Low-complexity Near-Optimal Detector for Multiple-Input Multiple-Output OFDM with Index Modulation.

Beixiong Zheng,Miaowen Wen,Ertugrul Basar,Fangjiong Chen

DOI: https://doi.org/10.1109/icc.2017.7996536

2017-01-01

Abstract:Multiple-input multiple-output orthogonal frequency division multiplexing with index modulation (MIMO-OFDM-IM), which provides a flexible trade-off between spectral efficiency and error performance, is recently proposed as a promising transmission technique for energy-efficient 5G wireless communication systems. However, due to the dependence of subcarrier symbols within each subblock and the strong interchannel interference, it is challenging to detect the transmitted data effectively while imposing low computational burden to the receiver. In this paper, we propose a low-complexity detector based on the sequential Monte Carlo (SMC) theory for the detection of MIMO-OFDM-IM signals. The proposed detector, which draws samples based on the importance weights at the subblock level, achieves near-optimal error performance with considerably reduced computational complexity. Simulation and numerical results in terms of bit error rate (BER) and number of complex multiplications (NCM) corroborate the superiority of the proposed detector.