Distributed massive MIMO: Algorithms, architectures and concept systems

Abstract

Making MIMO truly "massive" involves liberating it from the shackles of form factor constraints, by allowing arbitrarily large groups of neighboring nodes to opportunistically form virtual antenna arrays for both transmission and reception. Moving such distributed MIMO (DMIMO) systems from the realm of information theory to practice requires synchronization of the cooperating nodes at multiple levels. We are interested in all-wireless systems which are severely constrained in the amount of information that can be exchanged among the cooperating nodes, in contrast to recent proposals in massive MIMO (co-located antennas) or base station cooperation (which relies on a high-speed wired backhaul). The goal of this paper is to point out some of the research issues unique to scaling up such DMIMO systems. We briefly review the significant technical progress in design and demonstration over the past few years, and describe a research agenda for the next few years based on fundamental questions in attaining the "distributed coherence" required to realize concept systems such as DMIMO communication at large carrier wavelengths (e.g., white space frequencies for which standard antenna arrays are too bulky) and distributed 911 for emergency and rescue scenarios. © 2014 IEEE.