• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 290 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

AWR1843: Multi-path/Refraction/Diffraction while Radar @77GHz and install on the side of the bus for BSIS functionality

kuowei lin
Prodigy 190 points
Part Number: AWR1843

Multi-path/Refraction/Diffraction while Radar @77GHz and install on the side of the bus for BSIS functionality.

Test condition:

* Radar install on the side of the bus near right front wheel.

* AWR1843 with 3TX transmit and beam steering 22.5 degrees. (ME angle is 60 degrees) Main beam is 30-22.5 = 7.5 degrees near the bus. Below shows as antenna pattern.

   

* Target object is cyclic with rider, and start to ride from 50m far from radar and 1.5m/4.5m side of the bus (Two test cases).

   

* Rider start to ride from rear to front at speed ~15km/h, Radar output the detection result.

Test result:

1. 1.5m test result, the path should be fine. (0, 0) is radar position.

2. 4.5m test result, it seems caused by Multi-path/Refraction/Diffraction. (0, 0) is radar position.

If Radar does not install on the side of the bus, the test result is as following. There is no such issue.

We think it is caused by the Multi-path/Refraction/Diffraction, have any good idea to eliminate or improve it?

Many thanks.

  • 0
    TI__Mastermind 20605 points

    Hi,

    I am checking this with the team. Will get back to you by Monday.

    Regards,

    Kaushik

  • 0 in reply to Kaushik Gowda
    TI__Mastermind 20605 points

    Hi,

    Here is the feedback following discussion with the team. There are no methods to reduce the effect of multipath reflection at the lower layers. One idea is to categorize objects (after detection), into a group based on similar acceleration and trim them down. Typically, a robust clustering and tracking algorithm should be able to get this done.

    Regards,

    Kaushik

  • 0 in reply to Kaushik Gowda
    Prodigy 190 points

    Hi Kaushik,

    Thanks for the information.

    According to the comments, do you mean it can separate out the indirect and direct path by categorization objects?

    From the result of field try, it seems the indirect path is detected, and there is no direct path detected points due to the poor angle resolution (~30 degrees, 4 RX without MIMO mode). How do you think?

    Thanks.

  • 0 in reply to kuowei lin
    TI__Mastermind 20605 points

    Hi Kuowei,

    Can you confirm my understanding of your statement as if the primary detection is not made and only the secondary detections are made in this case? Can you also quantize the signal level of these multipath detections with respect to the noise and primary detection?

    Regards,

    Kaushik

  • 0 in reply to Kaushik Gowda
    Prodigy 190 points

    Hi Kaushik,

    Yes, it seems the primary detection is not made and only the secondary detection are made in this case.

    Any sorry that we do not have the signal level of these multi-path.

  • +1 in reply to kuowei lin
    TI__Mastermind 20605 points

    Hi Kuowei,

    kuowei lin said:
    According to the comments, do you mean it can separate out the indirect and direct path by categorization objects?

    Typically, these clustering and tracking algorithms group a bunch of detections within similar ranges, velocity, acceleration, SNR etc. Multipath detections would have a different feature profile compared to the primary detection which would make it a differentiating factor when it performs a cluster/track. This is why I mentioned that you might see an improvement in the post processing of your detection list.

    Regards,

    Kaushik