# Performance Analysis of Receive Antenna Selection in Multi-Antenna   Systems under Finite Constellation Size

**Authors:** Chongjun Ouyang, Hongwen Yang

arXiv: 1901.03669 · 2019-07-17

## TL;DR

This paper analyzes the performance of receive antenna selection in SIMO systems with finite constellation sizes, providing new closed-form approximations for capacity and error rates that reduce computational complexity while maintaining high accuracy.

## Contribution

It introduces novel asymptotic approximation formulas for symmetric capacity and SER in antenna-selection SIMO systems, applicable to various modulation schemes.

## Key findings

- Approximate expressions closely match simulation results.
- Asymptotic formulas significantly reduce computational complexity.
- Single-antenna selection capacity series are derived for the first time.

## Abstract

Antenna selection (AS) is regarded as one of the most prospective technologies to reduce hardware cost but keep relatively high spectral efficiency in multi-antenna systems. By selecting a subset of antennas to transceive messages, AS greatly alleviates the requirement on RF chains. This paper studies receive antenna selection in single-input multiple-output (SIMO) systems, namely the antenna-selection SIMO (AS-SIMO) systems, from the perspective of digital modulation. The receiver, equipped with multiple antennas, selects an optimal antenna subset to receive messages from the single-antenna transmitter. By assuming independent and identical distributed (i.i.d) flat fading Rayleigh channels, we first analyze the input-output mutual information, also referred as symmetric capacity, of AS-SIMO systems when the modulation style is BPSK/QPSK/16QAM. To reduce the computation complexity of the capacity, closed-form approximated expressions of the symmetric capacity based on asymptotic theory are given for the first time to approach the exact results. Compared with the conventional derivations, our approximation holds much lower computation complexity with the guarantee of high precision. Next, this asymptotic approximation technique is extended to estimate the symbol error rate (SER) of antenna-selection SIMO systems and approximated expressions for SER are proposed which indicates much lower complexity. Finally, a special scenario of single-antenna-selection is detailedly investigated and series expressions of the symmetric capacity are formulated for the first time. Beside analytical derivations, simulation results are provided to demonstrate the approximation precision of the derived results. Experiment results show that the asymptotic theory has a remarkable approximation effect.

---
Source: https://tomesphere.com/paper/1901.03669