Introduction to MIMO

MIMO (multiple input multiple output) is emerging as one of the most significant technical breakthroughs in modern communications. It is gaining prominence with inclusion in many international standards such as WiMAX, LTE, DVB-T2, 802.11n and 802.11ac.

MIMO is a generic term that refers to any radio system that uses multiple transmit and receive antennas. MIMO inputs and outputs are referring to a communications channel where there are multiple inputs to the channel from multiple transmit antennas and multiple outputs from multiple receive antennas. The core idea is that through time-space signal processing the multiple signals from the transmit antennas on one side of the communications channel combine with the receive antennas on the other side to provide improved performance for the MIMO user as either:

• improved data rates (bits/sec) via spatial multiplexing techniques and/or,
• improved data quality (lower BER for a given SNR) via spatial diversity techniques and/or,
• improved signal-to-noise ratios and reduced interference by way of beam forming.

MIMO has the prospect to increase wireless communication performance by many orders of magnitude with no spectrum penalty or increase in total transmitter power, only increased hardware and digital signal processing complexity.

Communication systems using MIMO can be one way or two way, symmetrical or asymmetrical, either frequency division duplex (FDD) or time division duplex (TDD). TDD simplifies spectrum planning as the up and down links are on the same RF channel and it also has the attraction that multiple users can dynamically share the increased data capacity of a MIMO system ie. either one user can benefit from the increased data capacity or multiple users can each share the increased capacity in time. For example, if a base station has a total capacity of 50Mbps, then either 1 user can use 50Mbps or 5 users can each have 10Mbps, or any combination up to 50Mbps.

1. Spatial multiplexing increases the data rate compared with a single antenna system. The capacity of a spatial multiplex MIMO system increases linearly with the number of antennas at the rate of the minimum of the number of transmit (N) or receive (M) antennas. Typically for good error performance the number of receive antennas exceeds the number of transmit antennas.

Typically only the receiver needs to have knowledge of the channel.

At the transmitter the input data sequence is split into N sub-sequences that are transmitted simultaneously using the same frequency. Therefore the data rate increases by N.

At the receiver the sub-sequences are separated by means of an interference cancelling algorithm, eg linear zero-forcing (ZF) or minimum-mean-squared-error (MMSE) detector, maximum-likelihood (ML) detector or successive interference cancellation (SIC) detector.

2. Spatial diversity decreases the error rates compared with a single-antenna system. The idea is to send and/or receive multiple redundant versions of the same data sequence and to combine at baseband.

For receive diversity there are a number of combining techniques, equal gain combining (EGC), maximum-ratio combining (MRC) and selection combining (SC).

Transmit diversity requires appropriate pre-processing to ensure coherent combining at the receiver. Well known techniques are Alamouti’s scheme for 2 transmit antennas, space-time trellis codes and orthogonal space-time block codes.

Again typically only the receiver needs to have knowledge of the channel.

3. Beam forming is used to improve signal to noise ratios (SNR) and signal plus interference to noise ratios (SINR) in multiuser scenarios. The improvement can be used to decrease errors or increase data rates. Beam forming applies to both a transmit array and a receive array, ie. reciprocity.

The SNR is improved by focusing the antenna pattern on the desired angles of reception and transmission for optimum signal strength.

The SINR is improved by steering antenna nulls towards co-channel users. This can enable space division multiple access (SDMA) as an alternative to time or frequency division multiple access (TDMA or FDMA).

In practical system the direction of significant scatterers must be estimated. If the channel is known at the receiver and the transmitter then the beamforming can be performed in the baseband domain, eigen-beamforming.