Fully Decentralized Approximate Zero-Forcing Precoding for Massive MIMO Systems

Research output: Contribution to journalArticle


We analyze the downlink of a massive multiuser multiple input, multiple output (MIMO) system where antenna units at the base station are connected in a daisy chain without a central processing unit and only possess local channel knowledge. For this setup, we develop and analyze a linear precoding algorithm for suppressing interuser interference. It is demonstrated that the algorithm is close to zero-forcing precoding in terms of performance for a large number of antennas. Moreover, we show that with careful scheduling of processing across antennas, requirements for interconnection throughput are reduced compared with the fully centralized solution. Favorable tradeoff between performance and interconnection throughput makes the daisy chain a viable candidate topology for real-life implementations of base stations in MIMO systems where the number of antennas is very large.


Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Telecommunications


  • daisy chain, decentralized processing, interference suppression, Massive MIMO
Original languageEnglish
Article number8607077
Pages (from-to)773-776
Number of pages4
JournalIEEE Wireless Communications Letters
Issue number3
Publication statusPublished - 2019 Jun
Publication categoryResearch