.png){kind=icon}
Return
Beam-structured precoding for network massive MIMO systems via Hamiltonian-based optimization
.png){kind=icon}
PRE
AI
DOI：10.1007/s11432-025-4666-y.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Massive multiple-input multiple-output (MIMO) has received widespread recognition for its substantially improved spectral efficiency, and it remains a fundamental technology in future wireless communication networks. Its extension to network massive MIMO enables joint transmission across base stations (BSs), but also introduces significant challenges due to the high dimension of the channel matrices and the associated optimization variables. Our work investigates the precoder design by leveraging beam-structured precoding under a Hamiltonian-based framework. We begin by introducing a beam-based channel model and formulating the precoder design problem in the beam domain. Then we show that the optimal beam-domain precoder for each user terminal (UT) only occupies beams corresponding to its non-zero beam-domain channel elements, a design referred to as beam-structured precoding, which results in a lower-dimensional optimization problem. The corresponding problem is handled using a Hamiltonian system, where the objective function is interpreted as potential energy, transforming the optimization problem into a physical system’s energy minimization task. The system’s dynamical equations are then solved numerically with a RATTLE integrator, providing a principled approach to explore the solution space, with reduced computational complexity and favorable performance. Through simulation results, we verify the effectiveness of our method by demonstrating notable complexity savings while maintaining high performance.
Keywords:
network massive MIMO transmission
beam-structured precoding
sum-rate maximization
Hamiltonian-based optimization
Journal
IF:
7.6
Papers:
4.9K
Citations:
8.9K
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSave
ShareSave
ShareSave
View more