IWR1443BOOST: Line-shaped noise in Range-Doppler Map

Matt

Prodigy 65 points

Part Number: IWR1443BOOST

We captured IWR1443BOOST ADC data in static scene.

Only zero Doppler component has been observed in almost all Tx-Rx pairs.

However, in specific Tx-Rx pairs (Tx4-Rx1 & Tx4-Rx2), non-zero Doppler line-shaped noise has been observed.

How can I reduce the noise?

We used following chirp parameters:

dfeDataOutputMode 1
channelCfg 15 7 0
adcCfg 2 1
adcbufCfg 0 1 0 1
lowPower 0 1
profileCfg 0 77.1 7 8 60 131586 0 60 1 512 10000 0 0 30
chirpCfg 0 0 0 0 0 0 0 1
chirpCfg 1 1 0 0 0 0 0 2
chirpCfg 2 2 0 0 0 0 0 4
frameCfg 0 2 128 0 33.33 1 0
dataHeaderCfg 0

Thanks,

over 5 years ago

0 zigang Yang over 5 years ago

TI__Mastermind 40335 points

Hi,

What kind of tool you are using for ADC data capture? Are you using out of box demo or DCA1000 capture board?

Best,

Zigang

0 zigang Yang over 5 years ago in reply to zigang Yang

TI__Mastermind 40335 points

Please provide more information to this thread. Otherwise, I will assume the problem is no longer exist and close this thread.

Best,

Zigang

0 Matt over 5 years ago in reply to zigang Yang

Prodigy 65 points

Hi,

We uses CSI streaming mode (MSS FW is developed based on CSI streaing demo).

We uses nvidia Jetson TX2 as a MIPI CSI-2 receiver.

We also have MMWAVE-DEVPACK & TSW1400.

We will check whether the same promlem occurs in ADC captured by TSW1400.

Thanks,

0 zigang Yang over 5 years ago in reply to Matt

TI__Mastermind 40335 points

Sounds good! Please let me know when you get the results with TSW1400.

Best,

Zigang

0 Matt over 5 years ago in reply to zigang Yang

Prodigy 65 points

Hi,

We noticed there is a individual difference.
We have three IWR1443boost rev.B (ES3.0) boards.
#1: Weaker line-shape noise in Tx4-Rx1 & Tx4-Rx2.
#2: No line-shape noise
#3: Stronger line-shape noise in Tx4-Rx1 & Tx4-Rx2 (as I showed before)

Furthermore, we designed custom PCB based on IWR1443boost rev.B (ES3.0).
We also observed this problem on our custom PCBs,
but noisy antenna pairs are different (Tx1-Rx1, Tx2-Rx4, & Tx3-Rx4).

We captured ADC data of some of above boards by TSW1400 (same chirp parameters were configured by RadarStudio).
We observed same line-shape noise in TSW1400 ADC data.

Thanks,

0 zigang Yang over 5 years ago in reply to Matt

TI__Mastermind 40335 points

Hi,

Can you check whether you have the latest mmwave radar studio 02.01.00.00?

http://software-dl.ti.com/ra-processors/esd/MMWAVE-STUDIO/latest/index_FDS.html

Can you see whether thiings improve if you enable RF calibratioin, check section 17 (CalibConfig Tab Operations) in radar studio user guide.

Can you see whether things looks diffferent if you enable internal loop back, check section 16 (Loopback Tab Operations) in radar studio user guide.

Best,

Zigang

0 Matt over 5 years ago in reply to zigang Yang

Prodigy 65 points

Hi,

We are using mmwave studio 02.01.00.

The noise still appeared regardless of RF calibration congig.

The noise disappeared when I enabled IF loop back.
The noise became weaker in some TRx channels when I enabled PA loop back.

Thanks,

0 zigang Yang over 5 years ago in reply to Matt

TI__Mastermind 40335 points

Hi,

Our hardware expert suggest to reduce Tx output power and Receiver gain.

If there is a coupling due to TXà Rx then receiver would go into compression, this phenomenon would be seen only for smaller range bins, at higher range bins it’s ok, You could also increase the HPF corner frequency to higher value to increase the attenuation due to coupling.

Best,

Zigang

0 Matt over 5 years ago in reply to zigang Yang

Prodigy 65 points

Thank you very much. I will try those solutions.

#I noticed my typo: noisy TRxs in IWR1443boost are "Tx1-Rx4 & Tx2-Rx4". The board doesn't have "Tx4"

Best,

0 Matt over 5 years ago in reply to Matt

Prodigy 65 points

Higher HPF reduces this noise.

Thanks.

Sensors

Sensors forum

IWR1443BOOST: Line-shaped noise in Range-Doppler Map