A Hybrid Optimization Approach for Interference Alignment in Multi-User MIMO Relay Networks Under Different CSI

For the next generation of wireless communication network, a heterogeneous network architecture with macro cells, small cells, and relays offers numerous advantages, such as higher spectrum efficiency, better fairness in resource allocation, and energy efficient usage as well as optimal coverage. Ho...

Full description

Saved in:
Bibliographic Details
Published inIEEE transactions on wireless communications Vol. 16; no. 12; pp. 7834 - 7847
Main Authors Chung Le, Moghaddamnia, Sanam, Peissig, Jurgen K.
Format Journal Article
LanguageEnglish
Published IEEE 01.12.2017
Subjects
Antennas
imperfect CSI
Interference
interference alignment
interference mitigation
MIMO
Multi-user AF-MIMO relay
Relay networks (telecommunications)
Robustness
Wireless communication
Online AccessGet full text

Cover

More Information
Summary:For the next generation of wireless communication network, a heterogeneous network architecture with macro cells, small cells, and relays offers numerous advantages, such as higher spectrum efficiency, better fairness in resource allocation, and energy efficient usage as well as optimal coverage. However, this approach encounters various network management challenges, including interference mitigation techniques. In this paper, we consider a multi-user half-duplex amplify-and-forward-MIMO relay system, where a macro basis station communicates with edge cell users via relays and each node has multiple antennas. In this context, we propose two interference alignment (IA) schemes, for which a hybrid approach of the problem formulation based on zero-forcing (ZF) and minimum-mean-square-error optimization criteria was developed to deal with interferences in perfect and imperfect channel state information (CSI) cases. In the case of perfect CSI, simulation results show that this approach provides a SNR gain of about 4 dB regarding BER performance in comparison to that achieved by the reference approach, where all filters are determined solely based on ZF criterion. Furthermore, the achieved capacity and effective signal-to-interference-plus-noise ratio at each edge node are improved by around 5 dB. Having CSI errors, the system performance is significantly improved using the robust design IA approach by minimizing the expected value of cost functions.
ISSN:1536-1276
1558-2248
DOI:10.1109/TWC.2017.2753776