Nonlinear Amplification Effects on Dual Band Multi-User MIMO-OFDM Systems

In this paper, we evaluate the performance of carrier aggregated dual band (DB) multi- user (MU) multi-input-multi- output (MIMO) orthogonal frequency division multiplexing (OFDM) system in the presence of transmitter nonlinearity and frequency-selective fading channels. As a benchmark, the performance has been compared with a linear DB MU-MIMO-OFDM system which employs a conventional transmit preprocessing technique to mitigate the inter-user interference. A two-dimensional (2D) memory polynomial (MP) is considered to model the transmitter nonlinearity. We show that transmitter nonlinearity leads to nonlinear interference in a DB MU-MIMO- OFDM system which can not be eliminated by the conventional transmit preprocessing techniques. As a consequence, there is an irreducible error floor which degrades the performance of the nonlinear DB MU-MIMO-OFDM systems. We have also shown that the irreducible error floor is the function of number of aggregated bands as well as number of users present in the system. The impact of number of bands and users has been shown by comparing the symbol error rates of a nonlinear single band (SB) MU- MIMO-OFDM system vis-a-vis nonlinear DB MU-MIMO-OFDM system for a multi-path frequency selective Rayleigh fading channel.