Making sense of MIMO

MIMO (multiple input, multiple output) has been around for a long time, but 5G has spawned a range of MIMO-related acronyms, says Sara Ghaemi of Avnet Abacus.

Stepping back into the world of long-term evolution (LTE), MIMO usually refers to single user MIMO (SU-MIMO). This is where each user/device waits in turn to use base station resources. The multiple antenna ports and antennas transmit simultaneously with multiple data streams using the same time/frequency resources. Increasing the peak throughput per user or device is achieved by using those simultaneous multiple streams, doubling (2×2 MIMO), or quadrupling (4×4 MIMO) the throughput.

The disadvantage to this architecture is the network congestion, where large numbers of devices, for example, in crowded urban situations such as train stations or airports, overwhelm individual base stations, especially in high bandwidth request situations. The advantage is that there is little interference for the small number of concurrent users that can be supported. Those users will get – at least initially – a high quality service in capacity terms.

MU-MIMO

Multi user MIMO (MU-MIMO) allocates an antenna port to each user or device. This increases the cell capacity while using the same time frequency resources. The same technique has been applied to Wi-Fi, enabling single routers to handle multiple devices more effectively. In an enterprise or business environment, or the home, this prevents unnecessary buffering.

MU-MIMO hinges on spatial separation between devices or users, which makes traditional broadcast methods less than ideal. It is here that beamforming comes in. Originally proposed in the 1940s to focus a wireless signal towards the receiving device, beamforming relies on densely packed multiple antennas, all broadcasting the same signal at slightly different times, creating an overlapping mesh of waves. This creates areas of interference that can be minutely controlled to focus the signal in the direction of a specific device.

That focus not only gives the end device an improved signal, but also reduces interference between devices, making it an ideal technology for dense, urban-style networks and a cornerstone of 5G. The downside to beamforming is that it is fairly compute-intensive, which originally meant that the time and power resources were prohibitive, but improvements in processor efficiency have made it viable in a range of situations.

mMIMO

Massive MIMO, or mMIMO is a MU-MIMO technology in that it supports multiple users. It is a core component of 5G networking and likely to play a starring role in upcoming wireless network standards too. The ‘massive’ refers to the number of antennas in a given base station array, rather than their physical size, which can serve many terminals simultaneously. Overall, it scales up the original MIMO concept considerably, which is an essential addition for the demands of IIoT and Industry 4.0. It’s thought the term was originally first used in Thomas Marzetta’s 2010 paper ‘Noncooperative Cellular Wireless with Unlimited Numbers of Base Station Antennas’ and subsequently popularised by other researchers.

It is so key that a recent report from MarketInsights valued the global mMIMO market at $1.09bn in 2019 and projected it to reach $15.79bn by 2027, growing at a CAGR of 35.3% from 2020 to 2027. In Europe, analysts tipped Germany to be among the leaders, with an above average CAGR of 36.9%.

Challenges remain

The real world demands of 5G mMIMO have proven to be something of a thorny issue for operators as they roll out networks as quickly and as cost effectively as possible. The temptation has been to use lightweight 5G radio units, featuring eight transmitters and receivers per unit (8T8R), as opposed to the much larger and powerful 64 transmitters and receivers per unit (64T64R) boxes that weigh more than 60kg. The 8T8R units need far less power to operate, but offer potentially shorter lifespans before needing to be upgraded. A newer configuration, 32 transmitters and receivers (32T32R), offers a more efficient balance between capacity, weight and power. This has become available only relatively recently, highlighting the difficulties of balancing network rollout today against potential demand tomorrow. Some industry experts estimate that 32T32R mMIMO will boost possible data rates by about 50% using Release 16 handsets, but could nearly double the data rate when Release 18 handsets come into play.

Even with today’s silicon technology, however, mMIMO remains an extremely demanding radio technology, pushing the envelope of what is currently practical in real world deployment.

As operators deploy mMIMO, lessons are being learnt and actively fed back into future 3GPP 5G releases to play a role in iterating and evolving the 5G NR standards.
Visit at electronica: C2-101