BER bounding techniques for selected turbo coded MIMO systems in Rayleigh fading

Abstract

II. SYSTEM MODEL Fig. 1 shows the block diagram of the considered turbocoded MIMO transmitter for uplink transmission in a typical urban environment, where n r transmit antennas are employed at the mobile and n R receive antennas at the base-station. Channel state information is assumed to be available at the receiver only and n r is assumed to be less than n R • The received symbol vector r, in a discrete-time MIMO channel model, is expressed as Abstract -We consider turbo-coded MIMO systems with Rayleigh fading, when channel state information is available at the receiver only. Using our recently proposed linearly complex MIMO signal detector [9], we derive two bit-error-rate bounds on performance for such systems by extending the union bound method from turbo codes on the AWGN channel [13] to timevarying MIMO fading channels. These bounds allow a designer to quantify the impact of MIMO system parameters, e.g., number of transmit / receive antennas, turbo code selection or an SNR change. Simulation results demonstrate that the derived bounds are within 0.2 -0.5 dB of the system performance in the error floor region.

Performance Analysis of Coded Massive MIMO-OFDM Systems Using Effective Matrix Inversion

“…In [26], an average bound has been proposed for the performance of turbo coded systems with correlated and uncorrelated Rayleigh fading channels. In [10], [27], the authors have proposed an upper-bound for the performance of the turbo coded MIMO system with correlated and uncorrelated Rayleigh, slow-fading channels, and the proposed bound approached the simulation results within 0.2-0.5 dB at a BER of 10 −5 .…”

“…In Section IV, we used the PDF of (16) to derive the BER equation of the massive MIMO-OFDM system with different modulation types. To verify the BER for 4-QAM and 16-QAM using(27) and(29), respectively, the theoretical performance of the massive MIMO-OFDM system is compared with Monte-Carlo simulations using N = 4, 10 and N r = 100, 200 and the obtained performances are illustrated in Figs. 7 and 8 both of which show close match between theory and simulations.…”