Abstract
Interfering manipulation is a growing number of bright aspects for 6th generation (6G) infrastructure, particularly to detect huge sequence quadrature amplitude modulation (QAM) signals. In this manuscript, an evolutionary gravitational Neocognitron neural network based differential spatial modulation detection scheme (EGNNN-DSMDS) is proposed for uplink multiple user huge multiple input multiple output (MIMO) systems. To prevent the wastage of antenna resources, massive MIMO uplink systems use the differential spatial modulation (DSM), here the transmitter is equipped with a small count of antennas, whereas the receiver is equipped with a huge count of antennas. By utilizing the asymptotic characteristics of massive multiple input multiple output channels, the transmitting signal can be uniquely decoded without instantaneous channel state information in the receiver. When analyzed with conventional coherent zero-forcing detector (ZFD), the proposed approach has less complication. The proposed EGNNN-DSMDS detection scheme achieves excellent bit error ratio performance and estimating channel errors. EGNNN-DSM attains higher spectral efficiency 99.56%, 96.33%, 98.11%, lower BER 2.5%, 4.13%, 1.19% and lower computational complexity 1.19%, 2.16%, 1.15% compared with existing methods, like MLPNN-MUD-UMM-MIMO, CNN-LASD-UMM-MIMO, and DNN-SRD-UMM-MIMO. Finally, the proposed EGNNN-DSM detection scheme attains the better outcomes.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call