Using nRF52811 for direction finding

RE: Using nRF52811 for direction finding

Simonr — Mon, 04 Nov 2019 13:42:57 GMT

Hi Fabrice

No, we at Nordic do not have a finished product with an antenna array at this moment. If you'd like more details on if and when this will be available, please contact your regional sales manager. If you don't know who this is, please fill out the sales related questions form here and they should get back to you with a detailed answer.

Best regards,

Simon

RE: Using nRF52811 for direction finding

bloobird0 — Mon, 04 Nov 2019 09:27:54 GMT

Hello,

what is the current status?`Is there any dev kit with antenna array that we could use to get into direction finding without having to make custom hardware for it?

RE: Using nRF52811 for direction finding

greentree — Sun, 07 Jul 2019 19:32:53 GMT

Hi Robwasab,

Thank you. As I see now, I missed “The IQ samples are recorded with respect to the RX carrier frequency” sentence in the spec and decided the sampling is made over demodulated signal without referring to the carrier. Strange, the developers did not point me that out. It is great if I was wrong. Hopefully PLL jitter will not make significant phase noise.

“I agree that this is not the cutting edge MUSIC algorithm”

MUSIC is a fundamental algorithm. It requires making distance between antennas not more the half of wavelength for unambiguous AOA detection. Max distance between antennas in array defines the accuracy of AOA estimate. We need many antennas if we need accurate AOA estimation with MUSIC. The more antennas, the more eigenvalues have to be calculated in the algorithm. Correspondingly, more resources and time for calculation are needed. It affects the system throughput quite well. Other methods would be more effective in low cost applications. For instance, methods based on disambiguation from one baseline to another one.

RE: Using nRF52811 for direction finding

robwasab — Sun, 07 Jul 2019 05:10:38 GMT

Hi Vlad,

"IQ sampling is made over demodulated signal having frequency of 250kHz or Since signal frequency is relatively low, samples taken in different antennas can be effectively synchronized."

No. While the baseband frequency is 250 kHz, the phase shifts seen by the antennas will be preserved from the baseband's perspective. If you do the math, a phase shift at RF will equate to the same phase shift at baseband. Dmitry was alluding to this earlier. See following:

"3. If question 1 and 2 are negative, what AOA estimation approach shall be used with nRF52811 ?"

Inferring from the product specification, section 6.14.12, or page 189, and that this is part of the Bluetooth specification, the transmitter appends a "continuous tone extension" to the end of a direction finding packet. The first part of the packet is a normal BLE packet, but then after the CRC, the transmitter continues transmitting the current bit it is on, which is a CW tone. It holds this tone for a duration configured by the receiver. The duration of the tone, according to the the details in the datasheet, consist of a guard time, and several switch slots, in that order. It is during the guard time that the receiver calibrates and locks its PLL, after which, disables the PLL feedback so the LO's phase can't change once we route another RF path from another antenna into the receiver - this is what is happening during the switch slots. The various RF paths from the different antennas are fed into the receiver, and the output of the IQ modulator will be indeed output phases equal to arctan(Q/I) relative to the antenna that was activated during the guard period.

Somebody please correct me if I am wrong.

Yes, I agree that this is not the cutting edge MUSIC algorithm. But I have yet to try their implementation to give it a thumbs up or thumbs down as to whether or not it can be used practically.

RE: Using nRF52811 for direction finding

greentree — Thu, 18 Apr 2019 13:15:25 GMT

Hi Dmitry,

Sorry for the late reply. The picture below is from here: https://web.wpi.edu/Pubs/E-project/Available/E-project-101012-211424/unrestricted/DirectionFindingPresentation.pdf . If you put real numbers in equations, you will get results I mentioned earlier.

RE: Using nRF52811 for direction finding

Simonr — Mon, 15 Apr 2019 06:03:52 GMT

Hi Vlad

Due to the Easter holiday you will have to expect a delay in answers, sorry for the inconvenience!

No, you can't. If you have private questions I suggest you create a Private ticket on the forum. This way the ticket will only be visible to you and Nordic engineers, and it will be treated with confidentiality.

Best regards and Happy Easter,

Simon

RE: Using nRF52811 for direction finding

greentree — Sat, 13 Apr 2019 22:13:58 GMT

Hi Simon,

I would like to ask you a question via email. Is that possible?

Best regards,

Vlad.

RE: Using nRF52811 for direction finding

Dmitry — Thu, 11 Apr 2019 12:36:47 GMT

Hi Vlad,

sampling is done on demodulated signal after I/Q mixers, so initial shift of RF carrier signal by, say, 30 degrees will give phase shift in sampled sequence by 30 degrees. Accuracy should be quite high.

RE: Using nRF52811 for direction finding

Simonr — Thu, 11 Apr 2019 08:21:56 GMT

No, it does not.

Best regards,
Simon

RE: Using nRF52811 for direction finding

greentree — Wed, 10 Apr 2019 14:27:52 GMT

Hi Simon,

I would like to clarify one more question to understand how it may work in a good way.

Does nRF52811 have a two channels receiver? Only "yes" or "not" would be enough.

Best regards,

Vlad.

RE: Using nRF52811 for direction finding

greentree — Wed, 10 Apr 2019 14:06:07 GMT

Thank you Simon,

Maybe somebody from DevZone tested that method using SDR?

