Low-Complexity Near-Field Channel Estimation and Subcarrier-Cooperative Hybrid Precoding for Wideband XL-MIMO OFDM Systems

arXiv:2609.14373 · eess.SP · Submitted 2026-09-13 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Electrical Engineering. The day in electrical engineering: signals, systems and the hardware that carries them.

Rosa: We're looking at the paper: "Low-Complexity Near-Field Channel Estimation and Subcarrier-Cooperative Hybrid Precoding for Wideband XL-MIMO OFDM Systems".

Dev: Here's the gist of the paper. Low-Complexity Near-Field Channel Estimation and Subcarrier-Cooperative Hybrid Precoding for Wideband XL-MIMO OFDM Systems addresses both accurate near-field channel acquisition and scalable precoding for wideband extremely large aperture MIMO (XL-MIMO) OFDM systems, with particular focus on reducing complexity Channel Estimation Challenges The paper tackles the challenge of acquiring accurate near-field wideband channel state information (CSI) in XL-MIMO OFDM systems, where the conventional planar-wave assumption breaks down and spherical wavefronts should be considered The high-dimensional grids of channel propagation parameters lead to dramatically increased complexity and overhead for conventional sparse recovery methods.

Rosa: To overcome this, the authors propose a decoupled off-grid channel estimation algorithm based on sequential sparse Bayesian learning (SBL) which exploits the separable structure of the OFDM nearfield atom to decompose high-dimensional searches into a sequence of low-dimensional operations.

Dev: A continuous-domain Newton refinement based on the exact marginal likelihood is incorporated to mitigate grid mismatch and improve resolution.

Rosa: Sequential SBL Framework The proposed sequential SBL approach avoids computationally intensive global operations over all atoms in classical Expectation–Maximization (EM)-based SBL by searching for the single most informative atom iteratively.

Paper summary: Rosa: So, we're looking at this paper called Low-Complexity Near-Field Channel Estimation and Subcarrier-Cooperative Hybrid Precoding for Wideband XL-MIMO OFDM Systems. Basically, they're tackling the hard problem of getting accurate channel information when you've got these massive systems with extremely large aperture MIMO.

Dev: It seems the main thrust is reducing the computational complexity involved in both figuring out that channel and then designing the precoding matrix for those wideband systems. They focus on making it scalable while still aiming for accuracy in a real-world scenario.

Taro: What's interesting is that they are dealing with channel parameters that have five continuous variables each, which makes conventional sparse recovery methods really hard to handle because of the grid mismatch issues <ref:2609.14373#pg1>.

Rosa: Right, so instead of trying to search through all those high-dimensional grids at once, they propose this decoupled off-grid channel estimation algorithm based on sequential sparse Bayesian learning, or SBL.

Dev: That approach seems smart because it exploits the separable structure of the OFDM nearfield atom to break down that massive five-dimensional search into a sequence of simpler, low-dimensional operations <ref:2609.14373#pg1>.

Taro: The idea of iteratively searching for the single most informative atom and then using a continuous-domain Newton refinement based on the exact marginal likelihood to fix any grid mismatch sounds like it's trying to get a good resolution without the brute force computation <ref:2609.14373#pg1>.

Rosa: And when you look at how they do that search, they have this decoupled coarse on-grid search which aims to find the best point by exploiting the nested structure of those five variables separately <ref:2609.14373#pg2>.

Dev: It sounds like a multi-step procedure where they start by processing things like the delay transform to eliminate one of those indices, and then use some angular prescreening to narrow down the options before finally doing an angle search <ref:2609.14373#pg2>.

Taro: I'm curious about that final UE angle search, since they say it eliminates the dependence on an unknown variable called psi by using a non-coherent combination across time slots <ref:2609.14373#pg2>.

Rosa: After they get those estimated parameters, they develop a subcarrier-cooperative transmission scheme for downlink data, which has extra degrees of freedom for interference suppression across different subcarriers <ref:2609.14373#pg2>.

