Understanding Timer Synchronization with BLE, using LE Audio Features with Isochronous Channels?

RE: Understanding Timer Synchronization with BLE, using LE Audio Features with Isochronous Channels?

Amanda Hsieh — Tue, 30 Jul 2024 12:40:38 GMT

Hi,

The isochronous events are scheduled independently of the normal BLE Links (ACL links), either before or after on a timeline. However, each CIS needs a corresponding ACL link. When the ACL link is disconnected, the corresponding CIS(es) are also disconnected. As you have already understood, this will limit the maximum throughput

To learn more about isochronous channels, here is a list of useful references:

The Bluetooth Core_5.2 feature overview document: Bluetooth® Core Specification Version 5.2 Feature Overview | Bluetooth® Technology Website
The Bluetooth Core specification in itself. Especially Vol 6, Part B, Section 4.5.13 and 4.5.14
The LE Audio book: Introducing Bluetooth® LE Audio | Bluetooth® Technology Website, chapter 4.

-Amanda H.

RE: Understanding Timer Synchronization with BLE, using LE Audio Features with Isochronous Channels?

dmk_weber — Mon, 29 Jul 2024 08:50:47 GMT

Thanks for your reply Amanda.

I still don't really get how the ISO channels are transmitted on the link layer. I'm totally aware of the concepts of Connection Intervals and Connection Events for the normal BLE link. I also found some information about the CIG/CIG and the ISO interval....

But when does the ISO interval happen when there is also a normal BLE link from a link layer point of view? I'm assuming that it's the same radio, so there is nothing going on in parallel... Is the ISO interval after the "normal" Connection Events within a Connection Interval? I'm asking this because the ISO channels might then affect throughput, when there is a lot of data to transmit within the Connection Events... (similar to the Timer Synchronization example from Nordic) Or are the ISO channels transmitted in a "normal" Connection Event?

Thanks!

RE: Understanding Timer Synchronization with BLE, using LE Audio Features with Isochronous Channels?

Amanda Hsieh — Fri, 26 Jul 2024 12:56:47 GMT

Hi,

[quote user=""]How is this done on the BLE link layer then?[/quote]

The short explanation is that incoming packets are timestamped. The timestamps are forwarded to the application as part of the isochronous data. The timestamps are used to set up timed triggers.

[quote user=""] Is data transmission on the Isochronous Channels only performed, when there's no other packet transfer?[/quote]

The time-sensitive data is transferred over the ISO channels. Other data (ATT) can be transferred over the normal Bluetooth link. When using CIS there is always an ACL connection + an ISO connection.

[quote user=""]Would this LE Audio approach of time synchronization als work in our setup, where we have to transmit sensor data alongside with synchronization / time data?[/quote]

The data that is sent over ISO channels has corresponding time data. Data that is sent over regular ACL Bluetooth Connections does not have synchronization/time data.

[quote user=""]Is there any other standardized time synchronization BLE service to date?[/quote]

No, this Bluetooth spec-defined way.

[quote user=""]What's the best way to perform time synchronization over BLE?[/quote]

Bluetooth: ISO time synchronization demonstrates how to perform time synchronization over Bluetooth.

However, this question also depends on application requirements:

How accurate do you want time synchronization to be?
How often do you want to perform time synchronization?
Is the timed/synchronized data flushable or not?
Does the device network consist of Nordic devices only?

I hope these answers allow you to make a well-informed decision on how to proceed.

Regards,
Amanda H.

RE: Understanding Timer Synchronization with BLE, using LE Audio Features with Isochronous Channels?

Amanda Hsieh — Thu, 25 Jul 2024 19:01:57 GMT

Hi,

We are currently understaffed due to the summer vacation period, so delayed replies must be expected. I am sorry about any inconvenience this might cause.

Regards,
Amanda H.