UL interference in big events

[Luis123]

Hi guys,
I want to ask if you can share what actions you take on your network on big events, like festivals, or soccer games where we have many user at same time. Do your network become instable or do you have many access problems?
Normally on such events RTWP hits very high values, >-70. Is there any recomended value of rtwp that we should overcome, or network becomes instable? Do you know what actions we cantake to avoid suchbig UL interference?
Im thinking on limit MS power and works parameters on RACH access, reducing maxretranspreamble and cycle. Anyother sugestions

[P.T.E.]

Dear,

High RTWP will highly impact PS DCR, I suggest in such events to increase CQI feedback period which is by default 2ms. this action will improve PS DCR on RNC level.

BR,

[dtvt2]

Hi friend
RTWP hits very high values, >-70. I think the network has external interference
dtvt2

[Optimustron]

There are some mitigation on changing some parameters but is very limited on "big events", the RTWP is high because of the quantity of users, you need to build a temporary site with many antennas, we do that in those events, we have installed antennas and we set equipment for that purposes, we set a site with 5 sectors and a special antenna with 5 beams, the solution works very well, but when the event is better the traffic is so high and we get a high RTWP, however we have a lot o traffic and the competitors too, we try wifi too for offload but the first time this didn't work well, we are going to make new trails on that. You can reduce the power of carrier and PICH too maintaining the dB difference in order to have good ECIO, however you need to put more antennas to cover the same area because you are reducing coverage and power capacity. You can try with METRO FEMTOCELLS not HOME/OFFICE FEMTOCELLS because this only can carry up to 5 users maximum. The solution involves a big effort and you need to convince all people in your company to set a good solution for a few days or maybe one day, if you convince them then ask for the best solution and make numbers. I hope this give you a perspective of what you need to do however almost I can tell you that traffic raises so high in that events and you aren't totally prepared instead you put a very big solution and capacity. We have some repetitive events annually however traffic always congest our solutions but we are thinking next time because for each new event we gain experience, more traffic and almost... almost satisfied clients, maybe next time. Best regards,

[yafawi79]

Good morning Dears; allow me to add one point. One way to reduce the high RTWP is to reduce the UL TX power of the UE that having high BE data rate by reducing the data rate. If the data rate reached to the maximum MBR. The RNC will reduce it to UL middle data rate. This will help to reduce ps call drop and enhance the number of Hspa users. Also, QoS interfreq and interrat HO can help to reduce the RTWP. Also; DCCC can be used to transit the low traffic volume UE from DCH to FACH state to reduce the data rate based on the UL TX power measurements and based on the high BLER which will be calculated by RNC mac layer and compared to the set threshold to help to reduce the data rate.
Thanks.

[Luis123]

Thanks for all comments and ideas.
On last big event, ive already reduce max MS power allowed from 24 to 20.
Ive read a good solution can be increase BLER from 1% to 10%, but have not test it. Aneone has feedback?
Also read increase CQI feedback seems a good solution to test.

My problem is most of changes i see possible are done at RNC level, and i want to act at cell levelonly, to avoid impact on other cells that has nothing to do with event.

[Luis123]

Hi friend
RTWP hits very high values, >-70. I think the network has external interference
dtvt2

It hasnt dtdt2. RTWP go so high due to traffic increase,that for sure.

[Luis123]

There are some mitigation on changing some parameters but is very limited on "big events", the RTWP is high because of the quantity of users, you need to build a temporary site with many antennas, we do that in those events, we have installed antennas and we set equipment for that purposes, we set a site with 5 sectors and a special antenna with 5 beams, the solution works very well, but when the event is better the traffic is so high and we get a high RTWP, however we have a lot o traffic and the competitors too, we try wifi too for offload but the first time this didn't work well, we are going to make new trails on that. You can reduce the power of carrier and PICH too maintaining the dB difference in order to have good ECIO, however you need to put more antennas to cover the same area because you are reducing coverage and power capacity. You can try with METRO FEMTOCELLS not HOME/OFFICE FEMTOCELLS because this only can carry up to 5 users maximum. The solution involves a big effort and you need to convince all people in your company to set a good solution for a few days or maybe one day, if you convince them then ask for the best solution and make numbers. I hope this give you a perspective of what you need to do however almost I can tell you that traffic raises so high in that events and you aren't totally prepared instead you put a very big solution and capacity. We have some repetitive events annually however traffic always congest our solutions but we are thinking next time because for each new event we gain experience, more traffic and almost... almost satisfied clients, maybe next time. Best regards,

I agree with you, but we must improve without investing and that is a big challenge

[yafawi79]

In such a big event You can enable F2 and F3 beside to the first carrier and distribute the traffic among the carriers through the service priority algorithm. I.e F1 for voice call only while F2 & F3 are dedicated for PS and R5 data.

[Optimustron]

Ok Luis, then apply the solutions provided by the people, for example the CQI reporting is well and limited but step by step you can do that, put CQI reporting to 8 ms and if you have free spectrum you can enable at least 2nd carrier, if not then reduce UE power more than you told us but you need to make some test in the worst place of the coverage and see the tx power of ms in a steady state or almost steady, then change the load of cell from 25% to 50% then to 75% this in order to see any changes on tx power of UE with the final result add 3 dB near to raleigh fading in order to set a secure value of Max UE TX POWER. Of course all of this before the event, I hope this can help you, best regards,

[P.T.E.]

