List your company
ResearchFormerly WizWand

Beamforming Sum-Rate Maximization on Rayleigh fading realizations (M=8, K=4, Pmax=1, sigma^2=0.1) (test)

14.81Sum-Rate (bits/s/Hz)PGD (200-it)
1.05084.62298.19511.7671Mar 18, 2026
Updated 7mo ago

Evaluation results

MethodLinks
PGD (200-it)training-size independent=true, Iterations=2002026.0314.81
Auto-PGDTraining set size (N)=10^3, Number of unrolled layers=52026.0314.63
Auto-PGDTraining set size (N)=10^2, Number of unrolled layers=52026.0314.49
Auto-PGDTraining set size (N)=10^4, Number of unrolled layers=52026.0313.49
ZFtraining-size independent=true2026.0313.33
PGD-NetTraining set size (N)=10^2, Number of unrolled layers=102026.0313.28
PGD-NetTraining set size (N)=10^3, Number of unrolled layers=102026.0312.74
Auto-PGDTraining set size (N)=5 × 10^4, Number of unrolled layers=52026.0311.87
PGD-NetTraining set size (N)=10^4, Number of unrolled layers=102026.0311.83
PGD-NetTraining set size (N)=5 × 10^4, Number of unrolled layers=102026.0311.64
LISTATraining set size (N)=5 × 10^42026.036.98
LISTATraining set size (N)=10^32026.036.46
MLPTraining set size (N)=5 × 10^42026.036.43
LISTATraining set size (N)=10^42026.036.14
MLPTraining set size (N)=10^42026.034.49
MLPTraining set size (N)=10^32026.032.72
LISTATraining set size (N)=10^22026.032.14
MLPTraining set size (N)=10^22026.031.58