Revisiting Degrees of Freedom of Full-Duplex Systems with Opportunistic Transmission: An Improved User Scaling Law
It was recently studied how to achieve the optimal degrees of freedom (DoF) in a multi-antenna full-duplex system with partial channel state information (CSI). In this paper, we revisit the DoF of a multiple-antenna full-duplex system using opportunistic transmission under the partial CSI, in which...
Gespeichert in:
Veröffentlicht in: | Entropy (Basel, Switzerland) Switzerland), 2018-03, Vol.20 (3), p.160 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | It was recently studied how to achieve the optimal degrees of freedom (DoF) in a multi-antenna full-duplex system with partial channel state information (CSI). In this paper, we revisit the DoF of a multiple-antenna full-duplex system using opportunistic transmission under the partial CSI, in which a full-duplex base station having M transmit antennas and M receive antennas supports a set of half-duplex mobile stations (MSs) having a single antenna each. Assuming no self-interference, we present a new hybrid opportunistic scheduling method that achieves the optimal sum DoF under an improved user scaling law. Unlike the state-of-the-art scheduling method, our method is designed in the sense that the scheduling role between downlink MSs and uplink MSs is well-balanced. It is shown that the optimal sum DoF of 2M is asymptotically achievable provided that the number of MSs scales faster than SNRM , where SNR denotes the signal-to-noise ratio. This result reveals that, in our full-duplex system, better performance on the user scaling law can be obtained without extra CSI, compared to the prior work that showed the required user scaling condition (i.e., the minimum number of MSs for guaranteeing the optimal DoF) of SNR2M−1 . Moreover, the average interference decaying rate is analyzed. Numerical evaluation is performed to not only validate our analysis but also show superiority of the proposed method over the state-of-the-art method. |
---|---|
ISSN: | 1099-4300 1099-4300 |
DOI: | 10.3390/e20030160 |