Open AccessInterference alignment in two‐hop multiple‐input multiple‐output interference channel under various CSI and antenna correlation
Author(s) -
HosseiniFakhr Seyyed Hossein,
Falahati Abolfazl
Publication year - 2020
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/iet-com.2020.0177
Subject(s) - computer science , telecommunications link , interference alignment , interference (communication) , algorithm , zero forcing precoding , channel state information , channel (broadcasting) , control theory (sociology) , topology (electrical circuits) , precoding , mathematics , telecommunications , mimo , wireless , control (management) , combinatorics , artificial intelligence
In this paper, the problem of interference management is investigated in the downlink of multi‐user multiple‐input multiple‐output two‐hop interference channel under various scenarios of channel state information (CSI) and antenna correlation. Interference alignment (IA) in the space dimension is a powerful tool to solve this problem. Iterative IA algorithms are adopted in both the first and the second hops transmissions independently. After aligning interfering signals, the recovery of the transmitted signal vector is performed based on zero‐forcing and semi‐definite relaxation detectors. Furthermore, the impacts of the imperfect CSI and antenna correlation on sum‐rate are investigated for decode‐and‐forward (DF) and amplify‐and‐forward relaying protocols. When the system is proper but IA is not completely feasible, the performance of minimum interference leakage (MIL) is compared with rank‐constrained rank minimisation (RCRM) algorithm in DF protocol. RCRM algorithm is more sensitive to CSI such that the channel uncertainty easily leads to an increase in the interference matrix rank. According to the simulation results, for the case of perfect CSI, RCRM makes further interference‐free dimensions which leads to better performance than MIL, but for the case of imperfect CSI, close performance is observed.