Dev: They set up a large-scale matrix-inversion-free alternating optimization framework to maximize the sum user rate, and then they suggest a low-complexity sub-optimal scheme that doesn't even need those alternating iterations <ref:2609.14373#pg2>.

Taro: That sounds like they are trying to keep the complexity down by avoiding heavy matrix inversion during the precoder design using a method based on minorization maximization principle <ref:2609.14373#pg2>.

Rosa: So, what we see here is that even with perfect channel information and accurate angle information, the benefit of that subcarrier cooperation mainly shows up in interference-limited regimes, giving extra nulling ability in the subcarrier domain <ref:2609.14373#pg2>.

Dev: And this low-complexity algorithm acts as a really good starting point for their alternating algorithm when the solution space is too big and complicated to handle directly <ref:2609.14373#pg2>.

Taro: The paper flags a limitation, specifically that the total cost of that decoupled coarse search is dominated by the BS prescreening operation, which they say is much smaller than classical approaches <ref:2609.14373#pg1>.

Rosa: They also point out that their proposed method achieves two levels of complexity reduction compared to classical methods by avoiding the need for global posterior computation and replacing it with additive lightweight operations <ref:2609.14373#pg1>.

Dev: So, we're seeing an effort here to keep the computational burden manageable for these wideband XL-MIMO systems while still aiming for good performance in interference-limited situations <ref:2609.14373#pg2>.

Taro: It seems like they’ve successfully decomposed a very complicated five dimensional search into a sequence of lower dimensional projections and then used refinement to handle the errors introduced by coarse grids <ref:2609.14373#pg1>.

Rosa: So, what this paper means for the real world is that we can potentially get accurate channel info in these massive systems without needing astronomical computational power to do it <ref:2609.14373#pg1>.

Dev: It gives us a more practical way to design precoders when the system size is huge, especially when we need to operate in real-time with low latency constraints <ref:2609.14373#pg2>.

Taro: If this works out in trials, it suggests that we can deploy these wideband XL-MIMO systems more robustly because the estimation part isn't crippling the whole operation <ref:2609.14373#pg1>.

Rosa: And after this paper, we still have to see how these low-complexity precoding schemes perform when you actually put them into a system with multi-user data transmission where you need that subcarrier cooperation <ref:2609.14373#pg2>.

Dev: That's the next step, figuring out the real-world latency and failure modes under those specific cooperative conditions <ref:2609.14373#pg2>.

Taro: So, we’ll keep an eye on how well these algorithms handle the unpredictable behavior of real-world environments when things go wrong <ref:2609.14373#pg1>.

Rosa: Right, so that’s our look at this paper on Low-Complexity Near-Field Channel Estimation and Subcarrier-Cooperative Hybrid Precoding for Wideband XL-MIMO OFDM Systems.

Conclusion: Rosa: So, we're wrapping up this look at the paper about low-complexity channel estimation and subcarrier cooperation for wideband XL-MIMO OFDM systems.

Dev: Yeah, that paper by Author Names focuses on two main things: figuring out those difficult near-field channels and then designing a precoder that uses subcarriers cooperatively.

Rosa: It really boils down to taking a problem with high dimensions and making it manageable for real-world hardware.

Dev: They used this sequential sparse Bayesian learning approach to tackle the channel estimation part, breaking down the big five dimensional search into smaller, less complex steps.

Rosa: And then after they get that channel info, they propose a low-complexity precoding scheme for downlink transmission that leverages subcarrier cooperation for better interference suppression.

Dev: The numbers show it works well compared to older methods, and the complexity reduction is significant because they avoid those heavy matrix inversions during the precoder design.

Rosa: It suggests that we can get accurate channel information in these huge systems without needing astronomical amounts of computational power just for estimation.

Dev: But there are still caveats, and we need to see how this performs when things go wrong in a real-world deployment scenario, like with unpredictable environments.

