Analog Least Mean Square Loop for Self-Interference Cancellation in In-Band Full-Duplex Radios

Publication Type:
Thesis
Issue Date:
2020
Full metadata record
Recently, In-band full-duplex (IBFD) transmission, which allows transceivers to transmit and receive simultaneously on a single frequency band, is regarded as a promising solution for the problem of frequency spectrum shortage. However, a fundamental challenge encountered in realizing IBFD communications is self-interference (SI), which is the strong interference imposed by the transmitter blocking its co-located receiver from the signal of interest. Therefore, to enable the IBFD mode, great efforts have been devoted to mitigate SI to beyond the noise floor level. Among various approaches proposed in the radio frequency (RF) domain, analog least mean square (ALMS) loop is a promising structure for SI cancellation (SIC) due to its simplicity and efficiency. However, the behaviours of the ALMS loop have not been fully understood and its application was proposed for single-carrier and single antenna IBFD communication systems only. This study aims at tackling the problem of SI in the RF domain for various IBFD systems using the ALMS loop. The contributions of this thesis are as follows. Firstly, the performances of the ALMS loop with different transmitted signals is investigated. It shows that due to the cyclostationary effect of the transmitted signals, SI cannot be removed completely by the ALMS loop but there exists an irreducible SI. The lower bounds of this irreducible SI are derived for both single-carrier and multi-carrier IBFD systems. Additionally, it proves that the ALMS loop also performs well with deterministic signals in full-duplex synthetic aperture radars. Secondly, by characterizing the ALMS loop in the frequency domain, the achievable levels of SIC by the ALMS loop in both analog and digital domains are revealed. Thirdly, the performance of the ALMS loop under hardware impairment conditions is investigated. More importantly, a degradation bound is found to determine how much of compensation should be obtained from other means of SI mitigation for a given level of imperfection. Fourthly, a novel beam-based analog SIC structure employing the ALMS loop is proposed for IBFD multiple input multiple output (MIMO) systems to significantly reduce the number of adaptive filters required for SIC in IBFD MIMO systems. Finally, a practical structure of the ALMS loop is proposed and a prototype is implemented using off-the-shelf components to provide experimental results confirming all the theoretical findings. The analyses and practical results in this thesis provide a comprehensive view of the ALMS loop and prove its potential application for SIC in different IBFD radios.
Please use this identifier to cite or link to this item: