Neural network-based multiuser detection for SDMA–OFDM system over IEEE 802.11n indoor wireless local area network channel models

Space division multiple access – orthogonal frequency division multiplexing-based wireless communication has the potential to offer high-spectral efficiency, system performance and capacity. This article proposes an efficient blind multiuser detection (MUD) scheme using artificial neural network models such as the radial basis function. The proposed MUD technique is consistently outperforming the existing minimum mean square error and minimum bit error rate (MBER) MUDs with the performance close to the optimal maximum likelihood (ML) detector. Besides that, the computational complexity of the proposed one is comparatively lower than both the MBER and ML detectors. Further, it can also outperform MBER MUD in the overload scenario, where the number of users is more than that of the number of receiving antennas simulation-based study showing BER performance and complexity are carried out to prove the efficiency of the proposed techniques. This analysis is carried through the IEEE 802.11n standard channel models, which are designed for indoor wireless local area network applications of bandwidth up to 100?MHz at frequencies 2 and 5?GHz.