NO can do.
In GSM, the transmitter is gated off while the reciever is on. So 7/1
is the limit.
>>Does nayone know of any handsets or Pccards that are higher than class
>>12 (more than 4 download timeslots)? I would really like a 8 timeslot
>>EDGE device.
> NO can do.
> In GSM, the transmitter is gated off while the reciever is on. So 7/1
> is the limit.
Hi,
as far as I know, there are no type 2 (simultaneous transmit & receive)
GSM/EDGE terminals or cards on the market.
However, there are terminals available with more than 4 timeslot reception
capability: for example, Nokia N93 has GPRS/EGPRS multislot class 32
supported. It does, however, only support 3 timeslot transmission (whereas
class 12 supports, supposedly, 4 timeslot transmit).
There is one thigh, though, to consider about the class 12 devices out
there. If they do not support extended dynamic allocation, in real network
they will never get 4 transmission timeslots allocated (because fixed
allocation method is not used by any live nework, and the whole feature has
been also removed from the newer specifications versions). So this is one
thing to check for those devices. More generally, any MS that supports
transmission on 3 or more timeslots for GSM/EDGE, really has the capability
to do so in real networks only if it supports extended dynamic allocation.
Of course, this is also the case with the network (not all networks support
it!).
For those interested, here's a small recap on multislot classes (can be
found from 3GPP TS 45.002 Annex B.1, see
http://www.3gpp.org/ftp/Specs/latest/Rel-5/45_series/45002-5d0.zip):
(I left out classes 13-29, which are irrelevant, as well as some other
parameters):
format: multislot class, max receive slots + max transmit slots = max. sum
of receive and transmit slots
1, 1+1 = 2
2, 2+1 = 3
3, 2+2 = 3
4, 3+1 = 4
5, 2+2 = 4
6, 3+2 = 4
7, 3+3 = 4
8, 4+1 = 5
9, 3+2 = 5
10, 4+2 = 5
11, 4+3 = 5
12, 4+4 = 5
(Rel-5 high multislot classes start here)
30, 5+1 = 6
31, 5+2 = 6
32, 5+3 = 6
33, 5+4 = 6
34, 5+5 = 6
35, 5+1 = 6
36, 5+2 = 6
37, 5+3 = 6
38, 5+4 = 6
39, 5+5 = 6
40, 6+1 = 7
41, 6+2 = 7
42, 6+3 = 7
43, 6+4 = 7
44, 6+5 = 7
45, 6+6 = 7
(note: classes 35 to 39 really are different from 31-34, even though the
amount of slots are the same)
Finally, disclaimer: it may be that e.g. the Nokia N93 (and/or other devices
aswell) has different versions (or variants) supporting different multislot
classes out there. This may be due to e.g. operator requests (I'm told e.g.
some operators do not want (/take) handheld devices with more than 2
transmission timeslot capability, due to e.g. power reduction applied in the
case where transmission occurs with 3 or more timeslots in the terminal)
So, it boils down to this:
- which is more important to you, uplink or downlink throughput? (typically
it's the downlink)
a) if it is the uplink, does your operator's network support extended
dynamic allocation, and does the device claiming support for e.g. multislot
class 11,12, or 32 (for example)?`
- does you operator support high multislot classes (to gain the extra
downlink timeslot for transfer)?
- how crowded is the network (or cell/cells you typically) use, so do you
have real-life possibility to gain from higher multislot class?
- how good are the radio conditions in the typical place of use (i.e. how
far are you typically from the basestation / how good is the signal level
and how high is the C/I)?
- does the operator support certain additional features, such as the (Rel-4
feature) Extended Uplink TBF mode (this is not to be confused with Exteded
Dynamic Allocation already mentioned)? (with this feature, short temporary
gaps in uplink transfer do not cause the release of the uplink connection;
this decreases the RTT for uplink transfer, e.g. TCP Ack's, which play an
important role in e.g. FTP download; some analysis suggests that overall
downlink download time for mid/large files over GPRS/EDGE can be sped up to
anything between 50 to 100 percent if the network supports extended uplink
TBF mode)
- all these factors (and possibly more, e.g. what are the operator-stored /
subscribed GPRS/EGPRS PDP context parameters) affect your real-life
experienced throughput speeds really will be
Regards,
Antti