Also Adding second carrier with CAC & LDR algo is a good solution for special events. but in case you are talking about events like new year, then you can increase CQI feedback period from 2ms to 8ms, also you can decrease GBR.

BR,

[sabila]

for huawei equipment
setting ENU to max 200 on ucellcac
setting backgroun noise to 160 on ucellcac
setting rxbranch attenuation for 18(9 dBm reduction)

thats for RTWP problem due to load traffic
for the rest u should maximise the CE,Code,IUB
reduce power so the coverage only cover the arena

[Luis123]

for huawei equipment
setting ENU to max 200 on ucellcac
setting backgroun noise to 160 on ucellcac
setting rxbranch attenuation for 18(9 dBm reduction)

thats for RTWP problem due to load traffic
for the rest u should maximise the CE,Code,IUB
reduce power so the coverage only cover the arena

I use Algoritm first, so ENU is not a question...which CAC algo do you guys use with best performance? is it the second?

What is the best solution: is it to use rxbranch attenuation or adjust initial rtwp? i talked with friends and some do attenuation, others adjust initial rtwp, so im curious to know your experience.

[FrankPintor]

Have you switched off HSUPA? Also, disable the higher data-rate R99 RABs. You can also modify the RACH parameters and reduce the impact of RRC signaling on the uplink.

[Optimustron]

Luis,
You need to state that you have a physical limitation on the number of users because of RTWP, codes, fach, channel elements, Iub, power, you can maximize the quantity of that but you need to see at the maximum number of users in a single UMTS channel with good experience, if you make too many changes in order to reduce the RTWP that comes because of traffic, you always will have a bottleneck and surely bad experience if your customers like high speeds, we agree that we need to put the less quantity of sites in order to reduce capex and opex, however you need to show this facts to the company and the complaints from customers, the KPIs for the site or sites involved in the solution, I understand you because our boss told us this in the past, in the first events and We show him the outcomes of the poor investment, now we are setting more capacity and sectors and of course doing some things the people said in this forum, by the way we have a big event recently, Aero Smith concert and we were successful first time, but we invest a lot of resources, time and efforts, please don't misunderstand me, take the good advices of the engineers in this forum, I'm thinking about some of them, but think in advance to make a good presentations about this experience, then you will have the support from the owners of your company. And thank you very much for this post you really move the people and this is the purpose of this forum, i.e. to share good ideas and successful stories and when you get there please tell us, best regards,

[rfsupp]

I use Algoritm first, so ENU is not a question...which CAC algo do you guys use with best performance? is it the second?

What is the best solution: is it to use rxbranch attenuation or adjust initial rtwp? i talked with friends and some do attenuation, others adjust initial rtwp, so im curious to know your experience.

Hi Luis,

The common adjustments to reduce the RTWP is:
Increase CQI Feedback
Increase the BLER for R99 and HSUPA(number rtx)
reduce the initial SIR
Adjust the HSUPA PO
Reduce the channel switch for higher RAB UL
adjust the RACH access parameters
Limit the max UL UE power
Adjust the T308/N308

If you want to let the user get in, for sure algorithm 2 is better and I saw in same cases even turn off.
Im not sure if to adjust the initial RTWP or rxbranch attenuation will solve the problem. Maybe will solve your kpis, but the user experience my friend will be the same(bad)

[Luis123]

Have you switched off HSUPA? Also, disable the higher data-rate R99 RABs. You can also modify the RACH parameters and reduce the impact of RRC signaling on the uplink.
Switching HSUPA makes a worst performance as we have more R99 traffic.
Thanks for the other 3 sugestions

[Luis123]

Luis,
You need to state that you have a physical limitation on the number of users because of RTWP, codes, fach, channel elements, Iub, power, you can maximize the quantity of that but you need to see at the maximum number of users in a single UMTS channel with good experience, if you make too many changes in order to reduce the RTWP that comes because of traffic, you always will have a bottleneck and surely bad experience if your customers like high speeds
Aneone has a tool to make this estimations how many users we can serve?

Of course on main events (concerts, company brand events...), we install a mobile unit with lots of capacity and even use wifi to drive traffic, but we cant do that for allin the country, and are mainly for that reasonwe need to implement low cost solutions trying to improve the best possible user experience.

[faisaladeem]

Gents, attached are some parameter changes that can be done to solve high RTWP issues (in H// RAN). Also you can turn off HSUPAOLSCHSWITCH on NB and reduce MAXTARGETULLOADFACTOR from 75 to 50 or less.

[Luis123]

Gents, attached are some parameter changes that can be done to solve high RTWP issues (in H// RAN). Also you can turn off HSUPAOLSCHSWITCH on NB and reduce MAXTARGETULLOADFACTOR from 75 to 50 or less.

Dont have permissions to download. I send you my email, please send me the attach