AWR1642BOOST: Dead Zone of mmWave

Divya Harikumar

Expert 2020 points

Part Number: AWR1642BOOST
Other Parts Discussed in Thread: TIDEP-0092

Hi,

I was referring TIDEP-0092 : Short range radar(SRR) reference design which says that it has a FoV of 120 degree.

I would like to know if there is a dead zone within this FoV?

Regards,

Divya Harikumar

over 5 years ago

0 Cesar over 5 years ago

TI__Guru**** 150344 points

No,

I assume "dead zone" means an area where a moving object is not detected.

No, there is no such zone.

However the range detection decreases when the angle increases. Optimal detection happens at 0 deg. (boresight)

thank you
Cesar

0 Divya Harikumar over 5 years ago in reply to Cesar

Expert 2020 points

Hi Cesar,

Thank you for your prompt reply!

I would also know if it applies to static objects like a car parked or trees within the given FoV?

Regards,

Divya Harikumar

0 Cesar over 5 years ago

TI__Guru**** 150344 points

Yes, it applies to static objects as well.

For static objects the detection range is expected to be shorter than for moving objects.

Thank you

Cesar

0 Divya Harikumar over 5 years ago in reply to Cesar

Expert 2020 points

Hi Cesar,

Will static objects be denoted by a tracker?

Regards,

Divya Harikumar

0 Divya Harikumar over 5 years ago in reply to Cesar

Expert 2020 points

Hi,

We got a scenario where pole was tracked while the car next to it was only seen as a detected point?

why is it so?

Regards,

Divya Harikumar

0 Cesar over 5 years ago in reply to Divya Harikumar

TI__Guru**** 150344 points

Hi,

Sorry for late response,

The tracking performance is dependent on detection quality.

If a moving object is not detected consistently it will not be tracked

thank you

Cesar

0 Divya Harikumar over 5 years ago in reply to Cesar

Expert 2020 points

Cesar,

I was asking in the case of static objects, as the pole was tracked instead of the car next to it.

What determines the detection quality?

Regards,

Divya Harikumar

0 Cesar over 5 years ago in reply to Divya Harikumar

TI__Guru**** 150344 points

It would be the RCS of the object.

It would be surprising that the pole is detected instead of the car.

Thank you
Cesar

0 Divya Harikumar over 5 years ago in reply to Cesar

Expert 2020 points

Cesar,

Yup,it was surprising.

What could have been the reason such a scenario happened?

Is there any algorithm implemented such that poles/signs boards will be tracked instead of a parked car?

Regards,

Divya Harikumar

0 Cesar over 5 years ago in reply to Divya Harikumar

TI__Guru**** 150344 points

I think that probably the interpretation of the data is not correct.

I still think that the car is detected not the pole.

You can perform some testing by removing the car , removing the pole and comparing.

To detect more details in the scene you would need sensor with higher angular resolution such as our imaging radar.

I will close this thread now

If there are new questions please start a new thread

thank you

Cesar

0 Divya Harikumar over 5 years ago in reply to Cesar

Expert 2020 points

Cesar,

We did remove the car and that's how we concluded that pole was being tracked instead of the car since the tracking still remained in the same location

Regards,

Divya Harikumar

0 Cesar over 5 years ago in reply to Divya Harikumar

TI__Guru**** 150344 points

It is possible that the car and the pole are in the same range bin. In that case, only the car will be detected since we have a static situation of small RCS object next to large RCS. The SRR algorithm will detect only one static object in same range bin.

In this specific use case we would definitely need raw data to perform more analysis

thank you
Cesar

Sensors

Sensors forum

AWR1642BOOST: Dead Zone of mmWave