- This topic has 20 replies, 1 voice, and was last updated 20 years ago by
.
-
Posts
-
Could someone explain to me what a trunk or line means with reference to erlang calculation?
Tell me the characteristics that a line or trunk should have to be erlang B formular compatible.
Lastly, I would like someone to provide information about the erlang B model…..This is rather an abstract concept especially in the wireless mobile world.
Thanks in advance..
Mwenya M
A trunk or a line is a circuit from a PBX or phone to a telephone exhange(CO). This type of cct. is suitable for erlang B calculations.
Hi Mwenya,
i have the exact question in my mind also. to my understanding, the trunk (american terminology) is used to connect the switches in the exchanges and is typically an E1 (lowest) in the telco world. However, my question is, in the Erlang B formula, it does not define what the trunk dimension is. Is it an E1, an E3 ??
The Erlang B formula basically says given the desired grade of service and the amount of traffic, we can work out the number of trunks required but what is this trunk size physically ?
Do you have any thoughts on this or have anyone enlightened you on it ?
thanks.
In this context, a trunk is one discreet voice connection. An E1 comprises 30 trunks. A T1 comprises 24 trunks. Together, trunks for a trunk group, and Erlang B assists in sizing a trunk group (ie. working out how many trunks a trunk group needs).
Sometimes ‘trunk’ and ‘line’ is interchangeable and when you said that, there is a reference point which is normally omitted for convenience. For example, a ‘trunk’ in a PABX is actually a ‘line’ with respect to the telco side. And a ‘trunk’ of the telco means the link between telcos.
How erlang per line is calculated of an EWSD exchange from the RAW data?
also what is standard design of erlang / line world wide?A little more on line terms…T1 aka Bell1 aka Digital Signal 0 or DS0 is an American standard, with 24 64k channels and a rate of 1.544k. E1-aka CEPT1 is the European standard with 30 64k channels and a rate of 2.048k, and J1 24 64K channels is the Japanese standard, same rate as US, but not compatible..line coding and framing is different.
Tre’
Can someone please give me the formula to calculate telephone trunk traffic manually if I knew the erlang. A little explanation would help.
Hello Mwenya…
I would direct anyone looking for the skinny on Erlang to look for a white paper on the subject. Perhaps I’m just lazy… Also couldn’t hurt to look up poisson(sp) to compare.
Anyway, the way line vs. trunk works, in my opinion, is that a trunk ties switches together, whether it’s a switch in a network or a PBX at a company’s premises is really unimportant to the definition of a trunk. Trunks are normally not rich in features, and are almost always brought in (In the US) by legacy, T-carrier facilities.
A line on the other hand, connects end users, (stations) to a switch, and most always has a concentrating device between the switch and the customer. Centrex service is provisioned on lines… and there is a trunk to line ratio (such as, three T1s (72 trunks) go in, 96 lines come out). Or something like that.
Hope this helps some.Lines connect users to networks, trunks interconnect switches.
Trevor, T1 is carrier, DS1 is the service provided, DS0 is the 64k timeslot.
Hello Ezra,
Your comment makes interesting reading and provokes thoughts about how one should dimension lines or trunks.From your explanation of lines and trunks,it is clear that the probability of lost calls is different for each case.
I’m made to understand that a trunk faces ROUTERS at each end and has thus many more possible combination of routes.Do you then expect, for the same traffic, a TRUNK to have a higher GOS than a LINE ?Thanks for that thought-provoking contribution !!
Thank you for bringing that mistake to my attention Ezra.
Is it true that the Erlang value is always between 0 and 1, and cannot be bigger than 1 ?
I have been told that Erlang is always between 0 and 1. However, after reading the whitepapers, this is does not appear to be the case. Does anybody have any idea where this confusion can come from ? Is there eg another parameter related with Erlang that is always between 0 and 1 ?
I have been told that Erlang is always between 0 and 1. However, after reading the whitepapers, this is does not appear to be the case. Does anybody have any idea where this confusion can come from ? Is there eg another parameter related with Erlang that is always between 0 and 1 ?
An Erlang is merely one hour of talk time, or 36CCS, or 3600 call seconds.
If you have two trunks and each of them carries 30 minutes worth of traffic in an hour, you had one Erlang of traffic that hour.
One trunk can not possibly carry greater than 1 Erlang of traffic, could that be the 0 and 1 thing you refer to?
Also, Grade of Service must be between 0 and 1. Either 0% of calls get blocked or almost 100% of calls get blocked.
