wise Amin man Fauzan - 7th November 2007 (06:43 GMT)
|Antenna hopping has already delivered on our vendor. It is diversity technique for DL path, namely Tx Diversity.|
Downlink performance is limited by multipath fading and interference “holes”. Using frequency hopping can only partly compensate for that limitation.
By combining Frequency and Antenna Hopping the multipath fading can be significantly reduced.
The peformance improvement is dependent on (multiplication of)
# of frequencies and antenna locations;
# frequency and antenna combinations.
It requires two trasnceiver module to handle single TRX funtionality.
Up to 3 dB gain by hopping on two uncorrelated antennas for low speed terminals in urban environment, even higher gain for quasi static mobiles and indoor coverage by outdoor BTS.
Keep bear in our mind that normal radio design is limited by capability of equipment being used when we're calculating the link budget which is normally higher path loss occured on the UL since power transmitted by the MS lower than BTS.
In case you enhance DL gain using Antenna hopping you have to compensate the UL gain.
In Karthi case, i assume that that it was frequency hopping since activation of antenna hopping is requires quite alot of effort.
deep_k - 5th November 2007 (14:06 GMT)
recently i also heard this term called Antenna hopping ......... i m also searching for it .
Pix - 2nd November 2007 (06:39 GMT)
It would be nice of you to give further details about "antenna hopping" ...
As far as I could find on internet, this is usually called "spatial diversity", isn't it ?
It's quite surprising to reduce the drops by implementing spatial diversity on "faulty" cells. I agree this solution will work, but :
1: The diversity should already be installed by default on all your GSM sites. It's almost a mandatory step in order to maximize the coverage and the "radio quality" of your cells.
2: The spatial diversity depends on your antenna configuration. Meaning that if you want to use spatial diversity, you need to replace the current antenna (from vertical polarization to cross polarization), or add a second antenna (+ 1 vertical polarized antenna). It's not just a radio parameter you can "enable" when you want.
I don't know... is antenna hopping = space diversity ?
Tufail - 2nd November 2007 (06:19 GMT)
Antenna Hopping provide some relief in Drops. its works similar to that of frequency hopping.
pix - 1st November 2007 (12:43 GMT)
i'm glad your problem was fixed.
antenna hopping oesn't exist (it would be funny to see antenna hopping onto each others though). I think your colleagues actually activated the frequency hopping :) Probably BBH.
It means that you were having path balance issues and interference issues. Time for a new frequency planning :)
Karthi - 1st November 2007 (08:23 GMT)
Thanks for ur valuable suggestions Pix.we observed some path balance issues with TRXs(sdcch) and changed the sdcch slot to other TRXs for observation.
Our planning guys enabled antenna hopping and got more improvement in sdcch drops.Do anyone aware of the significance of ant hopping?
pix - 30th October 2007 (10:27 GMT)
for sdcch drop, you can investigate the followings :
hardware/software problems (dynamic SDCCH ts allocation, TCU failures, RSL failures, TRX failures)
radio problems : Especially interferences on the frequency of the TRX's that is carrying the SDCCH's. Low coverage (low rxlev) can also have a big impact on the SDCCH drop rate (rural areas, mostly, as soon as avg rx lev of the cell is less than -88dBm)
Don't forget to check the path balance of your "SDCCH" TRX's.
handover problems : if SDCCH handover is enabled, then you might want to disable it and check the results.
You can also check the average duration of the SDCCH seizure : if it is above 5 seconds, then there might be "hanging" DTAP message during the SDCCH phase. If the SDCCH phase lasts too long, it increases the risk of drop.
That's about it...
just a reminder : the SDCCH congestion has no effect on the SDCCH drop.
Karthi - 30th October 2007 (07:35 GMT)
I hav been working in nokia systems and concern about sdcch drops which shows more than 3% for many cells. What could be the list of causes to figure out the issues?
pix - 25th October 2007 (15:18 GMT)
|welcome back mister !|
i tell you : your answer was clear...
but it's ericsson that is not very clear :)
Da Architect - 25th October 2007 (07:36 GMT)
|It's been a long time, but now I am back!|
Dear Pix you are right about tch drop and congestion when we talk from Alcatel point of view. But in ericsson the counter of TCH Drop [TNDROP] increment with another counter [TNRELCONG] that increment upon congestion on TCH Transcoder or Ater during assignment to better or worst cell. so the correct equation for call drop rate must be adjusted to be as following: 100 * (TNDROP - TNRELCONG)/(TCASSALL + ((SUMIHOSUCC - SUMIABSUCC - SUMIAWSUCC ) - (SUMOHOSUCC - SUMOABSUCC - SUMOAWSUCC))))
to reply for Sam questions:
subscriber tch drop must be the above equation, while the other tch drop rate is calculated without taking into consideration the TNRELCONG and HO and AB and AW.
also note tch drop due to sudden loss is not related to tch congestion... as i mentioned above, counter TNDROP increases with TNRELCONG.
I hope i made it clear.
Post a reply to this article