PARAFAC-Based Channel Estimation for Beyond Diagonal Reconfigurable Surfaces
Gilderlan Tavares de Araújo, Bruno Sokal, André L. F. de Almeida · 2026
Read PARAFAC-Based Channel Estimation for Beyond Diagonal Reconfigurable Surfaces by Gilderlan Tavares de Araújo, Bruno Sokal, André L. F. de Almeida (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-26384.txt.
Introduction
Araújo, Sokal, and de Almeida tackle channel estimation for beyond-diagonal reconfigurable surfaces in MIMO systems. The rich coupling that makes these surfaces strong also makes training and control hard. Their plan uses a two-timescale protocol and shapes the scattering as a low-rank PARAFAC tensor model. The link layout stays fixed across blocks while only phase shifts change. A receiver based on alternating least squares then recovers each channel, matching top tensor methods with far less design effort.
Catalog record
| CREATOR | Gilderlan Tavares de Araújo, Bruno Sokal, André L. F. de Almeida |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 24546 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.26384
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
PARAFAC-Based Channel Estimation for Beyond Diagonal Reconfigurable Surfaces Gilderlan Tavares de Araújo, Bruno Sokal, André L. F. de Almeida Abstract Channel estimation is a central bottleneck in BD-RIS-assisted MIMO systems. The richer inter-element coupling that enables large performance gains also makes training and hardware control substantially harder than in diagonal RIS architectures. Existing estimators either target only cascaded channels or require block-by-block reconfiguration of the BD-RIS interconnections, which is costly and difficult to implement in practice. To overcome this limitation, we propose a pilot-assisted tensor framework for group-connected BD-RIS under a two-timescale protocol, where the scattering structure is designed as a low-rank PARAFAC model with fixed factor matrices. This design keeps the interconnection topology constant across blocks and updates only phase shifts, enabling practical operation without sacrificing estimation quality. Building on this structure, we develop a PARAFAC-based alternating least-squares (PALS) receiver that recovers the individual channels. Numerical results confirm that PALS delivers markedly lower composite-channel NMSE than conventional LS, matches the accuracy of state-of-the-art tensor receivers, and sharply reduces BD-RIS design complexity. I Introduction Reconfigurable intelligent surfaces (RISs) are a key enabler for future wireless networks because they provide software-controlled manipulation of the propagation environment [ 8 , 19 ] . In RIS-assisted MIMO systems, however, channel estimation remains a critical bottleneck, as the passive nature of RIS elements and the large dimensionality of cascaded links make pilot design and parameter recovery particularly challenging [ 17 , 10 ] . A major advance in this direction was reported in [ 5 ] , [ 4 ] , which first established a formal connection between tensor decompositions and RIS channel estimation. Specifically, [ 4 ] showed that the received pilot measurements admit a structured multilinear model, enabling identifiability analysis and efficient tensor-based estimators to recover channel factors with reduced training overhead. This tensor-estimation perspective was subsequently extended to BD-RIS systems in [ 3 ] , where the richer scattering architecture (group-connected/fully connected) was incorporated into the estimation model. Therein, the authors demonstrated that tensor methods can estimate individual channel components beyond the cascaded representation, but it also highlighted a practical limitation: competitive performance often depends on repeatedly redesigning or reconfiguring the BD-RIS scattering pattern across training blocks [ 15 , 14 , 18 ] . Related semi-blind tensor methods [ 7 , 6 ] improve spectral efficiency, yet the hardware reconfiguration burden remains largely unresolved. Therefore, a key open issue is to jointly achieve: i) individual channel estimation accuracy; ii) low training overhead; and iii) hardware-feasible BD-RIS operation with limited reconfiguration. Motivated by this gap, this paper proposes a pilot-assisted tensor framework for channel estimation in group-connected BD-RIS systems under a two-timescale protocol. Our core idea is to model the BD-RIS training structure as a PARAFAC tensor [ 11 ] with fixed low-rank factors, so that the interconnection topology remains unchanged across blocks and only phase coefficients are updated. Compared with prior tensor-based designs that reconfigure the full scattering structure at each block, the proposed strategy is substantially more implementation-friendly. On top of this design, we derive a PARAFAC-based alternating least-squares receiver (ALS) that estimates the individual channels. Beyond its conceptual simplicity, PALS also leverages the well-known uniqueness properties of PARAFAC, which yield necessary and sufficient identifiability conditions for the involved channels---conditions not established in earlier works. Numerical results demonstrate that the proposed method preserves the estimation performance of state-of-the-art receivers while markedly reducing BD-RIS design complexity and outperforming conventional LS estimation. 