Large Intelligent Surfaces with Low-End Receivers: From Scaling to Antenna and Panel Selection

Ashkan Sheikhi,Juan Vidal Alegría,Ove Edfors

2024-11-07

Abstract:We analyze the performance of large intelligent surface (LIS) with hardware distortion at its RX-chains. In particular, we consider the memory-less polynomial model for non-ideal hardware and derive analytical expressions for the signal to noise plus distortion ratio after applying maximum ratio combining (MRC) at the LIS. We also study the effect of back-off and automatic gain control on the RX-chains. The derived expressions enable us to evaluate the scalability of LIS when hardware impairments are present. We also study the cost of assuming ideal hardware by analyzing the minimum scaling required to achieve the same performance with a non-ideal hardware. Then, we exploit the analytical expressions to propose optimized antenna selection schemes for LIS and we show that such schemes can improve the performance significantly. In particular, the antenna selection schemes allow the LIS to have lower number of non-ideal RX-chains for signal reception while maintaining a good performance. We also consider a more practical case where the LIS is deployed as a grid of multi-antenna panels, and we propose panel selection schemes to optimize the complexity-performance trade-offs and improve the system overall efficiency.

Signal Processing,Information Theory

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is how to evaluate and optimize the performance of large - scale intelligent surfaces (LIS) when using low - cost hardware in the receiving link. Specifically, the research focuses on the following aspects:

1. **Impact of hardware distortion on LIS performance**:
- The paper analyzes the performance of the system when there is hardware distortion in the receiving link of LIS. In particular, the authors consider the memoryless polynomial model to describe non - ideal hardware and derive an analytical expression for the signal - to - noise - plus - distortion ratio (SNDR) after applying maximum - ratio combining (MRC).
- The research also explores the impact of back - off and automatic gain control (AGC) on the receiving link.

2. **Scalability of LIS**:
- The derived expressions enable researchers to evaluate the scalability of LIS in the presence of hardware impairments. In addition, by analyzing the minimum expansion required to achieve the same performance, the cost of assuming ideal hardware is studied.

3. **Antenna selection schemes**:
- To improve performance, the paper proposes optimized antenna selection schemes for LIS. These schemes allow LIS to reduce the number of non - ideal receiving links used for signal reception while maintaining good performance.

4. **Panel selection schemes**:
- Considering a more practical situation where LIS is deployed as a multi - antenna panel grid, the paper proposes panel selection schemes to optimize the complexity - performance trade - off, thereby improving the overall efficiency of the system.

### Summary of mathematical formulas

- **Memoryless polynomial model**:
\[
y = f(x)=\sum_{k = 0}^{L - 1}a_{2k + 1}x|x|^{2k}
\]
where \(a_{2k+1}\) are the model parameters.

- **Bussgang gain**:
\[
g=\frac{E[y\bar{x}]}{E[|x|^{2}]}=\sum_{k = 0}^{L - 1}a_{2k + 1}(k + 1)!P^{k}
\]

- **Distortion power**:
\[
C_{\eta\eta}=E[|\eta|^{2}]=2\sum_{k = 1}^{L - 1}k!P^{k}\sum_{i = 1}^{k}a_{2i - 1}\bar{a}_{2k - 2i + 1}-|g|^{2}P
\]

- **SNDR expression**:
\[
\gamma=\frac{P\sum_{n = 1}^{N}|\tilde{h}_{n}|^{2}}{\sum_{n = 1}^{N}C_{n}|\tilde{h}_{n}|^{2}+\sigma^{2}}
\]

- **Simplified SNDR expression when considering AGC**:
\[
\gamma=\frac{P|\tilde{g}|^{2}\sum_{n = 1}^{N}|h_{n}|^{2}}{\kappa P\sum_{n = 1}^{N}|h_{n}|^{4}+\sigma^{2}}
\]

Through the research of these problems, the paper aims to provide theoretical basis and technical support for the design and optimization of large - scale intelligent surfaces in future wireless communication systems.