Unfortunately, theoretical estimation of the method doesn't look encouraging. Assume we have 1 meter distance between antennas and measure phase difference between signals of 500kHz. If AOA will be changed from 0 to 30 degrees, corresponding phase difference will be changed from 0 to 0.3 degree... Real phase errors may be over 1 degree. Utilization of multiple baselines will slightly improve accuracy of measurements, but we need to increase frequency of the signal and distance between antennas significantly to achieve acceptable accuracy of the system. If we increase frequency, it goes out of Bluetooth conception. Even 1 meter of distance is a quite large size of antenna array for indoor applications.

I appreciate any comments. Maybe I missed something?

RE: Using nRF52811 for direction finding

Simonr — Wed, 10 Apr 2019 09:01:55 GMT

Hi Vlad

We currently don't have these details/measurements in place as this is work in progress.

Best regards,

Simon

RE: Using nRF52811 for direction finding

greentree — Tue, 09 Apr 2019 14:25:35 GMT

Hi Simon,

Thank you. I think I got the idea of Direction Finding implemented in nRF52811. IQ sampling is made over demodulated signal having frequency of 250kHz or 500kHz. Since signal frequency is relatively low, samples taken in different antennas can be effectively synchronized. Yes, we can get an estimate of AOA. The accuracy of that estimation will be low.

Many factors affects the accuracy of direction finding. SNR, multipath propagation and relative distance between antennas are some of the most powerful factors. Influence of SNR is quite understandable. Effect of multipath is more complicated and it may be much more powerful than effect of low SNR. All errors will be aggregated in error of estimated phase difference between signals received in two spaced antennas. After all, accuracy of angle estimation is mostly an accuracy of phase difference measurement multiplied by ratio of wavelength to a distance between antennas.

If signal frequency is 500kHz, the wavelength is 600 meters. If we have two antennas allocated only at 1 meter from each other, error of AOA estimation (at least when real AOA is 1 ... 3 degrees) is defined as 600 * phase difference error. Multipath may (and will) shift phase estimate to some value. Shift of AOA estimate due to that only reason will be very significant. That is why it is better to make measurements at the carrier signal. If carrier frequency is 2.4GHz, wavelength is 0.125m.

However, that is all a theory.

Did somebody make indoor phase measurements (for instance with SDR) when the signal frequency is 500kHz?

How the real life environment affects the measurements?

Best regards,

Vlad.

RE: Using nRF52811 for direction finding

Simonr — Tue, 09 Apr 2019 10:26:49 GMT

Hi Vlad

I got some input from one of our 52811 experts, and here is what he said regarding the initial inquiry:

BLE AoA is based on IQ sampling from different antennae sequentially with a single receiver chain. For this a single nRF52811 is connected to an external antenna switch and then to multiple antennas. A system as described in questions 1 and 2 is not something we have investigated. We recommend to use the AoA method described in the Radio - Direction Finding (6.14.12) chapter of the nRF52811 product specifications.

I hope this is a satisfactory answer, and if you have any follow-ups, please ask!

Best regards,

Simon

RE: Using nRF52811 for direction finding

greentree — Thu, 04 Apr 2019 12:24:36 GMT

Thank you Simon.

I asked that question because developers, not sales, make the future. However, you are right, that is not the place for such discussion. I appreciate your answers.

Best regards,

Vlad.

RE: Using nRF52811 for direction finding

Simonr — Thu, 04 Apr 2019 06:59:02 GMT

Here on DevZone we never speculate or talk about what might happen in the future, so this question is better suited for your Regional Sales Manager (RSM). If you need his contact information, let me know and I will help you get in touch.

Best regards,

Simon

PS: I will of course update you when I have answers on your initial inquiry.

RE: Using nRF52811 for direction finding

greentree — Wed, 03 Apr 2019 13:36:37 GMT

Hi Simon,

Thank you for the fast answer. However...

Information over the link you provided doesn't include explanation regarding how one receiver can be used for coherent signal processing in direction finding. We can try to estimate direction of arrival based on magnitude of IQ samples for every antenna in array, but it will be highly vulnerable from multipath propagation especially in indoor environment. So, such approach will be highly inaccurate and comparable with accuracy of RSSI based methods. Whereas indoor location based services are starving for high accuracy.

Does Nordic have plans for making a Bluetooth 5.1 chip with two coherent receivers, if answers for questions 1 and 2 are negative currently?

There is no need of full duplication of whole current chip functions. Only second receiver working with the same LO and synchronized IQ sampling in both receivers are needed.

Best regards,

Vlad.

RE: Using nRF52811 for direction finding

Simonr — Wed, 03 Apr 2019 11:31:07 GMT

Sorry for the late reply, but I have not been able to find the answer to questions 1 and 2, so I am waiting for an answer from our experts on the matter.

In the meantime, I can give you an answer to question 3. You will not need to use two separate nRF52811 systems, as one should be sufficient for very accurate position finding. The nRF52811 antenna will be an antenna array, which allows it to find the AoA and AoD accurately without two separate receivers. So you will not need a specific approach to use the position finding feature. Please read more on how Nordic will implement Bluetooth 5.1 here.

Hopefully I'll be able to come back to you on questions 1 and 2 soon.

Best regards,

Simon