Table 8 Average runtime required to provide optimal beam pairs at \(T'=99\) sampled locations for different beam training algorithms
Beam training algorithms | Time (s) | Reward |
|---|---|---|
DRL-3 | 7.12 | − 0.17 |
DRL-5 | 7.11 | − 0.17 |
DRL-7 | 7.22 | − 0.18 |
MAB | 7.07 | − 0.20 |
RAND | 7.39 | − 0.60 |
MR | 46.56 | − 1.43 |