User Grouping for Massive MIMO in FDD Systems: New Design Methods and Analysis
Citations
1,899 citations
282 citations
Cites background from "User Grouping for Massive MIMO in F..."
...namely on inter-group precoding and intra-group precoding, which are capable of reducing the channel estimation overhead while guaranteeing fairness to the UEs [104]....
[...]
206 citations
196 citations
Cites result from "User Grouping for Massive MIMO in F..."
...More importantly, compared with TDD systems, FDD systems can provide more efficient communications with low latency [7], and it has dominated current cellular systems....
[...]
170 citations
References
6,248 citations
6,184 citations
"User Grouping for Massive MIMO in F..." refers background in this paper
...As an emerging and promising technology, large-scale MIMO also enjoys many advantages such as low-power, robust transmissions, simplified transceiver design, and simplified multipleaccess layer [1], [2], in addition to enhanced capacity....
[...]
...…the prebeamforming matrix Bg for group g shall be carefully designed based on all the group centers Rg, g = 1, 2, . . . ,G. Note that the group center can be obtained by averaging the subspace of all the groupmembers or by simply assigning one of the group members to be the group center....
[...]
5,158 citations
"User Grouping for Massive MIMO in F..." refers background in this paper
...However, due to the difficulties of acquiring channel state information at the transmitter side (CSIT), it is challenging to simultaneously support a large number of users [2]....
[...]
...As an emerging and promising technology, large-scale MIMO also enjoys many advantages such as low-power, robust transmissions, simplified transceiver design, and simplified multipleaccess layer [1], [2], in addition to enhanced capacity....
[...]
...…the prebeamforming matrix Bg for group g shall be carefully designed based on all the group centers Rg, g = 1, 2, . . . ,G. Note that the group center can be obtained by averaging the subspace of all the groupmembers or by simply assigning one of the group members to be the group center....
[...]
2,717 citations
1,347 citations
"User Grouping for Massive MIMO in F..." refers background in this paper
...It is therefore of great importance to investigate the massive MIMO design for FDD systems....
[...]
...;3 while ∣∣∣L(n)tot − L(n−1)tot ∣∣∣ > L(n−1)tot do4 Let S(n)g = ∅, g = 1, 2, · · · ,G ;5 for k = 1, 2, · · · ,K do6 for g = 1, 2, · · · ,G do7 Compute L(Rk ,V(n−1)g ) = ∥∥∥∥(Uk3 12k )HV(n−1)g ∥∥∥∥2 F ; 8 end9 Find g∗k = argmaxg′ L(Rk ,V (n−1) g′ ) and let10 S(n)g∗k = S(n)g∗k ∪ {k} ; end11 for g =…...
[...]
...It is useful to evaluate the various combinations to find the best one....
[...]
...B. WEIGHTED LIKELIHOOD SIMILARITY MEASURE Instead of chordal distance, we first propose a weighted likelihood function as the similarity measure between a user and a group, which is defined as L(Rk ,Vg) , ∥∥(Uk3 12k )HVg∥∥2F ....
[...]