IWR6843ISK: "Mirror"/image signal peak in data acquired in mmWave Studio

Timofey Savelyev

Part Number: IWR6843ISK
Other Parts Discussed in Thread: IWR1443, , , IWR6843

Dear experts,

I see a strong mirror/image signal peak in my measurements, which appears at (Maximal Distance - Distance of Fundamental Signal Peak).

Moving a corner reflector away from the radar results in a similar offset of the image peak with respect to the maximal distance. I tried several configurations and number of points but the parasitic peak is still there.

Presumably this is caused by a disbalance between the I and Q signals when their phase difference differs from 90 degrees at the signal frequency.

I also tested an ISK-ODS board for the same configuration, and the image peak is still there. This halfs the operational distance unfortunately... I didn't see it in the IWR1443 by the way.

The data was acquired in Complex1 format. In Complex2 the image appears as a peak at a negative distance simmetrically wrt 0.

Any suggestions? Thank you in advance.

Best regards,

Timofey

IWR6843ISK: 512 points, corner reflector at 1.1 m

1024 points:

Corner reflector at 1.7 m:

Corner reflector at 2.6 m:

over 6 years ago

0 Timofey Savelyev over 6 years ago

Intellectual 590 points

IWR6843ISK-ODS has the same effect:

0 Timofey Savelyev over 6 years ago in reply to Timofey Savelyev

Intellectual 590 points

I processed the acquired data in MATLAB and measured the phase difference between the I and Q spectra for the fundamental signal peak:

88.5 degrees for 512 points, corner reflector at 1.1 m;
76.2 degrees for 1024 points, corner reflector at 1.1 m;
84.6 degrees for 1024 points, corner reflector at 1.7 m;
90.7 degrees for 1024 points, corner reflector at 2.6 m;
91.7 degrees for 1024 point, ODS board.

Not exactly 90 degrees, although in a few cases the deviation is small.

0 Akash Gondalia over 6 years ago in reply to Timofey Savelyev

TI__Guru 50240 points

Hi Timofey,

I will need to talk to my team about this. We should have an answer for you sometime this week.

Cheers,
Akash

0 CHETHAN KUMAR Y.B. over 6 years ago in reply to Akash Gondalia

TI__Mastermind 25835 points

Hello,

Thank you for sharing your observation, As you rightly pointed out above observations are due to IQ imbalance in IWR6843, which is not as good as IWR1443. In ES2.0 version of the silicon this is being improved.

Thanks and regards,
CHETHAN KUMAR Y.B.

0 Timofey Savelyev over 6 years ago in reply to CHETHAN KUMAR Y.B.

Intellectual 590 points

Hi Chethan and Akash,

can you please indicate when ES2.0 becomes available? As for IWR1443, I started directly with ES2.0 whose perfect IQ balance was quite impressive. Looking forward to getting the same performance of IWR6843 in this regard.

BR
Timofey

0 Artem Aginskiy over 6 years ago in reply to Timofey Savelyev

TI__Intellectual 1360 points

Hi Timofey,

We are unable to share roadmap schedules on E2E forum. Please reach out to me directly @ <edited>

Thanks,

--Artem

Sensors

Sensors forum

IWR6843ISK: "Mirror"/image signal peak in data acquired in mmWave Studio