0
\$\begingroup\$

I am thinking about a device which is able to receive data from my computer (how, is currently not of interest for this question), and then transmit it via rf to multiple receivers. The receivers should obtain the signal approximately at the same time (approximately: several microseconds delay are acceptable, several tens rather not, if possible). The challenging part is that the data for the receiver is not identical, each of them should get different data. Data is a soundfile, length ~5 minutes.
What is the best way to transfer the data, if the receivers should be as cheap and small as possible? Should I rather operate at different frequencies, or multiplex the data, and the receiver have to find their own signal? Which ICs are suited best for that application?

\$\endgroup\$
  • \$\begingroup\$ Different carriers. Different channels. TDM. Send everything to all but get each to use there bit of the data. It's pretty much impossible to say from the details given. How much data? How fast? What format? How many receivers? What processing power have the receivers? etc. \$\endgroup\$ – Tom Carpenter Feb 27 '16 at 21:14
  • \$\begingroup\$ Data as stated: ~1 Mbyte. Processing power of the receivers: Not determined yet, but as cheap and simple as possible. How many receivers: ~6-8. Speed: >60 kbit/sec, if possible. \$\endgroup\$ – arc_lupus Feb 27 '16 at 21:16
  • \$\begingroup\$ And you might want to rethink your simultaneity requirement. 10's of microseconds simultaneity is unacceptable for audio? Really? Could you please justify this? \$\endgroup\$ – WhatRoughBeast Feb 27 '16 at 21:19
  • \$\begingroup\$ It should arrive at the receivers approximately at the same time in order not to let it be hearable that there are different transmissions. @WhatRoughBeast: If the requirement is too narrow, what delay would you suggest as acceptable? \$\endgroup\$ – arc_lupus Feb 27 '16 at 21:22
  • \$\begingroup\$ I'd need to understand the system in much more detail to set numbers. How far apart are the receivers, and what is the judgement basis for unacceptable performance? What exactly does "not to let it be hearable that there are different transmissions" mean? How would the judgement be made? And that's just a start. \$\endgroup\$ – WhatRoughBeast Feb 27 '16 at 21:29
2
\$\begingroup\$

A TDM (time-division multiplexing) approach would be the most straightforward method to implement. The transmitter would send out frames of information, each containing one audio sample for each receiver. The frame rate would equal the audio sample rate.

--+-----+------+------+------+------+------+------+------+------+-----+--
  |frame| ch 0 | ch 1 | ch 2 | ch 3 | ch 4 | ch 5 | ch 6 | ch 7 |frame|
--+-----+------+------+------+------+------+------+------+------+-----+--

Each receiver would synchronize to the frame flag and then pull out its assigned data. It would transfer that data to its DAC at the next frame flag, simultaneously with all of the other receivers. The only timing skew would be the difference in radio propagation delay to each of the receivers; less than 1 µs if they're all within 300 meters of each other.

A data rate of 60 kbps (I assume that this is per receiver) suggests a rather low audio quality; e.g., 8 bits per sample at 8 kHz sample rate, which is comparable to "telephone grade". This requires the transmitter to operate at about 500 kbps.

If you want to produce something more like "broadcast quality" or "CD quality" audio, you're going to want at least 16 bits per channel at a sample rate of 32 to 48 kHz. This would be 512-768 kbps per receiver, or an aggregate rate of 4-6 Mbps at the transmitter.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.