Datasets:
kkboy1
/
LeAudio

Dataset card Data Studio Files
xet
 Community
1
text
stringlengths
1
112
config urations on airtime and late ncy for bidirectional streams
chapter setting up unicast audio streams
qos configuration airtime
stereo latency
pd ms outbound inband
ms
ms
ms
table airtime and latency for various bidirectional qos configurations
it means that there i s a trade off between airtime qos and call setup time which designers
need to be aware of making this decision will usually depend o n other airtime resource
demands in the initiator it illustrate s the flexibility of bluetooth le audio but highlights the
fact that implementors need to understand more of the overall system than they did with
classic bluetooth audio profiles
whos in charge
one of the questions for designers is deciding who is in charge the development of
bluetooth le audio has seen some strong opinions from phone manufacturers who feel
that the phone is responsible for everything but also from speaker and hea rable
manufacturers who feel that more autonomy needs to b e given to their products leading to
the concept of the sink led journey the specifications give room for both but they need
developers to be aware of some of the consequences
its useful to lo ok at a simple example which is a pair of earbuds each of which ha s a
microphone a phone manufacturer might want to use both as that would allow them to
process both audio streams which should result in a better voice signal o n the other hand
earbud manufacturers might prefer to use only one conserving the battery life of the other
earbud they might even want to be able to swap which is being used during the course of
a call if they see the battery of the one with the active microphone startin g to decline
the question is how to enable these options assuming there is communication between
the earbuds one of them could consider not exposing its source ase although thats a
brute force approach especially as the phone could infer its presen ce by reading the pac
records as we saw above an acceptor should expose an ase for every audio stream it is
able to support even if it may not be able to support it at all points in time a far better
method is to indicate that the server is not avail able to transmit audio by using the
availableaudiocontexts characteristic in pacs with the availablesourcecontexts set
to x which is one of the few occasion s where you are allowed to set all context type
bits to zero
if the initiator phone can see that each earbud has an available source ase then it can
effectively take charge it may decide to select just one not least because processing two
incoming streams which are not time aligned is not trivial and increases its power
consumption it can use other information such as checking the battery status of each
section whos in charge
earbud to guide its choice of microphone if it only wants to use one
if its more intelligent it could look to see if phrases are regularly repeated during the call
infer that implies poor signal to noise and add in the second microphone to try to improve
the quality by gaining a few dbs of signal equally a phone app could log the duration of
calls with specific people and from that history deduce the likely length of the call the
earbud battery life and use that to inform its decision of how many audio streams to set up
and the qos settings that would let the battery survive through the anticipated call duration
so much opportunity for differentiation i suspect both o f these approaches are unlikely in
the short term as the phone will continue to consider the earbuds as a resource to use as it
likes
on the other hand supporters of the sink led journey would want more decision making
ability in the earbuds however that implies an ability for them to communicate with each
other which is currently out of scope other than the preset local synchronisation in haps
if the pair decide that they only want to use the microphone on one earbud they can decide
between the m which will be the one to take the battery hit and the other can set its
availablesourcecontexts set to x they can do this while theyre still in the battery
box so this doesnt have to use a separate sub ghz radio link
this approach has the advantage that the phone knows that the source ase is there for
both so it can configure a stream in the cis it cant enable it for the earbud showing
availablesourcecontexts as x because the earbud will re ject it at the config codec
stage however the initiator can schedule it as it can put whatever it wants in the hci le
set cig parameters command although weve seen that the process normally uses
information that the initiator receives from an accept or it can use cached or assumed
values for missing acceptors or ases that means that if there is a need to swap
microphones at some point that can happen otherwise the initiator would need to tear
down the cig and rebuild it resulting in a break in the call that shouldnt terminate the
call as the loss of a stream doesnt cause any change in the tbs state machine but its not a
good user experience
the point here is that the pacs and ascs characteristics allow a surprising amount of
ownership regarding how an audio stream is set up an initiator can act unilaterally but
that risks affecting how well devices work for the best performance and user experience
designers need to understand the options offered and how to use them
ooo
that concludes the methods for setting up unicast audio streams so we can now turn our
attention to broadcast
chapter setting up and using broadcast audio streams
chapter setting up and using broadcast audio streams
in this chapter well look at how to configure broadcast audio streams broadcast is the major
new feature of bluetooth le audio and has the potential to change the way that everybody
uses audio the exciting use cases it introduces include the ability to share audio along with
the ability to set up ad hoc private connections the latter is enabled by the broadcast
assistant features which are defined in bass and bring totally new control and user
experiences
once again the main specif ication is bap the basic audio profile but we now need to add
another one of the gaf specifications called bass the broadcast audio scan service bass
defines how an additional device can be used to help find broa dcast audio streams and
instruct one or more acceptors to synchronise with them
cap once again comes into play as well as providing procedures to set up and receive
broadcast streams it also defines the role of a commander a commander can be a phys ical
device or an application on a phone or a tv well look at the commander role in more
detail once weve gone through the basics of sending and receiving broadcast audio streams
caps main task is to lay down rules about associating context type s and content control
ids and repeat the order in which things need to be done
well start with the basics of setting up and receiving a broadcast audio stream then look at
what the commander brings to the picture in chapter well delve further into the
possibilities within broadcast audio examining some of the new use cases that it can enable
although the broadcast topology originated with a desire to improve the quality provided by
inductive hearing loops bluetooth le audio provides far more versatil ity than an inductive