Can the NRF52 output an unmodulated continuous tone at a specific channel?

[ray]

Hi all,

I'm a grad student trying to innovate on the BLE concept. I require a continuous signal as a source: preferably unmodulated, but a long duty cycle could also work. Would the NRF52 (or any other Nordic bluetooth device) be able to accomplish this?

Thank you, Ray

[ray]

There's recently several publications on using BLE in microwatts applications via backscattering. The idea is to incident a CW signal at a BLE beacon target, and modulate the CW signal on target-side to reduce the need to generate another carrier tone. The caveat is that the TX and RX has to be both enabled on the source-side.

[Øyvind Karlsen]

Hi,

This sounds extremely interesting, really cool thesis! I have found a paper that has done what you are trying to do using a nRF51822 chip that I think you will find very helpful.

Ensworth, Joshua F., and Matthew S. Reynolds. "Every smart phone is a backscatter reader: Modulated backscatter compatibility with bluetooth 4.0 low energy (ble) devices." 2015 IEEE International Conference on RFID (RFID). IEEE, 2015.

If I'm understanding you correctly that means that you are trying to compress the method described in this paper into two nodes?

For a functional test I think all three nodes described in this text can be realized partly by a nRF5 chip. CW output from the chip should be faily straight forward. You ccould also use it to generate the output frequency for the backscatterer. Finally it is already proven as a competent BLE receviver.

[ray]

Thank you! It's one part of a thesis focused on enabling applications in energy harvesting. And you are correct; a two-node solution with existing Bluetooth infrastructure (like in the Ensworth paper) would be a very desirable feature. In the worst case, my current plan is to use an NRF8001 as the CW source, and NRF52 as the receiver.

Would you know what Bluetooth chipsets are used for streaming audio? Those must be full-duplex, right?

[Øyvind Karlsen]

You could use a nRF52 for the CW source as well, anything the 8001 can do, the nRF52 can do as well. I am still not entirely sure why you want a full-duplex radio, the way I see it from the Ensworth paper there are no simultaneous receive and transmit operations. Please refer to the first image in this article, the difference in the article is that the backscatter node also alters the backscattered signal by a FET switch. I might very well be wrong here though, as I haven't had the time to properly research this topic.

It is very rare to use full-duplex as it requires expensive components in the antenna structure, ie. ensuring that the transmit does not self-interfere with the receive.

What is usually done is to time divide the transmit and receive and take turns, which is a form of half-duplex. I guess you are familiar with TDMA, this is the same thing.