Dear Karthik,
As my colleague and myself have already indicated, specifications of CDRs
in the context of GSM MSC vendors and their implementations are even more
specific to the vendors. Ideally, the information you request must be
sought with the Vendor. For e.g., if you are working with Ericsson CDR
Module, then you will have to consult with Ericsson for their Specs and
sample CDRs, as the implementations will definitely have some holes.
I guess, you ahve some clarity now.
regards,
Sathya Narayanan
LnT Infotech
|---------+----------------------------->
| | "Karthikeyan |
| | Ramnath" |
| | <[EMAIL PROTECTED]|
| | xgroup.com> |
| | Sent by: |
| | [EMAIL PROTECTED]|
| | |
| | |
| | 11/21/2003 10:29 |
| | AM |
| | Please respond to |
| | asn1 |
| | |
|---------+----------------------------->
>------------------------------------------------------------------------------------------------------------------------------|
|
|
| To: <[EMAIL PROTECTED]>
|
| cc:
|
| Subject: RE: [ASN.1] some doubts
|
>------------------------------------------------------------------------------------------------------------------------------|
guys, thanks for the inputs, but how do i handle a long run of 0xFF bytes?
This is something we saw in a GPRS source. Currently i am skipping any run
of non-tag bytes starting with a 0xFF. But according to the encoding
rules(BER) 0xFF is a valid encoding. It would represent a multibyte
constructed tag. But in my case I have to reject it. Is there any document
which says that the 0xFF is a special value or anything like that?
kr.
---------------------------------
The will to do, the soul to dare
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of KB
> Devaraj
> Sent: Friday, November 21, 2003 7:32 AM
> To: [EMAIL PROTECTED]
> Subject: Re: [ASN.1] some doubts
>
>
> Hi karthik,
>
> Here are the few tips to write a generic parser.
>
> Most of the CDRs genereted by MSCs are use Context
> Specific tag class.These tags are unique within the
> module but in within the specification.That means
> within the same specification itself, the same tag can
> occur in different modules.
>
> Most of the CDRs tags are encoded in single
> byte.Length of the cdr is encoded in two bytes in most
> of the call types.
>
> Length value for the indefinite lenght cdrs is 0x80
> and the indefinite length can be identified using End
> of Content (i.e. 0x00 0x00).when u write parser, make
> sure the the byte next to EOC is a tag.
>
> MSC cdr specifications are depends on operators.Its
> not the same for all operator.So it is better to check
> with the operator for CDR specification.
>
> TAP 3 uses Application tag class.In applciation tag,
> tags are unique within the specification.
>
> In tap3. some of the tags are encoded in two bytes and
> length as well.Indefinite length record is very much
> possible in TAP 3.Mostly it'll occur for Mobile
> Originating call and Mobile Terminating call.
>
> You can download all the TAP 3 specifications from
> www.gsmworld.com
>
> About filler,
> Almost in all cases filler will be 0x00.Most of the
> MSCs uses block structure of 2048 bytes.
>
> You need give more attention while parsing indefinite
> length cdrs.Example for indefinite length cdr
> generetad by MSC is CompositeCallDataRecord(is nothing
> but sequence of GSMCallDataRecord).
>
> Hope this will help you in some way.
>
> K B Devaraj
> Intec Telecom Systems.
>
>
> --- Karthikeyan Ramnath <[EMAIL PROTECTED]>
> wrote:
> > Dear Sir,
> > At work, I have to deal with parsing Call Detail
> > Records generated by GSM
> > MSCs, TAP 3 applications, etc which are in ASN.1
> > format.
> >
> > I wrote a generic parser/decoder which accepts the
> > tags and the value
> > encodings in a tree structure as configuration, and
> > ran into some doubts.
> >
> > Presently, I'm unsure of how to handle 0x00 and 0xFF
> > values between records.
> >
> > Also how are fillers/noise represented and handled?
> >
> > Also, is there any place from which we can get the
> > ASN.1 specification for
> > the different applications/enterprises? Just so that
> > we can shift to using
> > generic compilers.
> >
> > Thank you.
> > karthikeyan.r
> >
> > ---------------------------------
> > The will to do, the soul to dare
> >
> >
> > DISCLAIMER: This email is bound by the terms and
> > conditions described at
> > http://www.subexgroup.com/mail-disclaimer.htm
> >
>
>
> __________________________________
> Do you Yahoo!?
> Free Pop-Up Blocker - Get it now
> http://companion.yahoo.com/
>
DISCLAIMER: This email is bound by the terms and conditions described at
http://www.subexgroup.com/mail-disclaimer.htm