Taro: Exactly. The authors mention that the cooperative gain from subcarriers is really most valuable when the interference is already limited, so we're still looking at those specific operating regimes.

Rosa: So, while the estimation part looks promising and much lighter on computation, we still have to test how stable this whole system stays when it encounters unexpected channel behavior.

Dev: That's right. We need to see the latency and failure modes of this low-complexity precoding scheme when we push it into a live system with multi-user data transmission.

Technische Universitat Berlin · Huawei Technologies Duesseldorf GmbH

eess.SP

Submitted: 2026-09-13

Updated: 2026-11-07

Comments: Submitted to the IEEE journal

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 89/100

The gist: Low-Complexity Near-Field Channel Estimation and Subcarrier-Cooperative Hybrid Precoding for Wideband XL-MIMO OFDM Systems addresses both accurate near-field channel acquisition and scalable…

Key concepts

Near-Field Channel Estimation
Acquiring accurate channel information when the transmitter and receiver are very close together. In XL-MIMO OFDM, this is difficult because the signal wavefronts are spherical, not planar waves, leading to a high-dimensional search space that traditional methods struggle with.
Sequential Sparse Bayesian Learning (SBL)
A method used for channel estimation that iteratively searches for the single most informative parameter (atom) at each step. This approach avoids computationally expensive global searches by breaking down the complex problem into a sequence of simpler, low-dimensional operations, making it much faster.
Subcarrier-Cooperative Hybrid Precoding
A transmission scheme where multiple users share information across different subcarriers to suppress interference effectively. The paper uses this technique after estimating the channel to maximize user data rates in an interference-limited environment with minimal computational complexity.

Terminology

Summary

Low-Complexity Near-Field Channel Estimation and Subcarrier-Cooperative Hybrid Precoding for Wideband XL-MIMO OFDM Systems addresses both accurate near-field channel acquisition and scalable precoding for wideband extremely large aperture MIMO (XL-MIMO) OFDM systems, with particular focus on reducing complexity

Channel Estimation Challenges

The paper tackles the challenge of acquiring accurate near-field wideband channel state information (CSI) in XL-MIMO OFDM systems, where the conventional planar-wave assumption breaks down and spherical wavefronts should be considered The high-dimensional grids of channel propagation parameters lead to dramatically increased complexity and overhead for conventional sparse recovery methods. To overcome this, the authors propose a decoupled off-grid channel estimation algorithm based on sequential sparse Bayesian learning (SBL) which exploits the separable structure of the OFDM nearfield atom to decompose high-dimensional searches into a sequence of low-dimensional operations. p. 0 A continuous-domain Newton refinement based on the exact marginal likelihood is incorporated to mitigate grid mismatch and improve resolution. p. 0

Sequential SBL Framework

The proposed sequential SBL approach avoids computationally intensive global operations over all atoms in classical Expectation–Maximization (EM)-based SBL by searching for the single most informative atom iteratively. The search is decomposed into a sequence of cheap low-dimensional operations by exploiting the separable structure of near-field OFDM MIMO channels. p. 2 The algorithm proceeds by selecting a new grid point that maximizes the change to the cost function, which is achieved by choosing the grid point that maximizes Qi. p. 3 To mitigate mismatch error due to coarse grids, a local off-grid Newton refinement is applied after the on-grid search. p. 3

Decoupled Coarse On-Grid Search

The decoupled coarse on-grid search aims to find the best candidate grid point by exploiting the nested structure of Qi, which allows for an initial estimate for each parameter group separately. This is achieved through a multi-step search procedure that processes the three factors in (23) sequentially, starting from the delay transform to eliminate subcarrier index n. p. 4 The BS near-field codebook projection and search handles the remaining middle sum by performing an angular prescreening to select a single representative atom for each angular direction. p. 4 Finally, the UE angle search eliminates the dependence on the unknown ψ by using a non-coherent combination across time slots. p. 4

