Effect of HV1 Packets on Bluetooth Device Capacity
Original Post: HV1 packets using ALL Bluetooth capacity? (eGroups
Msg.) Date: 2000-08-22
If a master & slave are communicating using SCO HV1
packets, then all of the Bluetooth capacity is being used.
This conclusion comes from the fact that HV1 packets
have to be transmitted at every other packet , (and hence the return
duplex packets in between), i.e. a HV1 packet has to be sent every 2
slots.
Assume the master is sending voice data to a slave. The full
Bluetooth link capacity is used ,regardless of the slave device's
intervening TX slot (slave-to-master slot) .This intervening slot, and the
gap in voice data it causes, is handled by the fact that the time it take
to move from one slave's RX slot (master-to-slave) to another is 1.25ms
(2*625us) and each HV1 packet contains 1.25ms of voice.
No other slave can attempt to use the link as the
Master to Slave slot is directed at a single slave, so it is the only
slave that can respond.
In tests the Ericsson module switches to DV packets
if the ACL channel is doing data transmission when HV1 is requested for
voice. This can be explained by the fact that:
- All of the device's capacity is tied up servicing the HV1 link.
- When a new link has to be serviced , resources (capacity) has to be
taken away from the current voice link (HV1 packets), and given to the
new link.
Also by assumption, three simultaneous SCO links
must all use HV3 packets
It may be best to use a DM1 packet and drop the SCO
data if the Master wish to talk to other slaves, as the human ear will not
perceive 1.25ms of lost audio.
|
