Document Type

Theses, Ph.D


Available under a Creative Commons Attribution Non-Commercial Share Alike 4.0 International Licence

Publication Details

Successfully submitted for the award of Doctor of Philosophy (Ph.D) to the Technological University Dublin, June, 2015.


MIMO antenna array systems have been proposed as a means of increasing the spectral efficiency of wireless systems. However, their performance is likely to be sub-optimal if typical uniform antenna array structures are arbitrarily positioned; as they depend on spatial multiplexing. This is particularly true for indoor environments in which transmission ranges are short resulting in a strong correlation of the main propagation paths, especially the line-of-sight components. This makes it difficult to achieve successful spatial multiplexing which depends on a decorrelated set of signal components. Thus, the physical propagation channel and geometry of the antenna arrays, especially the inter-element spacing, can determine how effectively spatial multiplexing can be realised. This thesis investigates MIMO communications channels involving a single transmitter and receiver operating in a simple indoor environment using a ray-tracing simulation model. The results and analysis provide system designers with an understanding of the limits of MIMO system performance in the context of both the geometric properties of the arrays and the propagation conditions. These results serve to explain the often contradictory results that appear in the wider literature on MIMO systems. Guidelines for the deployment of standard array structures in an indoor environment are provided. An original solution to optimising MIMO system performance by adjusting the geometry of uniform linear arrays is described. This is done using an iterative search method based on the Metropolis algorithm in which individual array elements are repositioned. It is demonstrated through computer simulation that capacity levels, similar to those predicted by the theory for ideal Rayleigh channels, are possible to achieve with realistic modifications to uniform linear arrays.