Hybrid Precoding Design

For wideband multi-user downlink (DL) data transmission, a subcarrier-cooperative transmission scheme is proposed based on the estimated channel parameters, which has additional cross-subcarrier degree of freedom (DoF) for interference suppression. A large-scale matrix-inversion-free alternating optimization framework is established to maximize the sum user rate, followed by a low-complexity sub-optimal scheme without alternating iterations. p. 0 The BS digital precoder design involves fast iterative optimization under minorization–maximization (MM) principle to avoid heavy matrix inversion complexity. p. 8

Performance and Complexity Analysis

Simulations demonstrate the accuracy of the estimation algorithms compared to benchmarks and Cramer–Rao lower bound (CRLB), showing that the proposed algorithm significantly outperforms baselines despite oracle UE angle and delay information. The total cost of the decoupled coarse search is dominated by the BS prescreening operation in Section IV-A2, which is much smaller than classical approaches. p. 6 The proposed method achieves two levels of complexity reduction compared to classical approaches, avoiding global posterior computation and replacing multiplicative joint-grid evaluation with additive lightweight operations. p. 6

Subcarrier-Cooperative Gain

With perfect CSI and AO, the benefit of cooperative transmission mainly lies in the interference-limited regime, providing additional interference-nulling ability in the subcarrier domain. The proposed low-complexity algorithm serves as a high-quality initial solution for the alternating algorithm when the solution space is prohibitively large and complicated. p. 9 The scheme avoids iterative adaptation between the BS and users, reducing both computational complexity and signaling overhead. p. 9

Conclusion

In this paper, a decoupled off-grid sequential SBL algorithm is proposed to address the challenging near-field channel estimation problem in XL-MIMO OFDM systems. The authors effectively decomposed the computationally prohibitive 5-D joint parameter search into a sequence of low-dimensional marginal projections and proposed a continuous-domain Newton refinement to overcome severe grid mismatch. p. 0 After acquiring CSI, two low-complexity hybrid-precoding algorithms are proposed to improve the transmission in the interference-limited regime with subcarrier cooperation. p. 1 Simulations demonstrated the accuracy of estimation algorithms and communication performance of precoding based on estimated CSI in the challenging scenario of multi-user multi-subcarrier XL-MIMO. p. 1

APPENDIX A

ANALYTICAL FIRST AND SECOND DERIVATIVES IN (30)

Let the first and second partial derivative of an atom ϕpξq over each coordinate ξ be ϕξ fi Bϕpξq Bξ and ϕξxi fi B 2ϕpξq Bxi 2 respectively, then we have qxi “ ϕH xi C´1y, qxxi “ ϕH xi xC´1y, sxi “ 2Retphi H xi C´1phiu and sxxfi “ 2Retphi H xi xC´1phiu 2ϕH xi C´1phixi.

APPENDIX B

MATRIX-INVERSION-FREE POSTERIOR UPDATE

Derivation of LOO posterior covariance Σ´i: We first define the posterior precision of active set ΛS¯ fi Σ´1 S¯ “N´10 Φ H S¯ ΦS¯ Γ´1 S¯. Then for atom i P S¯, we write the full posterior quantities in blocks as ΛS̄ “ N´10 Φ H̄i λ̄i λH̄i λiȷ, ΣS̄ “ N´10 Σbarii σbaris σH̄i σiȷ, µS¯ “ N´10 µbarii µiȷ, where for notation simplicity, we define Λ¯ii fi rΛS¯sbi,bi, λ̄i fi rΛS¯si,i, λi fi rΛSbarisbi, ΣSbarii fi rΣSbarisbi σH̄i σiȷ, µbarii fi rµSbarisbi, µi fi rµSbarisbi. p. 5

APPENDIX A

ANALYTICAL FIRST AND SECOND DERIVATIVES IN (30)

Let the first and second partial derivative of an atom ϕpξq over each coordinate ξ be ϕξ fi Bϕpξq Bxi and ϕxixi fi B 2phiH xi C´1y, qxxi “ϕH xi xC´1y, sxi “ 2Retphi H xi C´1phiu and sxxfi “ 2Retphi H xi xC´1phiu 2ϕH xi C´1phixi.

