… As an alternative, in this paper we propose an approximated ML estimate using the
variational autoencoder (VAE) method [15], [16]. VAEs are widely used in the literature of deep …

… • We present a new method for blind channel equalization using variational inference,
and VAEs in particular. • We extend the method to the case of coded communications. …

… Throughout this work, we focus on adaptive channel equalization assuming an infinitely long
random data sequence. Hence, we do not have to worry about overfitting during training for …

… in this thesis concerns blind channel equalization, which we … Variational autoencoders
(VAEs) are variants of AEs that … blind channel equalization, which we review in Section 4.2. …

… Further, [7] presented a variational autoencoder (VAE)-based model, whose cost function …
(ie, channel equalization). While [3] extended this work to multi-carrier Channel AE, which …

… channel equalization and estimation using the vector quantized (VQ) variational autoencoder
… of the conventional VAE-based equalizer to nonlinear systems employing high-order …

… However, in our design, iCSI-SVAE will perform an additional channel equalization to
mitigate the distortion effect caused by the imperfect CSI ˆHAB and PNR. Therefore, we modified …

… blind equalization methods, VAE based receivers have similar performance and complexity
is not exponential in the channel … in [25] for blind channel equalization. And numerical results …

… effectiveness, for instance in channel equalization [11]. However, to our knowledge, there
are no approaches in the literature that use the VAE for channel estimation. Our contributions …

… and compare the results with standard autoencoders. Our findings suggest that variational
auto-encoders are more robust to channel degradation than auto-encoders. Furthermore, we …