Blind MIMO Semantic Communication via Parallel Variational Diffusion: A Completely Pilot-Free Approach
Hao Jiang, Xiaojun Yuan, Yinuo Huang, and Qinghua Guo

TL;DR
This paper introduces Blind-MIMOSC, a fully pilot-free MIMO semantic communication framework that uses a diffusion-based blind receiver with pre-trained score networks for joint channel estimation and source recovery, outperforming existing methods.
Contribution
It presents the first fully pilot-free joint channel estimation and source recovery method for MIMO systems using a diffusion-based approach with pre-trained score networks.
Findings
Achieves superior recovery accuracy compared to state-of-the-art methods.
Reduces channel bandwidth ratio significantly.
Operates without any pilot signals, simplifying system design.
Abstract
In this paper, we propose a novel blind multi-input multi-output (MIMO) semantic communication (SC) framework named Blind-MIMOSC that consists of a deep joint source-channel coding (DJSCC) transmitter and a diffusion-based blind receiver. The DJSCC transmitter aims to compress and map the source data into the transmitted signal by exploiting the structural characteristics of the source data, while the diffusion-based blind receiver employs a parallel variational diffusion (PVD) model to simultaneously recover the channel and the source data from the received signal without using any pilots. The PVD model leverages two pre-trained score networks to characterize the prior information of the channel and the source data, operating in a plug-and-play manner during inference. This design allows only the affected network to be retrained when channel conditions or source datasets change,…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsWireless Signal Modulation Classification · Blind Source Separation Techniques · Advanced MIMO Systems Optimization
