Learning a Common Dictionary for CSI Feedback in FDD Massive MU-MIMO-OFDM Systems
Pavan Kumar Gadamsetty, K. V. S. Hari, Lajos Hanzo

TL;DR
This paper introduces a novel common dictionary learning approach for compressing CSI feedback in FDD massive MU-MIMO-OFDM systems, significantly reducing feedback and memory requirements while improving reconstruction accuracy.
Contribution
It proposes a new CDL framework with two methods, CDL-KSVD and CDL-OP, for effective CSI compression across single- and multi-UE systems, outperforming traditional fixed dictionaries.
Findings
CSI feedback reduced by 50%
Lower NMSE than DFT-based dictionary
Memory usage decreased by at least 50%
Abstract
In a transmit preprocessing aided frequency division duplex (FDD) massive multi-user (MU) multiple-input multiple-output (MIMO) scheme assisted orthogonal frequency-division multiplexing (OFDM) system, it is required to feed back the frequency domain channel transfer function (FDCHTF) of each subcarrier at the user equipment (UE) to the base station (BS). The amount of channel state information (CSI) to be fed back to the BS increases linearly with the number of antennas and subcarriers, which may become excessive. Hence we propose a novel CSI feedback compression algorithm based on compressive sensing (CS) by designing a common dictionary (CD) to reduce the CSI feedback of existing algorithms. Most of the prior work on CSI feedback compression considered single-UE systems. Explicitly, we propose a common dictionary learning (CDL) framework for practical frequency-selective channels and…
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
TopicsPAPR reduction in OFDM · Advanced MIMO Systems Optimization · Advanced Wireless Communication Techniques
MethodsProcrustes · Balanced Selection
