Datasets:
kkboy1
/
LeAudio

Dataset card Data Studio Files
xet
 Community
1
text
stringlengths
1
112
dont need to be exposed i n production products silicon and stack vendors may provide
alternative apis to provide this functionality but understanding the hci commands helps to
demonstrate how the process works
parameter description
cigid assigned by the initiators host
sduintervalctop time interval between sdus generated from the
initiators host
sduintervalptoc time interval between sdus generated from the
acceptors host
worstcasesca worst case sleep clock accuracy of all of the
acceptors in the cig
packing preferred packing sequential vs interleaved
framing framed if set to otherwise up to the controller on a
percis basis
maxtransportlatencyctop the maximum transport latency for each direction
maxtransportlatencyptoc
ciscount the number of cises in this instance of this
command
cisidi unique id for each cis in this cig
maxsductopi the maximum sizes of sdus from the initiator s
and acceptors hosts maxsduptoci
chapter setting up unicast audio streams
parameter description
phyctopi bitfield indicating the possible phys to use for each
direction if more than one bit is set the controller decides phyptoci
rtnctopi the preferred number of retransmissions in each direction
the controll er can ignore this rtnptoci
table le set cig parameters hci parameters
in table the parameters highlighted in italics are recommendations to the controller
which it may choose to ignore
there are two values which are required in the le set cig parameters hci command table
which dont come from the ase configuration process the first is the
worstcasesca which is the worst case sleep clock accuracy of all of the acceptors the
initiator needs to determine this by requesting the current value of each devices sleep clock
accuracy before issuing the command normally by using the lerequestpeersca hci
command
the second is the packing parameter which determines whether the cis events will be sent
sequentially or be interleaved if set to they will be sequential if set to they will be
interleaved
normally the framing parameter value is set to so that the controller can dec ide but some
profiles such as hap may mandate that in some cases it be set to to force them to be
framed the recommended qos settings for unicast audio streams in table of bap state
that they should all be unframed apart from those with a kh z sampling frequency which
should be framed
at this stage the controller does not inform either the initiator or acceptors host of the
values it has chosen that information will be conveyed when the cises are established
instead the initiators hos t only gets confirmation that its requested configuration can be
scheduled along with connection handles for each of the cises
once the hcicommandcomplete event has been received confirming that the cises can
be scheduled the initiator should send th e ase control point operation command with the
opcode set to x to each of its acceptors in turn with the operation specific parameters
defined in ascs table and shown below in table this will move the ases to the
qos configured state the values that the initiator uses in the config qos command replicate
the parameter values that it used in its hci lesetcigparameters command
section configuring an ase and a cig
parameter value
number of ases i must be at least
cigid values used in the hci lesetcigparameters
command
note at this stage these may not be the values which
the controller has actually scheduled cisid i
sduinterval i
framing i
phy i
maxsdu i
retransmissionnumber i
maxtransportlatency i
presentationdelay i value which the acceptor will use for rendering or capture
table cis and presentation delay values used with x config qos opcode ascs table
the presentation delay is the only parameter which is not included in the hci le set cig
parameters command instead it is sent directly to the acceptor in almost all applications
the value of presentation delay will be identical for all of the sink ases the reason the
presentation delay value is the same for each sink ase is that this determines the point at
which audio will be rendered for earbuds and hearing aid s that would always be at the same
time in some situations such as where there are multiple speakers in a conference room or
auditorium there might be a requirement to assign different values to cope with latency related
to their relative position in the room but that is not the norm
if there are one or more source ases they are likely to use a different value of presentation
delay to the value for the sink ases not least because the overall lc encode time which is
part of the presentation dela y is significantly greater than the decode time around ms as
opposed to ms so more time is required to encompass that encoding however all source
ases would normally use the same value as each other
the presentation delay values chosen by the initiator for rendering should be within the
preferred presentation delay ranges exposed by the sink ases and the capturing values
within those exposed by the source ases the values should not extend beyond the common
range within the max and min value s that were exposed for each direction by the set of
acceptors and should ideally be within the range of preferred max and min values see table
if the context type is live it should work with the lowest common minimum value
as before the initiator sends these in a write without response command after which it
receives a notif ication of the updated ase control point characteristic confirming that each
ase has moved to the q os configured state followed by notifications of the updated ase
characteristic for each ase the parameters in these notifications reflect the values set in the
previous ase control point operation
chapter setting up unicast audio streams
parameter value
cigid values set by the initiator in its ase control point
operation
note at this stage these may not be the values which
the controller has actually scheduled cisid