Erlang software tools...
Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors

Erl/Sub

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • 2nd March 2007 at 10:41 #46759
    Arian
    Guest

    hi!
    I am qorking on two large networks at the moment. One network is quite huge interms of subs and NEs, the other one is very small interms of Subs and the NEs. There are only 12k subs in the whole network at the moment. The network was designed by our pre-sales for 20k subs. Now we are facing a problem since the access network is very very small and the terrain is such that its acting more like a WLL then of a GSM standard network. I will cut short and will state the problem here. The problem is : For 12 k subs the BHCA is 58771 and the MHT is 893 (Its just one figure, i understand that it keeps on changing). Now if i calculate the Erl/sub it come out to be 1.21 erl/sub, using the formula Erl/Sub= BHCA*MHT/3600*Active Subs.
    Now if i use the formula :
    Erl/Sub=BHCA*MHT/3600*Active Subs*X
    where X= 3600-MHT (in sec)
    the value is 0.4 Erl/sub.
    I have two questions which formula is correct.
    Secondly, dun u think the BHCA is too high??? i have anothe rcounter which says radio resource congestion and the value for that is nearly 30k for this particular time. Now this reflects the subs behaviour, if they are unable to make calls they keep on trying again and again thus building up a huge traffic on the network. Is this explanation plausible?

    2nd March 2007 at 19:27 #46760
    Pix
    Guest

    Hello Arian,

    that’s very interesting, let’s try to solve this together…

    MHT = 893 seconds.
    That’s the first thing that bothers me. Of course, your erl/sub is gonna be way too high. Average MHT around the world is about 30 to 60 seconds. Average erl/sub is about 20mErl (0.02 Erl)

    Can you imagine if on average, one subscriber spends about 13 minutes per call in one cell ? It is impossible. They’ll eat up their subscription fees in few days…

    I believe your MHT either :
    – is wrongly computed
    – doesn’t represent average MHT per cell and per call
    – shows a BIG problem of frozen TRX (a TCH keeps being allocated in your cell even though the call has been released)

    Could you please upload some QoS excel reports (values per cell and at the busy hour) and send them to me ? (my email is xav.storage (at) gmail dot com )

    The “radio resource congestion”…
    what is the unit ?
    can you try to find the “rate of radio congestion per cell at busy hour” instead ?
    include also the erlang per cell

    Your BHCA is very high (4 calls / subs at BH !) .. need qos report to understand why 🙂

    i’m waiting for your mail,
    regards,

    pix

  • Author
    Posts
Viewing 2 posts - 1 through 2 (of 2 total)
  • The forum ‘Telecom Design’ is closed to new topics and replies.