Dect coverage intermittent

[Vincent]

I work with a 4400 PBX, c5s3, ver R4.1-d1.311-ap-x1,
fitted with 6 Nodes. Omnivista 4760.
All DECT8 board on Node #3.

For a while, dect coms seem to be defective : intermittent "no answer from the system" is displayed,
but 1 minute later it can be OK. All areas affected, some of them maybe more than others.

Usually a [bascul] on Nodes 3 and 5 is enough to solve any tricky issue (don't know what node 5 has to do with it). This time it does not seem to be efficient.
I have tried [rstcpl] of all boards, and [outserv] - [inserv] of lots of bases.

* is this a software issue or something wrong with
boards ?
* is there some simple command to solve it ?
* some months ago I had to reduce carriers from 10
to 6 to comply with US regulations. If a call is
rejected due to this (all channels busy), will the msg
"no answr from system" be displayed ?
* I have read something about [dectisolon] : can it
be a usefull command in this case ?


Vincent

[torrentula]

Did you do a wireless audit before install? If you have too many base stations covering an area you can have intermittent problems. If you don't have enough base stations you will have issues too

[Vincent]

Well, the installation took place 5 years ago and I wasn't already here. As it was performed by Alcatel contractors I suppose they knew what to do...
About some areas not properly covered by bases : we have some but they are quite clearly
defined.
* It's now 2 days that I have switched back-on 4 additional carriers and it seem to improve,
but I still suspect something else.

[Eliott_DUP]

>If a call is rejected due to this (all channels busy), will the msg "no answr from system" be displayed ?
Yes. But don't think it's your problem. You can check the saturation of the RBS using traffic observation (dbgobs tool)
Maybe there is also an interference which makes some carriers unusable.
Normally you have many timeslots for each frequency so I don't believe that there is a saturation in the DECT itself.
Maybe for the RBS - but not in the carrier frequency.

I guess you have a synchronisation issue.
How many DTM card's are managed?
How is the synchronisation management?

Incidents in incvisu?
What's the result of infocs command?
...

PS: d1.311 is R4.1.1 and d1.305 is R4.1

[Vincent]

- there are 22 pieces of DECT8 boards on Node 3
- incvisu shows a few '1376' errors and lots of '3753 -1' but I should monitor it again
once I will switch back-off the 4 carriers.
- [infocs] answers "No synchronizing board is declared in the system, or synchronizing
board priority is 255"


I have no knowledge at all about synchronization, the only thing I know is that there are
two INTOF-B boards (Main and Standby) in the same shelf, and similar boards as well in other nodes, and I guess they have something to do with synchro. There no alarms related to them and they their red Led isn't flashing.


Thanks !

[Eliott_DUP]

What type of public access (trunks) is used?
As you have no CPU6 or CPU7 there should be set DTM function to True for two DECT8 cards.
(Directly in DECT8 board parameter)
Then you should set in DECT System - RBS System - DTM syncro to Yes and enter the ACT number with the DECT8 where the DTM is set to yes.
Caution! All RBS will reset after this management.

[Vincent]

- trunks are NDDI boards if that's what you mean, AWBC type if I remember well.
- DTM synvhro of RBS already set to YES and correct shelf displayed
- don't know how to enter DECT8 config...

1 thing more about synchro : Nodes are linked to our Master Clock System (MobaTime). The
procedure I follow is to adjust it Node by Node sometimes (does not seem logic to me but I've been told to do that). Anyway when checking with precision I see that the changes are not
effective : I use [swinst] and appropriate menu. At the end it says that both CPU are
updated and the correct time is displayed. When exiting this menu and running it again the
ancient time reappear. Time difference between Nodes does not seem to be more than 30 secs.

[Eliott_DUP]

- don't know how to enter DECT8 config...

mgr -> Shelf -> Go down -> Board -> Consult/Review -> Enter Shelf no and Board no of your DECT8 -> Ctrl+v -> Goto DTM function flag -> press Enter -> Ctrl+v

[Vincent]

...Well, after a week of test, situation is back to normal after setting the time on each Node as much
precisely as possible. Even areas wellknown for their poor coverage (i.e. small offices with entrance opposite and far from RBS) can get some signal now !

Thank you so much for driving me towards this synchronization topic...

[vali4715]

I work with 4400 r5.01 d2.314.7 ah, one node, 16 RBS, 2 x DECT8.
I have syncronisation with DTM from CPU6, I have 2 x PRA2 with 10 and 20 clock syncronisation.
I use only 100 pices Siemens c45.
Everything it`s normal, but appear the Vincent problems.
I have an error, I think 1387 with windi anomaly software, following the message that the RBS put in service, but time to time and not in the same time that my client reclams, I`m not 100% sure of that.
Any suggestions.
Did you have any technical communication that explain and solve this problem.
Major upgrade is very expensive and my client didn`t agree.

Thanks