Current browse context:
eess.SP
Change to browse by:
References & Citations
Electrical Engineering and Systems Science > Signal Processing
Title: Variational Autoencoders for Precoding Matrices with High Spectral Efficiency
(Submitted on 23 Nov 2021 (v1), last revised 5 May 2022 (this version, v7))
Abstract: Neural networks are used for channel decoding, channel detection, channel evaluation, and resource management in multi-input and multi-output (MIMO) wireless communication systems. In this paper, we consider the problem of finding precoding matrices with high spectral efficiency (SE) using variational autoencoder (VAE). We propose a computationally efficient algorithm for sampling precoding matrices with minimal loss of quality compared to the optimal precoding. In addition to VAE, we use the conditional variational autoencoder (CVAE) to build a unified generative model. Both of these methods are able to reconstruct the distribution of precoding matrices of high SE by sampling latent variables. This distribution obtained using VAE and CVAE methods is described in the literature for the first time.
Submission history
From: Evgeny Bobrov [view email][v1] Tue, 23 Nov 2021 11:45:45 GMT (68kb,D)
[v2] Wed, 1 Dec 2021 06:41:39 GMT (65kb,D)
[v3] Mon, 10 Jan 2022 14:44:33 GMT (208kb,D)
[v4] Tue, 15 Feb 2022 15:20:44 GMT (213kb,D)
[v5] Sat, 26 Feb 2022 12:38:20 GMT (218kb,D)
[v6] Sat, 5 Mar 2022 13:03:58 GMT (162kb,D)
[v7] Thu, 5 May 2022 09:11:54 GMT (290kb,D)
Link back to: arXiv, form interface, contact.