University of Oulu

S. Jamshidiha, V. Pourahmadi, A. Mohammadi and M. Bennis, "Link Activation Using Variational Graph Autoencoders," in IEEE Communications Letters, vol. 25, no. 7, pp. 2358-2361, July 2021, doi: 10.1109/LCOMM.2021.3076190

Link activation using variational graph autoencoders

Saved in:
Author: Jamshidiha, Saeed1; Pourahmadi, Vahid1; Mohammadi, Abbas1;
Organizations: 1Department of Electrical Engineering, Amirkabir University of Technology, Tehran, Iran
2Centre for Wireless Communications, University of Oulu, Finland
Format: article
Version: accepted version
Access: open
Online Access: PDF Full Text (PDF, 0.4 MB)
Persistent link:
Language: English
Published: Institute of Electrical and Electronics Engineers, 2021
Publish Date: 2021-10-14


An unsupervised method is proposed for link activation in wireless networks by identifying clusters of interfering users. A k-nearest neighbors interference graph is first defined for the wireless network which is then mapped to a stochastic latent space. The users are then clustered in the latent space using a Gaussian mixture model, and one user from each interfering cluster is activated while the rest of the users in that cluster remain idle. The proposed framework is scalable, works across several network topologies such as device to device (D2D), and is close to the optimal solution in performance.

see all

Series: IEEE communications letters
ISSN: 1089-7798
ISSN-E: 2373-7891
ISSN-L: 1089-7798
Volume: 25
Issue: 7
Pages: 2358 - 2361
DOI: 10.1109/LCOMM.2021.3076190
Type of Publication: A1 Journal article – refereed
Field of Science: 213 Electronic, automation and communications engineering, electronics
Copyright information: © 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.