IWR1843: About the noise floor slope

user4825004

Part Number: IWR1843

Hello,
We checked the A scope with the FMCW radar equipped with IWR1843.
The noise floor rises at short distances.
I found a question about this phenomenon in E2E.
e2e.ti.com/.../iwr1443boost-slope-noise-floor
The following three answers were given regarding the cause.
[The slope is indicating you get an Rx reflection close to the Tx and Rx , the frame-configuration-chirp is not correct, the radar power supply is noisy (AC ripple and noise comparison measurements to EVM), or possibly that the BSS and MSS software needed should use the latest version of the mmwave SDK compatible with your device.]
When we investigated, power supply noise and software were not candidates.
'an Rx reflection close to the Tx and Rx' is candidates.
Is this phenomenon due to the reflections that occur inside the IWR1843 device?
Are there any countermeasures?

over 3 years ago

0 Alec May over 3 years ago

TI__Genius 17265 points

Hi,

Could you provide more context as to what your setup look like (in terms of what the radar is looking at when testing) and how you are measuring the noise floor in this situation?

Best Regards,
Alec

0 user4825004 over 3 years ago in reply to Alec May

Intellectual 590 points

Hi,
Thank you for your support.
We are measuring without a radar target.
Specifically, the radar is surrounded by a radio wave absorber.
The measurement result is attached.

The horizontal axis is m and the vertical axis is dB.

Best Regards,

0 Alec May over 3 years ago in reply to user4825004

TI__Genius 17265 points

Hi,

Could you share your chirp configuration?

Best Regards,
Alec

0 user4825004 over 3 years ago in reply to Alec May

Intellectual 590 points

Hi, Alec

(1) The chirp configuration is as follows.

(2) Is this a known symptom?

Best Regards,

0 Alec May over 3 years ago in reply to user4825004

TI__Genius 17265 points

Hi,

I just tested this on my end using the mmwave demo visualizer, and achieved the following noise plot

As you can see, it is much more typical than what you have shown, which is why I am trying to understand the steps you took in order to produce your plot.

Best Regards,
Alec

0 user4825004 over 3 years ago in reply to Alec May

Intellectual 590 points

Hi, Alec
Thank you for testing.
I'm curious what is the difference between us.
Is this the result at speed 0km/h?
Best Regards,

0 Alec May over 3 years ago in reply to user4825004

TI__Genius 17265 points

Hi,

Below is a screenshot from the mmWave SDK Out of Box Demo Doxygen, which can be found at <mmwave_sdk_03_05_00_04>/packages/ti/demo/xwr68xx/mmw/docs/doxygen/html/index.html#tlv3

Best Regards,
Alec

0 user4825004 over 3 years ago in reply to Alec May

Intellectual 590 points

Hi, Alec
Thank you for your reply.
Our A-scope is 0km/h.
Other speeds were fine.
I don't see the range profile in your tested plot.
Could you show me your range profile?
Best Regards,

0 Alec May over 3 years ago in reply to user4825004

TI__Genius 17265 points

Hi,

Here is a plot of both

It's worth noting that this is NOT in an anechoic chamber, so the peak at ~1.7 meters is actually the ceiling of the room which I am currently in. I would also like to point out that this plot is displayed in the mmWave Demo Visualizer here: https://dev.ti.com/gallery/view/mmwave/mmWave_Demo_Visualizer/ver/3.6.0/ with the Out Of Box Demo from the mmWave SDK in case you wanted to try this method yourself.

Best Regards,
Alec

0 user4825004 over 3 years ago in reply to Alec May

Intellectual 590 points

Hi, Alec
Thanks for the plot.
Q.1
Do you have any range profile acquired in absence of any targets?
For example, the data will do if it was acquired when the radar antenna faced up to open sky, or when the radar antenna faced directly up to an absorber.
Q.2
In the range profile you gave me, the received power seems to be lowered gradually from 2m through 5m, instead of dropping down to the noise floor immediately.
What is the cause of the slant of received power?
Q.3
It looks like there is a slope from 0m to 1.7m to me.
Do you know the cause of this slope and how to eliminate it?
We suspect multiple reflections inside the IC or at the pins and substrate.
Best Regards,

0 Alec May over 3 years ago in reply to user4825004

TI__Genius 17265 points

Hi,

1. I don't currently have quick acces to a space for this, but you could quickly test this on your end to get a plot using the visualizer.

2, 3. So, it is expected that there will be some antenna coupling on the device. This is referring to the signal from the TX antenna going directly into the RX antennas, which causes typically higher power in the first few range bins in the 0 doppler domain, which I believe is why we use the max doppler bins for noise. You can also reduce this effect by applying some backoff to the TX power.

Best Regards,
Alec

Sensors

Sensors forum

IWR1843: About the noise floor slope