MIMO, or multiple input multiple output, is a technique where multiple antennas are used at both the transmitter and the receiver to increase the link reliability, the spectral efficiency, or both. This concept has been around for many years but its use in wireless standards is more recent. This is probably due in part to the fact that OFDM (orthogonal frequency-division multiplexing), which facilitates the implementation of MIMO, is now commonly used in today’s wireless standards. MIMO techniques are used today in technologies like Wi-Fi and LTE, and new techniques are under study for future standards like LTE Advanced.

Coping With Signal Attenuation

A signal propagated between a transmitter and a receiver is often affected by fading. One major source of fading is multipath propagation, where different copies of the signal partially cancel each other out at certain times and points in space. This attenuation decreases the signal to noise ratio (SNR), which in turn increases the number of errors. This increase in errors decreases the effective throughput of the signal, thereby weakening the RF link.

Diversity is one technique that is used to mitigate this effect. By transmitting multiple versions of a signal with different characteristics, each one being affected differently by the propagation path, we therefore we increase the chances that the signal will reach the receiver. Time and frequency diversity are two types of diversity that are used. The use of multiple antennas adds a new type, space diversity.

Space Diversity

Starting with a standard SISO (single input single output) implementation, we can create transmit diversity by using multiple antennas at the transmitter. This is configuration is called MISO (multiple input single output). Similarly, we can add receive diversity by using multiple antennas at the receiver. This configuration is called SIMO (single input multiple output). Real MIMO starts when multiple antennas are used both at the transmitter and the receiver. The more antennas, the better, but the cost, space and processing power required to handle the multiple streams of data limit the number of antennas that can be used in practice.

Diversity Coding and Spatial Multiplexing

The two main forms of MIMO are diversity coding and spatial multiplexing. Diversity coding is used to increase link reliability in the presence of fading conditions. With this technique, multiple versions of the same data are encoded and transmitted over multiple antennas. The encoding adds some level of time diversity. The multiple signals being propagated using different paths are affected differently by the fading. The receiver then recovers the original stream either by selecting the best received signal or by combining all the information received. The Alamouti code is one of the most popular space-time block codes used for 2×2 diversity coding.

Spatial multiplexing exploits the same concept of multiple propagation channels, but to increase the spectral efficiency instead of the link reliability. For spatial multiplexing, instead of transmitting multiple versions of the same signal, the data to be sent is split into multiple streams and transmitted using multiple antennas. Here, the coding is very important because the receiver needs to be able to separate the data sent on each channel to recover the original signal. Obviously, this technique is only used when the propagation conditions between the transmitter and the receiver are good.

The consequence of this multiplexing is an increase in throughput. By using a 2×2 (two transmit and two receive) configuration, it is in theory possible to double the throughput using exactly the same RF channel bandwidth. Using 4×4, the throughput would quadruple. For example, in LTE the peak throughput using SISO is about 100 Mbps. Using 2×2 MIMO and 4×4 MIMO this number reach 172.8 Mbps and 326.4 Mbps, respectively, while using the same channel bandwidth. LTE Advanced has even added a new transmission mode that uses 8×8 MIMO.

The concept of multiplexing can be pushed even further by allowing different users to transmit at the same time over the same channel. This is called multi-user MIMO or MU-MIMO. This technique is very interesting because it increases the capacity of a cell without using more bandwidth.

OFDM with MIMO

Even though the MIMO technique can be used with any modulation, OFDM and its derivatives are particularly well suited to it because for each sub-carrier they transform the channel selective fading into near-flat fading, which is easy to model and equalize. This greatly simplifies the implementation of MIMO that can work individually on these sub-carriers. When added to an OFDM system, which by definition is already spectrally efficient and robust, MIMO adds even more robustness and data throughput capacity, which explains why it’s used in almost all current wireless standards and studied for future ones.