Technology Category
Machine Learning: Distributed Machine Learning & Federated LearningSearch and Optimization: Distributed SearchMultiagent Systems: Mechanism Design
Adaptive Local Combining with Decentralized Decoding for Distributed Massive MIMO
🤖 AI Summary
To address the poor local combining performance, high fronthaul overhead, and excessive computational complexity in uplink distributed massive MIMO, this paper proposes a decentralized decoding architecture: each access point independently designs interference-aware local combining weights and performs local signal combining, abandoning the conventional large-scale fading decoding paradigm reliant on global channel state information. The core contribution is the Adaptive Generalized Local Zero-Forcing (AG-LZF) framework, which dynamically classifies strong and weak users via closed-form spectral efficiency optimization—eliminating the need for fixed thresholds or global coordination. Additionally, a pilot-dependent weighted vector sharing mechanism is introduced, accompanied by a closed-form spectral efficiency analysis. Simulation results demonstrate that the proposed scheme achieves negligible performance loss (<0.5 bps/Hz), reduces fronthaul overhead by ~70%, lowers computational complexity by an order of magnitude, and significantly outperforms fixed-threshold benchmark methods in spectral efficiency.
Technology Category
Application Category
📝 Abstract
A major bottleneck in uplink distributed massive multiple-input multiple-output networks is the sub-optimal performance of local combining schemes, coupled with high fronthaul load and computational cost inherent in centralized large scale fading decoding (LSFD) architectures. This paper introduces a decentralized decoding architecture that fundamentally breaks from the conventional LSFD, by allowing each AP calculates interference-suppressing local weights independently and applies them to its data estimates before transmission. Furthermore, two generalized local zero-forcing (ZF) framework, generalized partial full-pilot ZF (G-PFZF) and generalized protected weak PFZF (G-PWPFZF), are introduced, where each access point (AP) adaptively and independently determines its combining strategy through a local sum spectral efficiency optimization that classifies user equipments (UEs) as strong or weak using only local information, eliminating the fixed thresholds used in PFZF and PWPFZF. To further enhance scalability, pilot-dependent combining vectors instead of user-dependent ones are introduced and are shared among users with the same pilot. The corresponding closed-form spectral efficiency expressions are derived. Numerical results show that the proposed generalized schemes consistently outperform fixed-threshold counterparts, while the introduction of local weights yields lower overhead and computation costs with minimal performance penalty compared to them.
Problem
Research questions and friction points this paper is trying to address.
Improving sub-optimal performance in distributed massive MIMO uplink
Reducing high fronthaul load and computational costs
Eliminating fixed thresholds in user classification for combining
Innovation
Methods, ideas, or system contributions that make the work stand out.
Decentralized decoding with independent interference-suppressing local weights
Adaptive local ZF frameworks classifying UEs as strong or weak
Pilot-dependent combining vectors shared among same-pilot users
🔎 Similar Papers
No similar papers found.
