🚀 go-pugleaf

To add some more information, the GSM timing advance is measured in
symbol-periods.

The GSM symbol rate for GMSK and 8PSK modulation is 156.25 symbols per burst
* 8 bursts per timeslot =frame * 26 frames per 26multiframe (120ms) 270833
1/3 symbols per second.

The TA is actually an 8-bit field, not a 6-bit field, however the maximum
allowable value for DCS1800 and GSM900 is 63 (00111111b).

At 270833+1/3symbols/sec, light (radio) would travel c/[270833+1/3]m, so a
value of 63 would result in a maximum two-way distance of 63c/[270833+1/3] =
~7x10km (70 kilometres)

The reason for the maximum value of TA is because transmit and receive
bursting is skewed 3 bursts (Tx 3bursts after Rx). Hence, this leave a gap
of 468.75 symbol periods, less the TA. The MS has to perform neighbour
measurements and synchronisation with whatever is left. 63 was chosen to be
a reasonable value.

Additionally, I believe the pseudo-synchronised handover scheme may have
influenced the maximum TA, although I'm not sure. (The psuedo-synchronised
handover scheme is the most complex handover procedure)

--Jeremy Quirke



"John Henderson" <jhenRemoveThis@bigpond.com> wrote in message
news:b334e9$1fo3ki$1@ID-83062.news.dfncis.de...
> "Pascal" wrote:
>
> > I am considering that option to gain reception in remote areas.
> > Anyone with on mounted on his car ?
>
> Yes.  I've tried various, and find a high-gain whip antenna the
> best.  But get one that's wired from phone to antenna - not an
> RF-coupled "through the glass" one.  A whip aerial is
> omnidirectional (if it's not physically obstructed).
>
> > What is the maximum gain I can expect ?
>
> > I have seen 7dB and if my memory is good from uni, it means 4
> > times better as it doubles every 3 decibels.
>
> I've tried various from 3 to about 5, and find the 5 excellent.
>
> Bear in mind that you're limited by Timing Advance (TA) in a GSM
> phone.  This is the compensation made for the delay caused by the
> speed of light, to ensure that the bursts transmitted by your
> phone get to the tower at exactly the right time for
> synchronisation with your allocated timeslot.
>
> It's a 6-bit field, so values range from 0 to 63.  At 553.5
> metres per step, that gives GSM an absolute maximum range of 35
> km.
>
> There's a workaround to this limitation, using pairs of timeslots
> in half-capacity cells.  Extended Range, and Enhanced Extended
> Range cells have maximum ranges of 72 and 121 km respectively.
> But they're rare.  In general, you might "see" cells further then
> 35 km away, but you won't be able to use them.
>
> > I have found some Omnidirectional antenna designed for both GSM
> > and GPS.
> > Is that a better alternative to the vertical antenna ?
>
> If these are the compact, stick-on, "phased array" types, I've
> found them reasonably good, but whip antennae better.
>
> John
>
>