- This topic has 15 replies, 1 voice, and was last updated 10 years ago by franky.
2nd March 2010 at 02:43 #61251dharmendraGuest
what is link imbalance10th March 2010 at 16:16 #61252JasonGuest
Some please correct me if I am wrong.
Link Imbalance is the difference between you Downlink (tower to mobile) and you Uplink (mobile to the tower). Normaly this is to strong on the side of the downlink. I belive the basic ways to correct this is: power down the site, downtilt the site or add some type of TMA or GMA to the site.10th March 2010 at 16:30 #61253pixGuest
good definition 🙂 just 2 things to add : the link budget is usually rather balanced, or sometimes a little weak on the DL side. That’s because in a cell with more than 4 TRX, there are 2 combining stages. One combining stage removes 4.5dB of DL power. 2 stages remove 9dB of DL power !!
In UL, there are not such losses due to combining (indeed, there is no “combining” loss at all in UL).
In other words, cells with high number of TRXs are DL limited. Cells with less TRXs are usually balanced or slightly UL limited (depends on BTS sensitivity…)
2nd thing : downtilting the antennas are not going to improve UL or DL : the antenna has exactly the same pattern in UL and in DL. downtilt modifies the UL and the DL path loss in exactly the same way.
TMA, powering down, high power TRX, less combining stages, Rx diversity, Tx diversity… all those methods can favour one direction compared to the other.
pix11th March 2010 at 09:22 #61254BHTGuest
With all due respect to PIX.as he has helped me a lot in optimization and i am keen follower of this forum.
I want to share my experience that we can improve KPI’s and uplink quality by using downtilt.
e.g if site is in country side and tower height is 60 with antenna at 55.if TA of the site is around 35.
which means the MS at 35 TA is able to camp on cell.As we know downlink signal is more strong and uplink signal strength is low as MS transmitting power is less then that of BTS antenna.so there may be a chance that due to downlink signal MS camp on it and HO command is sent to MS and due to poor uplink condition MS reply wasn’t reached to BTS the BTS and HO failure can occur.It also degrades uplink quality in KPI’s.
I will be very glad if some one can correct if i am wrong.
Thanks11th March 2010 at 11:25 #61255PixGuest
your action is correct, but what you are doing is actually reducing the coverage of your cell. You are not fixing the imbalance, just “hiding it”.
The imbalance is still here, but now the traffic is located closer to your BTS. Therefore both directions are good enough for all users, even though the UL is still a little weaker than the DL.
A “proper” action (but expensive) to improve your coverage up to TA=35 is to add a TMA or improve your RX-Diversity (use Space Polarization with a distance between the 2 antennas of about 6m) (ensure both antennas are facing same direction)
Cheers and thanks for the nice words, i appreciate it very much 🙂
pix11th March 2010 at 17:02 #61256JasonGuest
I am in learning mode now. I have a question about the link imbalance. If I have a site that has a link imbalance according to my stats, what other statistics should I see problems with and to what degree? Can you give some examples? Thank you in advance.12th March 2010 at 03:41 #61257vishnu rGuest
the linkimbalance in site is having a differance of 5db , we done our best by changing the RRU & jumber cables .but it is still their.
plz help me to rectify the issue12th March 2010 at 06:54 #61258PixGuest
5dB is acceptable, no problem !
it is above 8 or 10dB that you should start worrying.
KPI impact : smaller coverage, CSSR and CDR are a little bad (especially at the cell border, when coverage starts to decrease). High amount of HO due to bad level (UL or DL, depending on the direction of the imbalance)
And if the link imbalance is only on one TRX, you’ll see a poor TCH allocation success rate on this specific TRX, while all other TRX are good.
pix12th March 2010 at 12:17 #61259BHTGuest
Yes u r right that we are hiding the problem but right now company is not ready to invest money in the network.
SO we that is why we are going for “Cost effective” solutions.
I want to ask on thing and that is about channel allocation.
BSC allocates channel and send activation request to BTS and then BTS open power amplifier on that TS.
Is my this what you are calling allocation.
SO how we can see allocation on trx level mean success & attempt?
because actually activation is done on BTS level not the allocation.
Hope you understand my question.12th March 2010 at 12:39 #61260PixGuest
it is the story of a MS that does a call setup…
after the sdcch phase, the BSC looks for a TCH to give to the MS : that is counted as a “TCH REQUEST”.
if the BSC finds a TCH in the cell, then it will ask the BTS to do the channel activation. It is only possible if the cell is not congested, and if there are no Abis failures, or other software failures.
This is counted as a “TCH ALLOCATION”.
If the cell is congested, the BSC doesn’t send the channel activation, and there is a “TCH ASSIGNMENT FAILURE CONGESTION”. The KPI “TCH ASSIGNMENT PREPARATION SUCCESS RATE” decreases.
Once the TCH is allocated, the BSC sends the Channel Assignment message to the MS. The MS then tries to hop onto this TCH ts.
If it is succesful, then it is counted as a “TCH SUCCESS”.
Otherwise, it is probably because there was a radio issue… then it is counted as TCH ASSIGNMENT FAILURE RADIO…. and the KPI “TCH ASSIGNMENT EXECUTION SUCCESS RATE” decreases.
These problems also occurs when the TRX on which the TCH is ALLOCATED suffers from imbalance. Then the executions are probably going to fail, because the imbalance actually translates into “radio problems”.
Hope it clear things up…
pix13th March 2010 at 17:06 #61261BHTGuest
thanks for detailed response.18th April 2011 at 11:37 #61262frankyGuest
I am working on a network with both link imbalance and sync problems. But the cells with link imbalance perform poorly on some days and meet the requirements on some days… Also link imbalance reduces on the days when they meet the criteria.. How is it possible given that link imbalance is due to harwdware issues which are permanent??? I initially thought it might be due to HO_UL_QUALITY being affected by sync but HO_UL_QUALITY_ATTEMPTS AND SUCCESS both increase on days when performance is good?? Any methods to get around the problem?? The sync issue might take time to resolve..18th April 2011 at 16:02 #61263pixGuest
the only time were there are strong variation of link balance is if there is too little traffic in the TRX. In that case you might see “crazy” values.
typically, you need at least 0.1 Erlang @ BH to see “reliable” results. The more the better.
Apart from this case, I agree with you : the link balance should not change too much over time, since the only cause of link unbalance is faulty hardware.
Pix19th April 2011 at 11:58 #61264frankyGuest
@pix: thanks for the reply.. the traffic is same… I guess its cos of the changing traffic distribution(TA) and sync variation… will check on TA distribution changes…20th April 2011 at 21:06 #61265pixGuest
timing advance does not impact the link balance.
traffic distribution does not impact the link balance.