1. Home
  2. Embedded Programming
  3. AVR & RFM70 based wireless guitar system - two RFM70 in parallel to increase the bandwidth?

AVR & RFM70 based wireless guitar system - two RFM70 in parallel to increase the bandwidth?

Hi,

I've just finished my RFM70 based wireless guitar system

(I hope to publish it soon, but code and documentation needs some polishing - the system is designed to be very cheap - just CS5343/4 as ADC, ATmega88 and RFM70 in the transmitter, and RFM70 with ATmega32u4 in the receiver. The system is visible as USB MIDI/audio device so you can connect it thorough "alsa_in" to "jakcd" in Linux and then to your virtual guitar amp/ soubd processor - e.g. rakarrack or guitarix. Additionally the transmitter contains 4 switches and

4 potentiometers, which are mapped to MIDI controls)

The system works acceptably with 24bit/48kHz without acknowledge/ retransmission. Anyway some packets are dropped and this increases the noise level, so probably the acknowledge/retransmission will be needed. Other users reported that 500 kb/s is achievable with ACK.

I think, that I'll need to use two RFM70 in parallel in the Tx and in Rx to provide the bandwidth sufficient for may needs (OK. maybe I'll reduce the resolution to 20bits, which gives 960kb/s).

My questions are. Is it reasonable to expect, that two RFM70 will work correctly together at distance of ca. 10 cm? How should I allocate the frequency to them? I have 83 1MHz channels. At 2mb/s air bitrate each RFM70 occupies

2MHz. Should it be OK to separate both channels by 4MHz or it is better to assure the distance of ca. 40 MHz? (But then I'll more significantly spoil the 2.4MHz band in my viccinity ;-) ). What do you think about it? Any hints are appreciated.

-- TIA & Regards, Wojtek

Reply to
wzab
Loading thread data ...

No sure how much you can adjust these modules, but you will have serious interferences with two transmitting at the same time. For higher data rate, you might need higher carrier frequency (i.e. 5GHz).

Reply to
linnix

Hmmm, in fact I'm rather afraid if one module will be able to receive the acknowledge, when the another one is transmitting. I'm afraid, that the input RF amplifier may get saturated by the signal sent by the neighbouring module, even though it is tuned to another freuency... The big advantage of RFM70 is their good availability and very good price. It can be done at 2.4GHz, as Nordic offers nRF24Z1 (

formatting link
= ) and AUREL offers the audio links 16b/44.1 kHz (
formatting link
formatting link
) working in this band. Unfortunately I'm not able to buy the nRFZ24Z1 breakout board in Poland and AUREL's modules are quite expensive (and you need to solder wires to the chip pins to extract the signal in a digital I2S form from them).

Reply to
wzab

Yes, you need some settling delays, which would cut into your data rate further. You need to either compress your data or go to the 5GHz ISM band.

Reply to
linnix

Diversity receive is the usual solution to this sort of problem. You might contact the manufacturer and see if their protocol can be adapted to it.

Reply to
Jim Stewart

Thanks. For now I have simply reduced the sampling rate by factor of two (quick & dirty solution - simply discarding every second sample without antialiasing filtering, but my source is supposed not to contain signals above 12 kHz), and the resolution to 20 bits. Now it is possible to send the bitstream via RFM70 with autoacknowledge/retransmission.

The noise level has decreased significantly because no packets are lost.

Reply to
wzab

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.