Finding MIMO: how did it get Massive, and what’s it all about?
Many of you will be familiar with the term ‘multiple-input, multiple-output’, or MIMO. It took root with the introduction of LTE, and evolved into higher order MIMO, then ‘full dimension’ MIMO. The latest iteration is ‘Massive MIMO’, which is often referenced in the 5G narrative.
But what’s the difference between all these MIMO variants? Is it better to “beamform” or “beamsteer”? And what performance benefits can we expect in the traditional frequency bands and new mmWave bands?
Multi-path propagation
The first MIMO specifications emerged in 3GPP standards at the end of the 3G era, but never quite caught on. It was only in 2008 when LTE was introduced that MIMO in cellular radio started to be taken seriously. The basic principle is using multiple antennas at each end of the radio channel, to ‘cheat’ the Shannon-Hartley channel capacity theorem by exploiting multi-path propagation. The goal is to increase data rates by spatial multiplexing or sending multiple data streams at the same time in the same frequency.
In a single antenna system, sending multiple data streams simply results in interference. But with Multi-User MIMO, the signals transmitted from each antenna take different paths to the receivers. By applying the right mix of each data stream to each transmit antenna, the signals received at each receiving antenna will only see one of the original data streams, which have become unscrambled by the channel’s propagation characteristics.
An example, of a simple system that can support spatial multiplexing is one in which the base station (BS) has two transmitters each with a separate physical antenna and the user equipment (UE) has two receivers—also each with a separate physical antenna. The simplest scenario in this system is one in which the path from BS transmitters 1 and 2 towards UE receiver 1 are identical, the path from BS transmitter 2 to UE receiver 1 is also the same, but the path from BS transmitter 2 to UE receiver 2 has a 180-degree phase inversion.
In such a scenario, if we have two data streams (A and B), and we transmit A + B from transmitter 1 and A - B from transmitter 2, then receiver 1 will see:
(A+B) + (A-B) = 2A
and receiver 2 will see:
(A+B) – (A-B) = 2B
This increases the overall data rate for the path between the two systems (channel capacity) and improves the signal-to-noise ratio (SNR) for each stream.
Seeking perfection
Such a perfect multipath channel scenario rarely exists. For MIMO to work, it’s necessary to know the actual channel propagation conditions, and these will vary between each transmitter and each receiver in time, frequency, phase and interference level.
Provided the channel between the base station (BS) and user equipment (UE) is not identical for each receiver, it is mathematically possible to precode the transmitted signals with the inverse of the channel, as shown in the simple example above. However, the inability to precisely know the channel conditions, coupled with the existence of noise or interference, means the ability to mathematically recover both streams is a function of 1) how well the channel supports orthogonal paths; and 2) the signal to interference and noise (SINR) level.
The 2x2 principle
In a practical system, for any given SINR, if 2x2 MIMO is used, the signal power has to be shared between the transmitters, which decreases the signal to noise and interference ratio (SINR) by 3 dB. As such, MIMO can only provide a meaningful gain over single input single output (SISO) when the SINR of the channel gets higher than is needed to support the maximum SISO data rate.
These high-SINR conditions occur when the user is near the cell center, or when interference from adjacent cells is low. In a typical urban macro environment, a 2x2 MIMO setup provides around 20% gain over SISO, not the 2x (100%) theoretical gain. However, MIMO gains do move towards the 2x limit as channel conditions improve.
This 2x2 principle can be increased to higher orders, by adding more antennas at each end of the link. In the 2008 3GPP Release 8 LTE standard, 2x and 4x operation was specified, and 8x8 was added later.
User equipment matters
However, there are limitations with devices such as smartphones. It’s increasingly difficult to support higher-order MIMO because the handsets are simply not large enough to support multiple receive antennas. But this limitation does not mean the end of MIMO.
Instead, it points to an alternative form of MIMO where, rather than transmitting multiple streams of data to one user to increase peak data rates, the same number of streams can be transmitted towards multiple users, with each getting one stream. This is known a multi-user-MIMO (MU-MIMO) and has the effect of increasing cell capacity but does not increase peak data rates to any given user.
Evolving to full dimension
The next MIMO evolution came in 3GPP Release 13 with full dimension MIMO (FD-MIMO), also known as ‘elevation beamforming’. In previous MIMO incarnations, the cellular world was ‘flat’, existing in two dimensions. Only the azimuth domain was considered in channel models, and base stations had no active elevation controls on their antenna systems.
But with FD-MIMO, BS antenna configurations can exploit the elevation domain too. Antenna arrays of up to 32 cross-polarized elements are specified, leading to the development of 3D spatial channel models. This degree of beam control also gave the possibility of simple beamsteering of signals towards specific directions.
Beamsteering is a subset of general MIMO beamforming using channel inversion, and it occurs when the channel is predominantly line-of-sight and the UEs are spatially separated. If beamsteering is enabled through the availability of sufficient BS antenna elements and a sparse line of sight channel, it has the advantage that precise channel state information is not required: just the direction of the UE, which can be established from simple phased array antenna systems. In contrast, the more complex MIMO beamforming requires full digital control of every element in the antenna array, to minimize crosstalk between MIMO streams at the UE.
Massive MIMO
The final piece of the MIMO story is when it gets “massive”. The term “Massive MIMO” has been around for a few years now and has come to mean many things. For the sake of this discussion, it is best explained as a form of multi-user MIMO, where 1) the number of BS antenna elements (backed up by independent-but-coordinated transceivers) is much greater than the number of UE’s; and 2) this typically means tens and likely over 100 BS antenna elements with the number of UE’s at less than 30% of this. In the context of 5G NR, there are updates to the approach of sounding and estimating the channel which are not feasible in LTE.
Experiments in research lab conditions have shown the potential of M-MIMO to offer around 50 times the spectral efficiency of a single input, single-output system. Initial field trials are promising, but the big question is whether the lab conditions that demonstrated the theory can scale to commercial systems.
That brings the MIMO story up to date. To summarize, massive MIMO is best suited to what are being called “mid-band” 5G frequencies (3-6GHz), whereas SISO beamsteering is more suited to high-frequency mmWave bands where the channel is sparse, and beamforming offers minimal gains over simpler beamsteering. Time will tell how far along the M-MIMO road the industry will go.
Find out more about MIMO, and how Keysight can accelerate 5G testing and innovation here.