• State TI Thinks Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 689 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

CCS/AWR1642BOOST: Intereference issue

Sreenath Vipin
Intellectual 530 points
Part Number: AWR1642BOOST

Tool/software: Code Composer Studio

Hi,

I want to know how the AWR 1642 sensor hande the interference issue if 2 such modules are fixed on 2 cars and both are moving in a traffic one following the other. How is the reflected signal interefence is handled by the 2 devices?

BR

Sreenath

  • 0
    TI__Guru 58035 points

    Hello Sreenath,

    There are various algorithms that are being used to detect and mitigate interference.  Some IEEE papers that talk about these schemes are listed below:

    •M.Wagner , F. Sulejmani, A. Melzer, P. Meissner, M. Huemer,

      „Threshold-Free Interference Cancellation Method for Automotive FMCW Radar Systems”

    •J. Bechter, K. Deb Biswas, C. Waldschmidt,

      „Estimation and cancellation of interferences in automotive radar signals”

    •M. Barjenbruch, D. Kellner, K. Dietmayer, J. Klappstein, J. Dickmann,

      „A Method for Interference Cancellation in Automotive Radar“

    •J. Bechter, F. Roos, M. Rahman, C. Waldschmidt

      „Automotive radar interference mitigation using a sparse sampling approach”

    •J. Bechter, C. Waldschmidt

      „Automotive radar interference mitigation by reconstruction and cancellation”

    Regards,

    vivek

     

  • 0 in reply to Vivek Dham
    Intellectual 530 points

    hi,

        Two cars car 1 and car 2 , each fixed with mmwave radar device is moving ina traffic in 2 lanes one behind the other, they are continuously transmiting the signals. In front of both cars another car 3 is moving. both cars are transmitting the signals to car 3 and the reflected signl for each device is coming back. How will both devices know the reflected signal is for that particular device in car 1 or car 2? how will they distinguish between that reflected signal, whether it is for car 1 or for car 2?

    BR

    Sreenath

  • 0 in reply to Sreenath Vipin
    TI__Guru 58035 points

    Hello Sreenath,

    There is no direct way of knowing if the reflected signal is from car 1 or car 2. If the signals do not overlap in the exact IF bandwidth then any way one radar will not be able to "see" other radars signal , since anything outside its IF bandwidth will be filtered heavily.

    If the signal does happen to over all lap even in IF band then you need to use some higher level algos (referenes provided above) to detect that there is interference happening and then use one of the mitigation techniques discussed in the papers.  

    regards,

    vivek

  • 0 in reply to Vivek Dham
    Intellectual 530 points

    hi

    thank you for the reply.

    What if the car1 moving in the traffic has the mmwave radar device on the back of the car and the car2 following the car 1 has a device on the front . car 2 device transmit asignal and car 1 device on the back will also transmit signlas. Then there is a chance of interefrence of signals between the 2 devices on each car how will this be mitigated, how will the signals will be identified correctly?

    BR

    Sreenath

  • 0 in reply to Sreenath Vipin
    TI__Guru 58035 points

    Hello Sreenath,

    Yes, there is chance of these sensors interfering with each other. As mentioned above you would need to implement higher levels algos , some of them discussed in the papers I provided links to , to detect interference and mitigate the effect.

    Regards,

    Vivek

  • 0 in reply to Vivek Dham
    Intellectual 530 points

    hi 

    So if i flash the SRR demo code into the AWR 1642 and test it on highway will it properly work i mean is there  any alogorithm  already present in the device to mitigate the interference problem  or do we have to develop it?

    BR

    sreenath

  • 0 in reply to Sreenath Vipin
    TI__Guru 58035 points

    Hello Sreenath,

    The SRR demo does not include the interference detection and mitigation algo implemented. This has to be added separately.

    Regards,

    Vivek

  • 0 in reply to Vivek Dham
    Intellectual 530 points

    Hi Vivek,

    So we have to develop the algo for interference detection and mitigation.Does the chip already contatin the algo for clustering and tracking? also when we tested in an open space wind has significant effect on the points shown in the SRR visualizer and it seems to have some stability issues. Te board we used was AWR 1642 Boost ODS.

    BR

    sreenath

  • 0 in reply to Sreenath Vipin
    TI__Guru 58035 points

    Hello Sreenath,

    The SRR demo application includes an example clustering and tracking code.

    Could you provide more details on the stability issues? What is the scenario/scene in which you are doing the measurement? what is the observation?

    Regards,

    Vivek

  • 0 in reply to Vivek Dham
    Intellectual 530 points

    Hi Vivek,

    Like the demo visualizer we cannot adjust the field of view(FoV) in the SRR visualizer right?

    So what is the field of view of AWR 1642 Boost that we get as a default?

    We placed the device on a car which was stationary and it was a windy day, so we got some ghost points which we coudn't identify and the points were erratic  like randomly changing . The device was placed in open space . It identified a person walking and also another car which was moving in its FoV,but we couid not distinguish all the points?

    Some points in the doppler plot was below the origin line like below zero with a doppler showing about 10m/s and was constantly staying there?

    Some times if a person is moving away from the radar it is showing 2 clusters moving away from the radar one through the zero line and another above the zero line with some velocity.

    We are little confused because of these issues

    BR

    Sreenath

  • 0 in reply to Sreenath Vipin
    TI__Guru 58035 points

    Hello Sreenath,

    >>We placed the device on a car which was stationary and it was a windy day, so we got some ghost points which we coudn't identify and the points were >>erratic  like randomly changing

    The detection depends on the detection threshold. You can change that in the mmwave OOB demo by editing the config file to adjust the detection threshold value.

     >>It identified a person walking and also another car which was moving in its FoV,but we couid not distinguish all the points?

    If the car and the person are close to each other then it could be difficult to detect the person always. You can get a better resolution by using 1843 device with 3 Tx.

    >>Some points in the doppler plot was below the origin line like below zero with a doppler showing about 10m/s and was constantly staying there?

    What is the velocity resolution in the configuration you had set? If the velocity resolution itself is 10m/sec then it might just be one bin error.

    Regards,
    vivek