MIMO is the acronym that refers to multiple input multiple output. MIMO exploits a previously mentioned RF phenomenon called multipath. Multipath transmission implies that signals will reflect off walls, doors, windows, and other obstructions. A receiver will see many signals each arriving at different times via different paths. Multipath tends to distort signals and cause interference, which eventually degrades signal quality (this effect is called multipath fading). With the addition of multiple antennas, a MIMO system can linearly increase the capacity of a given channel by simply adding more antennas. There are two forms of MIMO:
- Spatial diversity: This refers to transmit-and-receive diversity. A single stream of data is transmitted on multiple antennas simultaneously using space-time coding. These provide improvements in the signal to noise ratio and they are characterized by their improvement of link reliability and coverage of the system.
- Spatial multiplexing: Used to provide additional data capacity by utilizing the multiple paths to carry additional traffic; that is, increasing the data throughput capability. Essentially, a single high-rate data stream will be split into multiple separate transmissions on different antennas.
Using multiple streams does impact overall power usage. IEEE 802.11n includes a mode to only enable MIMO when the effect will have a performance benefit, thus saving power at all other times. The Wi-Fi Alliance requires all products to support at least two spatial streams to receive 802.11n compliance.
A WLAN will split data into multiple streams called spatial streams. Each transmitted spatial stream will use a different antenna on the transmitter. IEEE 802.11n allows for four antennas and four spatial streams. By using multiple streams sent separately for antennas spaced apart from each other, spatial diversity in 802.11n gives some confidences that at least one signal will be strong enough to reach the receiver. At least two antennas are needed to support MIMO functionality. Streaming is also modulation agnostic. BPSK, QAM, and other forms of modulation work with spatial streaming. A digital signal processor on the transmitter and receiver will adjust for the multipath effects and delay the line of sight transmission by just enough time to have it line up perfectly with the non-line-of-sight paths. This will cause the signals to reinforce.
The IEEE 802.11n protocol supports a four-stream single user MIMO (SU-MIMO) implementation, meaning that the transmitters would work in unison to communicate to a single receiver. Here 4 transmit and 4 receive antennas deliver multiple streams of data to a single client. Right: the effect of spatial diversity MIMO in 802.11n. Following is an illustration of SU-MIMO and multipath use in 802.11n.
In the figure, four transmit and four receive antennas deliver multiple streams of data to a single client (SU-MIMO). On the right, two transmitters spaced a fixed distance apart communicate to two receivers. Multiple paths exist as reflections from the two transmitters. One line-of-sight path is stronger and will be favored. DSPs on the transmitters and receiver side also mitigate multipath fading by combining signals so the resulting signal exhibits little fading.
802.11n also introduced the optional feature of beamforming. 80211.n defines two types of beamforming methods: implicit feedback and explicit feedback:
- Implicit feedback beamforming: This mode assumes that the channel between the beamformer (AP) and the beamformee (client) are the reciprocal (same quality in both directions). If that is true, the beamformer transmits a training request frame and receives a sounding packet. With the sounding packet, the beamformer can estimate the receiver's channel and builds a steering matrix.
- Explicit feedback beamforming: In this mode, the beamformee responds to a training request by computing its own steering matrix, and sends back the matrix to the beamformer. This is a more reliable method.
Below is a diagram illustrating the effects of beamforming in a situation with no line-of-sight communication. In the worst case, the signals arrive 180 degrees out of phase and nullify each other. With beamforming, the signals can be adjusted in phase to strengthen each other at the receiver.
Beamforming relies on multiple spaced antennas to focus a signal at a particular location. The signals can be adjusted in phase and magnitude to arrive at the same location and reinforce each other providing better signal strength and range. Unfortunately, 802.11n did not standardize on a single method for beamforming and left that up to the implementor. Different manufacturers used different processes and could only guarantee it would work with identical hardware. Thus, beamforming was not adopted heavily in the 802.11n timeframe.
We will cover MIMO technologies in many other areas, such as 802.11ac, and in the next chapter on long-range communication using cellular 4G LTE radios.