How does a bluetooth headset work near another one?

Archived from groups: alt.cellular.bluetooth (More info?)

P T Wang wrote:
> Hello,
>
> What happens if two bluetooth headsets are in range of the base station
> (e.g. a cell phone)? Will both headsets be able to talk? Or, is there an
> association/authentication between the headset and the base station?
>
>
>

Usually the two headsets will be connected to different cellphones. The cellphones act as
the master for the network, headset as slave.

Each network has a different frequency hopping routine and each mini-network or piconet as
it is termed, has a different access code.

So if headset came into range of another it would not be aware of it, and because of
frequency hoppping (at 1600 times every second) the two networks would collide

eg

piconet A communicates on example channel sequence:-

10 48 34 27 58 1 9 30 18 52 14 25.......

and piconet B communicates on example channel sequence:-

24 61 8 19 37 41 20 4 38 19 14 71.......

Once a slave is connected to it's master it can predict the hopping sequence and thus know
the next channel to hop to,

If two channels communicate on the same freq (see example above for hop 11, ch 14) then it
would not receive any valid data be cause the transmissions interfere with each other, In
this case the network hop to the next frequncy and retransmit the previuos packet...

the Bluetooth specification version 1.2 has Adaptive frequncy hopping so if, for example,
the piconet is in range of a wireless LAN network they will interfere with each other also.

However the lan equpment only uses a single channel in the entire range. The bluetooth
units in the piconet will notice they get interference every time they hop to this
particular channel and will negotiate that if they hop to this channel in the future they
will ignore it and just hop again.

pretty clever huh...

I'ts all to minimise interference and maximise the number of networks that can coexist in
the same physical space.

hope this helps

Rob

Archived from groups: alt.cellular.bluetooth (More info?)

"P T Wang" <news1003@paultwang.com> wrote:

>What happens if two bluetooth headsets are in range of the base station
>(e.g. a cell phone)? Will both headsets be able to talk? Or, is there an
>association/authentication between the headset and the base station?
>

They are supposed to work.

However there ARE issues with bluetooth. I have a Nokia 6310i, a Tosh
E800 PDA, some D-link bluetooth-USB adaptors, a socketcom bluetooth
GPS receiver, and basically the stuff doesn't work at all reliably if
the bluetooth device one is talking to is a lot further away than
another (presently not used but still "online") bluetooth device.

There are so many "issues" with bluetooth, which some people get and
some people don't get, that I think BT should be regarded as a point
to point comms solution for a specific purpose, not a peer to peer
local area network which it is advertised as...


Peter.
--
Return address is invalid to help stop junk mail.
E-mail replies to peter4400@peter2000XY.co.uk but remove the X and the Y.
Please do NOT copy usenet posts to email - it is NOT necessary.

Archived from groups: alt.cellular.bluetooth (More info?)

On Tue, 22 Jun 2004 09:29:54 +0100, Rob Shepherd wrote:

> Usually the two headsets will be connected to different cellphones. The cellphones act as
> the master for the network, headset as slave.
>
> Each network has a different frequency hopping routine and each mini-network or piconet as
> it is termed, has a different access code.
>
> So if headset came into range of another it would not be aware of it, and because of
> frequency hoppping (at 1600 times every second) the two networks would collide
>
> eg
>
> piconet A communicates on example channel sequence:-
>
> 10 48 34 27 58 1 9 30 18 52 14 25.......
>
> and piconet B communicates on example channel sequence:-
>
> 24 61 8 19 37 41 20 4 38 19 14 71.......
>
> Once a slave is connected to it's master it can predict the hopping sequence and thus know
> the next channel to hop to,
>
> If two channels communicate on the same freq (see example above for hop 11, ch 14) then it
> would not receive any valid data be cause the transmissions interfere with each other, In
> this case the network hop to the next frequncy and retransmit the previuos packet...
>
> the Bluetooth specification version 1.2 has Adaptive frequncy hopping so if, for example,
> the piconet is in range of a wireless LAN network they will interfere with each other also.
>
> However the lan equpment only uses a single channel in the entire range. The bluetooth
> units in the piconet will notice they get interference every time they hop to this
> particular channel and will negotiate that if they hop to this channel in the future they
> will ignore it and just hop again.
>
> pretty clever huh...
>
> I'ts all to minimise interference and maximise the number of networks that can coexist in
> the same physical space.

What actually happens if two established Bluetooth piconets come into the
range of each other, which are not LAN equipment.

Say person-A is using a BT headset and paired phone at for example a
railway station, and person-B is using the same, and person-B walks into
range of person-A. Would the BT headsets and paired phones resequence
themselves to avoid collisions ?

--
Michael Turner

Email (ROT13)

zvxr.gheare1963@ivetva.arg

Archived from groups: alt.cellular.bluetooth (More info?)

Henryk Plötz <henryk.nospamplease@ploetzli.ch> wrote:

>> Say person-A is using a BT headset and paired phone at for example a
>> railway station, and person-B is using the same, and person-B walks
>> into range of person-A. Would the BT headsets and paired phones
>> resequence themselves to avoid collisions ?
>
>No, they don't need to. The pseudorandom hopping sequence is determinded
>by the hardware addresse of the master, which means there is only a
>_really_ small chance that two piconets use the same hopping sequence.
>And even if they did, they would surely be at two different positions in
>the sequence.
>
>So there is a chance of 1 to 2^28 for two piconets to have the same
>hopping sequence times a chance of 1 to 2^27 for two piconets with the
>same hopping sequence to be at the same position in it.
>
>All taken together the probability for two different piconets to follow
>exactly the same sequence is pretty close to non-existent, so all there
>is is a chance of about 1 to 79 (or 23 in France) for two piconets to
>collide _in one timeslot_ only. Then after 625us both will use a
>different frequency and the packet is retransmitted.

In practice I can demonstrate anytime that there are serious
interference problems with BT devices.

These could be due to faulty protocol implementation, or whatever, but
the bottom line is that the problems do exist.

I don't know about headsets but making a pile of devices like GPS
receivers, BT-USB adapters, BT access points, BT PDAs, communicate
concurrently is very unreliable.


Peter.
--
Return address is invalid to help stop junk mail.
E-mail replies to peter4400@peter2000XY.co.uk but remove the X and the Y.
Please do NOT copy usenet posts to email - it is NOT necessary.

Archived from groups: alt.cellular.bluetooth (More info?)

On Wed, 23 Jun 2004 09:19:10 +0100, Peter wrote:

> I don't know about headsets but making a pile of devices like GPS
> receivers, BT-USB adapters, BT access points, BT PDAs, communicate
> concurrently is very unreliable.

I suppose it's not so much an issue with BT headsets, given their typical
usage. Which usually no more than a couple of feet from the paired
phone. So unless another BT headset user stood or sat right next to the
first headset user, the BT data packets wouldn't get stomped-on if
collisions did occur.

--
Michael Turner

Email (ROT13)

zvxr.gheare1963@ivetva.arg