1 1 1 Notation : Scalars are denoted by lowercase letters ( a ), vectors by bold lowercase letters ( \mathbf{a}) , matrices by bold capital letters ( \mathbf{A}) , and tensors by calligraphic letters (\mathcal{A}) . The transpose and Hermitian transpose of a matrix \mathbf{A} are denoted by \mathbf{A}^{\textrm{T}} and \mathbf{A}^{\textrm{H}} , respectively. D_{i}(\mathbf{A}) is a diagonal matrix holding the i -th row of \mathbf{A} on its main diagonal. \|\cdot\|_{\text{F}} is the Frobenius norm. The operator \textrm{diag}(\mathbf{a}) forms a diagonal matrix from its vector argument, while \diamond denotes the Khatri-Rao product. \mathbf{A}_{i.} denotes the i -th row of \mathbf{A} , and \mathbf{I}_{J} is an identity matrix of size J\times J . II System Model Consider a multiple-input multiple-output (MIMO) system assisted by a beyond diagonal reconfigurable intelligent surface (BD-RIS) , where the transmitter and receiver are equipped with M_{T} and M_{R} antennas, respectively, and the BD-RIS has N elements. For simplicity, we assume that the direct transmitter–receiver link is blocked. We adopt the two-timescale protocol in [ 4 ] , [ 3 ] , in which the transmission interval is divided into K consecutive blocks, each composed of T time slots. The pilot matrix is repeated within each block, whereas the BD-RIS response is kept fixed during a block and updated only across blocks. We assume a group-connected BD-RIS architecture [ 15 , 13 , 12 ] , where the N reflecting elements are partitioned into Q groups, each containing \bar{N} interconnected elements [ 15 ] , such that N=\bar{N}Q . Under this structure, the scattering matrix at block k is written as \bm{S}_{k}=\text{bdiag}(\bm{S}^{(1)}_{k},\ldots,\bm{S}^{(Q)}_{k})\in\mathbb{C}^{N\times N} . The pilot signal matrix received at block k is then given by \bm{Y}_{k}=\sum\limits_{q=1}^{Q}\bm{G}^{(q)}\bm{S}^{(q)}_{k}\bm{H}^{(q)\text{T}}\bm{X}^{\text{T}}+\bm{B}_{k}\in\mathbb{C}^{M_{R}\times T}, (1) where \bm{H}^{(q)}\in\mathbb{C}^{M_{T}\times\bar{N}} and \bm{G}^{(q)}\in\mathbb{C}^{M_{R}\times\bar{N}} are the involved channels associated with the q BD-RIS group, respectively, that correspond to the block q -th of \bm{H}\in\mathbb{C}^{M_{T}\times\bar{N}Q} and \bm{G}\in\mathbb{C}^{M_{R}\times\bar{N}Q} , respectively, defined as follows: \displaystyle\bm{H}^{(q)} \displaystyle=\bm{H}_{.\,[(q-1)\bar{N}+1,\ldots,q\bar{N}]}\in\mathbb{C}^{M_{T}\times\bar{N}},\,\,q=1,\ldots,Q (2) \displaystyle\bm{G}^{(q)} \displaystyle=\bm{G}_{.\,[(q-1)\bar{N}+1,\ldots,q\bar{N}]}\in\mathbb{C}^{M_{R}\times\bar{N}},\,\,q=1,\ldots,Q. (3) Hence, channel matrices can be seen as a concatenation of smaller submatrices such that \bm{H}=[\bm{H}^{(1)},\ldots,\bm{H}^{(Q)}]\in\mathbb{C}^{M_{T}\times\bar{N}Q} and \bm{G}=[\bm{G}^{(1)},\ldots,\bm{G}^{(Q)}]\in\mathbb{C}^{M_{R}\times\bar{N}Q} . Additionally, \bm{S}^{(q)}_{k}\in\mathbb{C}^{N\times N} is the corresponding scattering matrix used in the k -th training block, and \bm{B}_{k} is the additive noise in the receiver modeled as a complex Gaussian random variable with zero mean and unitary variance. The design of BD-RIS training matrices was first addressed in [ 15 ] , which proposed an orthogonal construction for \bar{\bm{S}}\in\mathbb{C}^{K\times\bar{N}^{2}Q} under the block-length condition K\geq\bar{N}^{2}Q . Although this design is effective for cascaded-channel acquisition, it does not provide separate estimates of the Tx–RIS and RIS–Rx channels. A subsequent work [ 3 ] introduced a modified training strategy by modeling the BD-RIS training structure as a third-order tensor, enabling individual channel estimation with reduced training overhead. Motivation : In BD-RIS-assisted systems, the training scattering tensor is intrinsically high-dimensional, and designing it independently at each block leads to a large number of control variables and substantial reconfiguration overhead. This motivates a structured low-dimensional representation that preserves the essential multilinear coupling while reducing implementation complexity. In this work, we adopt a low-rank PARAFAC parameterization of the BD-RIS training tensor to achieve this objective. This modeling choice plays a dual role in the paper: it yields a hardware-feasible training design with fewer parameters, and it also induces a tractable tensor structure in the received pilots that can be exploited for channel estimation. In the following, we derive the tensor signal formulation and then build PARAFAC-based estimators for recovering the individual channel factors. III PARAFAC-based channel estimation This section formulates the proposed channel estimation method and its physical implications for the BD-RIS architecture. By exploiting the tensor structure of the received signal, we formulate a PARAFAC-based iterative block alternating least-squares method that decouple the estimation of \bm{G} and \bm{H} . Then, we discuss solution-existence conditions and their implications for