… unfolding Abstract: This paper investigates the use of hybrid model-and-data-based deep
learning on a recently proposed doubly-iterative turbo equalizer for handling inter-symbol …

… can be optimized. In this context, we propose a joint optimization through deepunfolding of
the hyper-parameters that naturally arise in this kind of doubly iterative turbo VEP receivers. …

… to design an equalizer for the complex time-frequency doubly-… , the deep unfolding approach
is used to unfold the iteration of … We use classification ideas to optimize the equalizer. The …

… deep unfolding methodology that can be applied to iterative … Here, the optimization problem
P1 is ADMM. The Lstep … - FD radios - have the potential to double the spectral efficiency of a …

… -iterative vector approximate message passing (VAMP) soft frequency-domain equalizer (SFDE)
is a preferred choice for turbo equalization … applied in turbo equalization achieves extra …

… as multi-layer deep feedforward networks to optimize the necessary … lth iteration of EP, and
the maximum number of iterations is … Studer, “Deep unfolding for communications systems: A …

… neural network following the idea of deep unfolding [2]. Although the … Note, the double arrows
in Fig. 2 (and all other figures) … iterations lead to a lower BER; we used 96 iterations during …

… we use a deep unfolded NN structure that represents iterative … trained by the LSTM optimizer
by using the generated dataset. … , “A double EP-based proposal for turbo equalization,” IEEE …

… In [6], a frequency domain turbo equalizer (FDTE) aided by low… The learnable parameters
in DL-GAMP are optimized using … We unfold the iterations of Sim-GAMP and introduce several …

… is obtained by first unfolding the EP iterations into layers and … omit the superscript ι of the
turbo iteration index for simplicity. … follow the configurations of EP and double EP (DEP), which …