Efficient Channel Prediction for Beyond Diagonal RIS-Assisted MIMO Systems with Channel Aging

Nipuni Ginige,Arthur Sousa de Sena,Nurul Huda Mahmood,Nandana Rajatheva,Matti Latva-aho

2024-11-21

Abstract:Novel reconfigurable intelligent surface (RIS) architectures, known as beyond diagonal RISs (BD-RISs), have been proposed to enhance reflection efficiency and expand RIS capabilities. However, their passive nature, non-diagonal reflection matrix, and the large number of coupled reflecting elements complicate the channel state information (CSI) estimation process. The challenge further escalates in scenarios with fast-varying channels. In this paper, we address this challenge by proposing novel joint channel estimation and prediction strategies with low overhead and high accuracy for two different RIS architectures in a BD-RIS-assisted multiple-input multiple-output system under correlated fast-fading environments with channel aging. The channel estimation procedure utilizes the Tucker2 decomposition with bilinear alternative least squares, which is exploited to decompose the cascade channels of the BD-RIS-assisted system into effective channels of reduced dimension. The channel prediction framework is based on a convolutional neural network combined with an autoregressive predictor. The estimated/predicted CSI is then utilized to optimize the RIS phase shifts aiming at the maximization of the downlink sum rate. Insightful simulation results demonstrate that our proposed approach is robust to channel aging, and exhibits a high estimation accuracy. Moreover, our scheme can deliver a high average downlink sum rate, outperforming other state-of-the-art channel estimation methods. The results also reveal a remarkable reduction in pilot overhead of up to 98\% compared to baseline schemes, all imposing low computational complexity.

Signal Processing

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to provide efficient and accurate channel state information (CSI) estimation and prediction for the beyond - diagonal reconfigurable intelligent surface (BD - RIS) - assisted multi - input multi - output (MIMO) system in a rapidly changing environment with channel aging. Specifically:

1. **Challenges**:
- Due to its non - diagonal reflection matrix and a large number of coupled reflection units, BD - RIS complicates the CSI estimation process.
- In a rapidly changing channel environment (such as the Doppler effect caused by user movement), the channel aging problem exacerbates this challenge.
- Traditional CSI estimation methods have large pilot overhead and computational complexity in BD - RIS - assisted MIMO systems.

2. **Solutions**:
- A joint channel estimation method based on Tucker2 decomposition and bilinear alternating least squares (BALS) is proposed to reduce the dimension of parameter estimation and improve the estimation accuracy.
- A channel prediction framework combining convolutional neural network (CNN) and autoregressive (AR) predictor is introduced to deal with the channel aging problem and further reduce the pilot overhead.

3. **Objectives**:
- Achieve low - overhead, high - accuracy CSI estimation and prediction.
- Maximize the total downlink rate and optimize the RIS phase - shift configuration.

### Specific Problem Description

- **Channel Estimation**: Use Tucker2 decomposition to decompose the cascaded channel of the BD - RIS - assisted system into an effective channel, thereby reducing the dimension and improving the estimation efficiency.
- **Channel Prediction**: Predict the channel state through the CNN - AR model, reduce the need for frequent pilot training, and deal with channel aging.
- **Optimize RIS Phase - shift**: Use the estimated / predicted CSI to optimize the RIS phase - shift configuration to maximize the total downlink rate.

### Mathematical Formula Representation

- **Tucker2 Decomposition**:
\[
X = G\times_1 A^{(1)}\times_2 A^{(2)}
\]
where \(X\in\mathbb{C}^{I_1\times I_2\times I_3}\) is the original third - order tensor, \(G\in\mathbb{C}^{R_1\times R_2\times I_3}\) is the core tensor, and \(A^{(1)}\in\mathbb{C}^{I_1\times R_1}\) and \(A^{(2)}\in\mathbb{C}^{I_2\times R_2}\) are factor matrices.

- **Mode - Unfolding Matrix**:
\[
[X]_{(1)} = A^{(1)}[G]_{(1)}(I_{I_3}\otimes A^{(2)})^T
\]
\[
[X]_{(2)} = A^{(2)}[G]_{(2)}(I_{I_3}\otimes A^{(1)})^T
\]

- **Channel Model**:
\[
e_k[l]=\sqrt{\beta_e}R_{\text{RIS}}^{1/2}q_{e,k}e^{j2\pi f_dT_sl}
\]
\[
H = \sqrt{\beta_H}R_{\text{BS}}^{1/2}Q_HR_{\text{RIS}}^{1/2}
\]
where \(R_{\text{RIS}}\) and \(R_{\text{BS}}\) are the deterministic correlation matrices of RIS and base station respectively, \(q_{e,k}\) is the Rician fading vector, and \(\beta_e\) and \(\beta_H\) are the large - scale fading factors respectively.

Through these methods, the paper aims to provide an efficient, low - overhead and accurate CSI estimation and prediction scheme for BD - RIS - assisted MIMO systems, thereby improving the performance and reliability of the system.