I have a very specific question about 3GPP TS 23.040 GSM Spec. In
section 9.2.3.24.8, the spec indicates that the short message within
the TP-UD is 151 characters (160-9), for concatenated SM 16-bit
reference number. However, UDH takes 6 byes plus UDHL 1 byte in the
concatenated SM 16-bit reference number. This means, 7 bytes will be
taking from the 140 bytes total for the SM. So, we have 133 byes
multiply by 8bits = 1064 bit. 1064 divides by 7 segments = 152
characters. So this leaves us with 152 characters for short message
instead of 151 as stated! My question is: is that extra character that
is being taking is used as a fill-bit? if so, why? Or I'm miss
understanding something here?
I would appreciate any hints or replies.
Thanks.
John Henderson - 13 Sep 2003 09:36 GMT
> I have a very specific question about 3GPP TS 23.040 GSM Spec.
> In section 9.2.3.24.8, the spec indicates that the short
[quoted text clipped - 8 lines]
> being taking is used as a fill-bit? if so, why? Or I'm miss
> understanding something here?
I see nothing wrong with your reasoning. I suspect that the
statement "the maximum length of the short message within the
TP-UD field is 151 (160-9) characters" in clause 9.2.3.24.8 is
incorrect. The corresponding statement concerning 8-bit
reference numbering (clause 9.2.3.24.1) correctly says that 153
7-bit characters are available in that case (it's actually 153
with one spare bit left over).
Because 16-bit referencing takes an extra byte, that's 1 and 1/7
less characters available. And 153 1/7 minus 1 1/7 = 152. But
it would be 151 if the figure for 8-bit referencing messages was
153 exactly (without that one extra bit left over and unused).
I've found a few other mistakes in GSM 23.040 when working on EMS
a couple of years ago. It doesn't seem to be up to the usual
high standards.
It would be interesting to analyse a concatenated message sent
from a handset which uses 16-bit referencing - just to see how
manufacturers are interpreting the standard.
John