Enabling Signal Space Diversity for MU-MIMO/OFDMA Cellular Systems That Employ Frequency Diversity

Abstract

Signal space diversity (SSD) is a technique to obtain diversity without loss in spectral efficiency. Despite this, SSD has been minimally studied in multiuser scenarios. In this paper, SSD is operated in the uplink of orthogonal frequency division multiple access (OFDMA) cellular systems that exploit frequency diversity, use the optimum maximum-likelihood detector (MLD) and employ antenna arrays at the receivers. Thus, a multiuser multiple-input-multiple-output (MU-MIMO) system that works in presence of own-cell and co-cell interference (CCI) is taken into account. In order to diminish CCI, fractional frequency reuse (FFR) is used in the cellular system. For a feasible MLD implementation, a matrix structure of the received signals is obtained in order to use the sphere decoder algorithm. An exact closed-form expression to calculate the pairwise error probability between two multidimensional symbols is derived. From this, a bit error rate (BER) upper bound is found for single user scenarios. This BER expression is an accurate lower bound for high diversity orders in multiuser scenarios. The BER asymptotic analysis shows that our approach maintains the SSD multidimensional diversity, the spatial diversity of MIMO and the frequency diversity. This allows to overload the system by minimizing the BER and increasing the spectral efficiency, simultaneously.

Keywords

• Diversity methods
• MIMO
• mobile communication
• multiple access interference
• OFDM