BLIND DEINTERLEAVING OF SIGNALS IN TIME SERIES WITH SELF-ATTENTION BASED SOFT MIN-COST FLOW LEARNING

CAN O., GÜRBÜZ Y. Z., Yildirim B., ALATAN A. A.

IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), ELECTR NETWORK, 6 - 11 Haziran 2021, ss.3295-3299 identifier identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/icassp39728.2021.9415025
  • Basıldığı Ülke: ELECTR NETWORK
  • Sayfa Sayıları: ss.3295-3299
  • Anahtar Kelimeler: Deinterleaving, attention, min-cost flow, IMPROVED ALGORITHM
  • Orta Doğu Teknik Üniversitesi Adresli: Evet

Özet

We propose an end-to-end learning approach to address deinterleaving of patterns in time series, in particular, radar signals. We link signal clustering problem to min-cost flow as an equivalent problem once the proper costs exist. We formulate a bi-level optimization problem involving min-cost flow as a sub-problem to learn such costs from the supervised training data. We then approximate the lower level optimization problem by self-attention based neural networks and provide a trainable framework that clusters the patterns in the input as the distinct flows. We evaluate our method with extensive experiments on a large dataset with several challenging scenarios to show the efficiency.