کنفرانس مهندسی برق ایران

صفحه اصلی / سی و چهارمین کنفرانس بین المللی مهندسی برق

Newton-Based Mini-Batch Detection in Decentralized Large-Scale Antenna Systems

نویسندگان :

Ghazal Mousavi¹ Mojtaba Amiri² Ali Olfat³

1- School of Electrical and Computer Engineering, University of Tehran 2- School of Electrical and Computer Engineering, University of Tehran 3- School of Electrical and Computer Engineering, University of Tehran

کلمات کلیدی :

Large-scale MIMO،decentralized architectures،Newton algorithm،Detection

چکیده :

Large-scale multiple-input multiple-output (MIMO) systems are a key technology for 5G and beyond (5GB), as they have demonstrated significant improvements in both power and spectral efficiency. However, algorithms for massive MIMO uplink detection typically rely on centralized processing, which introduces bottlenecks in computational complexity and interconnect bandwidth. To address this challenge, decentralized architectures provide an effective solution. In such architectures, the antenna array is partitioned into clusters, each supported by dedicated computing hardware to enable parallel processing. In this paper, a decentralized architecture is considered, in which a mini-batch Newton-based algorithm is developed and applied using Markov Chain Monte Carlo (MCMC). The proposed method achieves remarkable performance and high efficiency compared to existing distributed methods. Additionally, the approach offers advantages in computational latency and interconnect bandwidth compared to centralized implementations.