Effect of feedback delay and channel gain difference on distributed beamforming

The performances of distributed beamforming (BF) with feedback delay have been evaluated for the several scenarios of the distributed channel gain differences. The performance degradations by feedback delay and by using codebook-based BF are analyzed according to the channel gain difference. As the most complexity-reduced alternative, selective power allocation (PA) is also compared with the full feedback-based BF and the codebook-based BF. Based on the results, we note several distinguishable behavior of the distributed BF compared to the clustered BF. 1) Large differences among the channel gains (most typical scenario of distributed BF) suppress the feedback delay effect on performance degradation compared to the case of no differences among the channel gains (similar to the case of clustered BF). 2) However, performance degradation by codebook-based BF becomes more severe in the case of large differences among the channel gains compared to the case of small differences among the channel gains irrespective of feedback delay. With large differences among the channel gains, codebook-based BF gets even worse than the selective PA in distributed BF. 3) For the case of large differences among the channel gains, selective PA is a good alternative to the full feedback-based BF. However, this is only for the case of small feedback delay. The performance gap becomes non-negligible as the feedback delay increases. On the other hand, for the case of no or small channel gain differences, selective PA is a good alternative to the full feedback-based BF as the feedback delay increases.