Dear,

For free-space optical (FSO) transmission, the effects of MIMO is shown that spatial diversity can combat the fading effects due to scattering and scintillation caused by atmospheric turbulence. Ongoing research activities intend to increase the capacity of indoor optical wireless communication (OWC) systems by MIMO techniques. However, for indoor OWC it is still not clear to what extent MIMO techniques can provide gains. This is because in indoor environments there are no fading effects caused by turbulence, especially if line-of-sight (LOS) scenarios are considered. Therefore, indoor optical wireless links are envisaged to be highly correlated, enabling only minor diversity gains. As MIMO techniques mostly rely on spatially uncorrelated channels, it is unclear whether the optical propagation channel in indoor environments can offer sufficiently low channel correlation.

A major performance degrading factor in free-space optical communication (FSO) systems is atmospheric turbulence. Spatial diversity techniques provide a promising approach to mitigate turbulence-induced fading.

To mitigate the deleterious effects of scattering and turbulence, multiple transmitters and receivers can be used. Hence, it would be possible to benefit from spatial diversity and receive multiple independent copies of the same signal. 

This paper should answer some of your questions:

http://cictr.ee.psu.edu/facstaff/kavehrad/JCM-490.pdf

Note that the paper talks about use of laser sources, and outdoor propagation. So it would not apply to an indoor LiFi system, which typically uses LED (incoherent) light sources.

But I'd say that one can certainly imagine an indoor LiFi system that makes use of MIMO techniques, much like all the newer versions of WiFi do. For example, super simplistic here, aim narrow beams in different directions around the room, then have the user equipment provided with multiple narrow beam sensors, also aimed in various directions. You'd have to have a way of aiming the user equipment sensors, perhaps with light training signals used for this purpose.

With incoherent light, you won't be able to exploit the same mathematical magic that RF or laser MIMO can use, however the main concept, that of making use of highly uncorrelated propagation paths, might still be exploited?

Dear,

These links may be useful for you :

https://www.osapublishing.org/jlt/abstract.cfm?uri=jlt-34-9-2158

http://www.engr.uconn.edu/~jcui/UWSN_papers/wireless-overview_wcmc_2008_v1.pdf

https://arxiv.org/pdf/1502.00365.pdf