PRECODER DESIGN BASED ON INTER-ANTENNA LEAKAGE MINIMIZATION FOR MU-MIMO

Multiple Input Multiple Output (MIMO) is an important technological development
that makes use of the spatial dimension hitherto underutilized to increase the channel
capacity. MIMO today is an integral part of many up to date wireless standards, such
as: IEEE 802.lln for wireless LAN, WiMAX 2, and even the latest ITU-T mobile
communication/ L TE-Advanced standard adopted by 3GPP in 20 II. The performance
of the Multi-user MIMO (MU- MIMO) broadcast channel highly depends on the
precoding scheme used for each user at the transmission front end. The MU-MIMO
precoding design process is broadly classified in either of the following two
frameworks, namely linear and nonlinear. Given the practical implementation
challenges of the nonlinear precoding methods, preference is given to linear methods
resulting in simple but high performance and efficient transmission methods.
In this thesis, a new MU-MIMO precoding method is developed based on
optimizing what is termed as Per-Antenna Signal to Leakage plus Noise Ratio (PA�SLNR). The optimization is facilitated using Fukunaga-Koontz Transform (FKT) that
not only overcomes the problem of singularity but also helps reduce computation.
Furthermore, we investigate the proposed method on a typical and practical wideband
channel where OFDM is used to convert the wideband frequency selective channel
into a set of narrow band flat sub-channels. The performance of the proposed method
is validated against other methods available in the literature. finally, we extend the use
of our proposed method to the extremely important and practical deployment scenario
of cellular communication where MU-MIMO precoding is devised for the inter-cell
interference cancelation by utilizing, what is known as, Multi-Cell Processing (MCP).
We demonstrate the importance of precoding in cellular scenario by the capacity
enhancement of both the cell edge users and the overall cellular system.