Joint Turbo Decoding for Overloaded MIMO-OFDM Systems

To achieve an excellent error-correcting capability, the Third-Generation Partnership Project (3GPP) Long-Term Evolution (LTE) employs turbo codes as its forward error correction (FEC) standard. In addition, to achieve higher throughput, the LTE also implements multiple-input-multiple-output (MIMO) systems. Although a conventional turbo decoding scheme gives satisfactory performance in ideal MIMO systems, significant performance degradation occurs in an overloaded MIMO system when the number of transmit antennas is larger than that of receive antennas. In this paper, a joint decoding scheme for turbo codes, i.e., so-called joint turbo decoding, is proposed. In joint turbo decoding, calculations of soft information are conducted for each combination of bits from all streams instead of separately between each stream. In addition, to combine the trellis diagrams from all the streams, a super-trellis diagram is employed. The numerical results obtained through computer simulation show that the proposed scheme performs better than the conventional scheme, particularly in the case of an overloaded MIMO system. In the MIMO system with four transmit and two receive antennas, up to 2-dB performance enhancement can be attained at a bit error rate (BER) of 10^-5. The proposed scheme also obtains improved throughput with a throughput/(SNR/stream) ratio close to the theoretical capacity of overloaded MIMO systems.