Controlling the Media Clock

[jade23]

Hi,

for my bachelor's thesis I want to use two i.MX6UL Eval Kits to implement a demonstrator for low-latency streaming over IP using the AES67 standard. The eval boards are supposed to be the the audio sinks (receivers). I want to connect speakers to the analog audio out. The receivers need to be tightly synchronized with a PTP master clock which in this case is also the audio source (transmitter). I already know that the i.MX6UL supports IEEE1588 hardware time-stamping.

Is it possible to take control over the clock (Media Clock) that drives the audio codec using the eval boards? I would like to use the network clock (PHC) to drive the audio clock. That way I hope to achieve highly synchronized audio output. Any information is very welcome.

Please bear with me if this not the right place to ask.

Regards

Jan

#audio_codec #audio_pll #audio_transmission streaming #ptp master clock external clock #ieee1588‌ #aes‌ #media_clock‌

[gusarambula]

Hello Jan Deinhard,

The audio pll generates the Audio clock so you may not use the PHC for the regular operation of the audio codec on the board, as far as I know.

I am not familiar with AES67 but what you are trying to do sounds a lot like AVB (Audio Video Bridging). The i.MX6UL is not supported on the AVB stack that is available from NXP (link). However, it does support IEEE1588 on HW and you should be able to do the streaming control in SW.

http://www.nxp.com/products/software-and-tools/run-time-software/professional-services-software-tech...

Rather than controlling the media clock that drives the codec the idea behind AVB is controlling the time at which the samples will be reproduced.

I hope this helps!

Regards,

[jade23]

Hello,

thank you very much for your input. AES67 is indeed related to AVB. Though both operate on different network layers to solve a similar problem. AES67 is a standard for audio-over-IP interoperability. It tries to find a common denominator for audio networks like Dante and Ravenna.

I would like to be able to use the PHC as the reference for the audio PLL.

A software solution is what I want to avoid because that will involve resampling AFAIK. Sample rate conversion add latency and will deteriorate the audio signal.

Regards

Jan