• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 30 subscribers
  • Views 814 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

CC8531 ganged masters feasibility

Michael Conroy
Prodigy 135 points
Other Parts Discussed in Thread: CC8531

I have an application in which I'd like 16 slave units to connect and stream audio to 4 masters.  They must do this in an environment where there's likely to be a fair bit of WiFi interference (auditorium).  Similar posts in this forum lead me to believe that 3-5 masters is possible, but not in a noisy environment.  Here are my questions:

1. Can I dynamically define what frequency constitutes a channel?  If there's enough room in the 2.4 GHz band, I should be able to offset those frequencies so each master operates in its own 18-channel space without risk of interference from the other masters.

2. Can I dynamically alter the hop sequence? 

3. If I can dynamically define the hop frequencies and sequence, I will need to tweak the pairing algorithm to ensure a slave gets an appropriate assignment on a well-defined channel (similar to a "listening socket" in TCP/IP).  Do I have the ability to tweak the firmware to make this happen, or can I make this happen simply micromanaging the device through a host controller?

Finally, I would like to take the I2S stream from each master and do some mixing and light signal processing.  Does TI have a recommendation for a Swiss army chip, or am I better off mixing separately and allocating a single DSP to each 8531? 

Regards,

Mike

  • 0
    Prodigy 25 points

    Hi Mike,

    No answer by TI .. for what ever reasons....??

    My guess would be:

    1. No, you cannot define dynamic channels because the range is fixed in silicon. I.e. there is a 1MHz (or what ever pre-defined bandwidth) window built into silicon. You may dynamically re-assign channels by som software algorithm, but the bandwidths are fixed. IMHO. 

    2. Hop sequence: Maybe you can alter, but I suspect it will be an implementation nightmare!

    3. Not sure if TI will give you the firmware source code. Best to assume not and develop your own. Again it may be

    an implementation nightmare.

    4. Mixing the audio streams on TLV320AIC devices may be possible -- for two streams, by first converting the I2S out of CC8531 to audio (via yet other TLV320AICs) then feeding input to mixing-TLV320AIC. Too complex a system, too many A2D, and D2A's introducing too many opportunities for noise/distortion. You are probably better off using a commercial mixer.

    As I stated above, the answers are my $0.02 worth guesses... hope TI can provide better answers.. 

    HK

  • 0 in reply to Hemanth Kanekal
    TI__Mastermind 21125 points

    Mike and Hemanth, 

    Hemanth is not far off with his response. Thanks for stepping in! 

    1. Not possible. You need to alter the masters to have a sub-set of the 18 channels (minimum 6) spread across the band or it might actually be that they would behave better just letting the AFH algorithms figure out the channel usage. That being said, sounds like you are pushing the boundaries here with 4 maximum PPW networks in an environment that has other interferes. If you on top require minimum latency etc. this might not be doable (to many audio drop-outs).

    2. Not possible due to #3

    3. We do not release the source code and this can't be done over the external host interface.

    4. To many channels etc to have a easy and simple solution. DSP with enough inputs sounds like the best choice or a commercial mixer. C553x series from TI comes to mind

    REgards, 
    Kjetil

  • 0 in reply to Kjetil
    Prodigy 135 points

    Hemanth and Kjetil,


    Thank you for your responses.  I'm disappointed that TI doesn't have an alternative offering...  PPW seems like such a time-saver for development and all of the robustness features are things I would've wanted to bake into the system myself had I not stumbled across this chip first.  That said, I may be able to get away with 8 channels.  Do either of you have any experience running 8 concurrent channels with success?  Anything less than that is just too constraining and I need another solution.  I'm open to suggestions there as well. 

    Best,
    Mike