APPENDIX B

MATRIX-INVERSION-FREE POSTERIOR UPDATE

Derivation of LOO posterior mean µ´i: First define gS¯ fi N´10 Φ H S̄y, then µS¯ “ ΣS̄gS̄ and gS̄ fi ΛSBµSbarisbi, given in blocks „gbarii giȷ, where gbarii fi rgSbarsbi and gi fi rgSbarsbi.

APPENDIX B

MATRIX-INVERSION-FREE POSTERIOR UPDATE

Derivation of new posterior after adding new atom: The same block matrix inversion identity is used when adding a newly selected atom or adding back an updated atom after LOO refinement.

APPENDIX A

ANALYTICAL FIRST AND SECOND DERIVATIVES IN (30)

Let the first and second partial derivative of an atom ϕpξq over each coordinate ξ be ϕxi fi BphiH xi C´1y, qxi “ϕH xi xC´1y, sxi “ 2Retphi H xi C´1phiu and sxxfi “ 2Retphi H xi xC´1phiu 2ϕH xi C´1phixi.

APPENDIX B

MATRIX-INVERSION-FREE POSTERIOR UPDATE

Derivation of LOO posterior covariance Σ´i: We first define the posterior precision of active set ΛS¯ fi Σ´1 S¯ “N´10 Φ H S¯ ΦS̄ Γ´1 S̄. Then for atom i P S̄, we write the full posterior quantities in blocks as ΛSbarisbi “ N´10 Φ Hbarii λbaril λiȷ, ΣSbarisbi “ N´10 Σbarii σH̄i σiȷ, µS¯ “ N´10 µbarii µiȷ, where for notation simplicity, we define Λ̄ii fi rΛSbarsbi,bi, λ̄i fi rΛSbarsbi si,i, ΣSbarii fi rΣSbarisbi σH̄i σiȷ, µbarii fi rµSbarisbi, µi fi rµSbarisbi.

APPENDIX B

MATRIX-INVERSION-FREE POSTERIOR UPDATE

Derivation of LOO posterior covariance Σ´i: We first define the posterior precision of active set ΛS¯ fi Σ´1 S̄ “N´10 Φ H S̄ ΦSbarisbi Γ´1 S̄. Then for atom i P S̄, we write the full posterior quantities in blocks as ΛSbarisbi “ N´10 Φ Hbarii λbaril λiȷ, ΣSbarisbi “ N´10 Σbarii σH̄i σiȷ, µS¯ “ N´10 µbarii µiȷ, where for notation simplicity, we define Λ̄ii fi rΛSbarsbi si,i, ΣSbarisbi fi rΣSbarisbi σH̄i σiȷ, µbarii fi rµSbarisbi, µi fi rµSbarisbi.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Decoupled off-grid sequential SBL-based channel estimation can be implemented to acquire accurate near-field wideband channel state information for XL-MIMO OFDM systems with low complexity by decomposing the five-dimensional joint search into a sequence of low-dimensional operations and active set inference.

  2. The system can perform high rate communication based on estimated CSI with low precoding complexity by utilizing a subcarrier-cooperative transmission scheme that offers additional cross-subcarrier degrees of freedom to suppress interference.

  3. A hybrid precoding framework can be developed exactly enabled by the estimated wideband MIMO-OFDM channels, allowing for high-capacity yet computationally efficient multi-user wideband beamforming.

  4. The proposed low complexity algorithm can serve as a high quality initial solution for the alternating optimization algorithm when the solution space is prohibitively large and complicated, providing a robust starting point for full optimization.

  5. The improved system reduces computational cost by replacing multiplicative joint-grid evaluation with additive lightweight operations in the decoupled search, leading to a total cost of OpGangMrfGtauTq which is lower than classical approaches.

Sources

Related papers