counters (something with .TP. )which may show you the quality of the signal at different distances from the site.

Also I have found a calculator which calculates.the.distance between to points, give the lat / long of each point. I hope someone will find it usefull. :-)

http://www.movable-type.co.uk/scripts/latlong.html

mirc84
2013-11-21, 04:50 AM
Hi,

I think it is possible to get the lat/long for the UE if we have the TA/Propagation Delay data, Serving Cellid,RNC,LAC,MCC by using triangulation calculation.




Hi, I'm not sure about the prop delay (because in 3g you have soft ho). But the TA in gsm is only for the serving cell. So you cannot do a triangulation. Of course nothing is impossible ( if you find a way tl force the ue to query the distance to all the seen neighbours, good for you :-))
And also there are the 3gpp stamdardised positioning methods : a-gps, rtt, etc

You could also try using rxlev, rscp, but you need to know the antenna horisontal characteristic and you need to be in los.