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.
@article{arxiv.2010.12972,
title = {Blind Deinterleaving of Signals in Time Series with Self-attention Based Soft Min-cost Flow Learning},
author = {Oğul Can and Yeti Z. Gürbüz and Berkin Yıldırım and A. Aydın Alatan},
journal= {arXiv preprint arXiv:2010.12972},
year = {2023}
}