Open AccessStabilised blind interference suppression of constant modulus‐based adaptive array over Rayleigh fading channels by initial antenna selection
Author(s) -
Maruta Kazuki,
Ahn ChangJun
Publication year - 2019
Publication title -
iet microwaves, antennas and propagation
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.555
H-Index - 69
eISSN - 1751-8733
pISSN - 1751-8725
DOI - 10.1049/iet-map.2018.5795
Subject(s) - rayleigh fading , interference (communication) , antenna (radio) , electronic engineering , quadrature amplitude modulation , qam , computer science , fading , control theory (sociology) , acoustics , telecommunications , channel (broadcasting) , physics , engineering , bit error rate , control (management) , artificial intelligence
This article proposes an initial antenna selection for constant modulus‐based blind interference suppression algorithms to stabilise their prominent interference suppression performance. Constant modulus algorithm (CMA) is a well‐known blind adaptive array scheme, but it cannot suppress the interference when signal‐to‐interference power ratio (SIR) is <0 dB. It faces strong limits on their applicable region over Rayleigh fading channels where instantaneous channel gain fluctuates over 10 dB range. Even if the expected desired signal strength is larger than interference, CMA may still miscapture the interference and incorrectly suppress the desired signal. The authors’ proposal is simplified approach to select the antenna element whose reception power is maximal. Certain antenna element can be expected to capture the desired signal precisely under the condition where SIR is statistically positive, so that the CMA processor can utilise it as an initial input. This condition will be ensured by increasing antenna element number, supported by the trends that base station antenna elements are going massive. Computer simulations verifies improved interference suppression performance provided by the proposed scheme. This article also verified applicability for multi‐modulus algorithm to accommodate M‐ary quadrature amplitude modulation (QAM) use.