A known signal, signal 1 got overlapped with an unknown signal. Likewise, signal 2 overlapped with another unknown signal. The problem I face now is how to classify the overlapped signals based on the signal source and label them as from signal 1 and signal 2 based on the available source training data of signal 1 and 2. Can you please help me in this?

  • $\begingroup$ Are you looking to classify the 'unknown' as class 3,4,5 etc? If so, how many different classes can there be? Is it finite and known up-front (closed-set classification)? $\endgroup$
    – Jon Nordby
    Commented Apr 25, 2019 at 19:45
  • $\begingroup$ Or you just want to classify signal 1 or 2 on noisy samples? $\endgroup$
    – Jon Nordby
    Commented Apr 25, 2019 at 19:45


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.