In 5G, Massive MIMO is considered as the holy grail to increase individual datarates, to improve cell capacity, to extend range, or any combination of these. mMIMO is not a single technique, however, there are a number of different flavors that can to a degree be mixed and matched to achieve different things.
I recently came across this interesting primer by Keysight on the RCR Wireless News website that puts together an overview and high level description of the different forms of mMIMO with 32 or 64 individual antenna elements per base station sector and how they are likely to be used in practice.
mMIMO for Beam Forming
In a few paragraphs, the article gives an an intro on how mMIMO is used for beam sweeping and beam forming to increase the range of a cell and to improve data rates of users closer to the cell by focusing the RF energy into their direction. The important thing to realize about this form of mMIMO is mobile devices no longer only see cells but many beams of a single cell pointed in different directions, some of which they can detect and some of which they can’t. Each beam is identified by a Synchronization Signal Block Beam id (SSB id). In other words, the UE has to no longer only keep track of the cell id of a cell site, but has to measure and keep a list of cell id + several SSB ids per site. And to put it yet into other words, a single cell is now split up into several smaller parts.
mMIMO for Simultaneous Data Transmissions
Another use of mMIMO is to direct beams with different data streams to users that are located in different directions from the point of view of the base station. Transmit power has to be shared so this does not increase range but allows to serve several mobile devices that are close-by simultaneously, which improves overall cell capacity.
2×2 or 4×4 single user MIMO used in LTE today is already quite computationally expensive compared to previous systems, and at least from a layman’s point of view, it looks like mMIMO will once again increase processing requirements of base stations by an order of magnitude. But then, almost 10 years have passed since the introduction of 2×2 MIMO with LTE so it can be expected that new specialized processing hardware in the baseband and RF part of base station will be able to cope. It will be interesting to see over the coming years, if mMIMO will bring the same real world speed and capacity improvements like 2×2 MIMO did over 3G when it was introduced in LTE.