Graph traversal aided detection in uplink MBM massive MIMO based on socio-cognitive knowledge of swarm optimization

Media-based modulation (MBM) plays a crucial role in enhancing the spectral efficiency and energy efficiency of massive MIMO (mMIMO) systems for 5G and beyond wireless communications. In MBM, multiple parasitic elements, also termed as radio frequency (RF) mirrors, are placed near the transmit antennas for generating different channel fade realizations. These realizations are obtained by ON/OFF switching of RF mirrors. One of those channel fade realizations is selected (using a part of the incoming information bits) for transmitting a part of the information bits utilizing a symbol chosen from the conventional constellation set (using another part of the incoming information bits). Transmission of a symbol through one of the available channel realizations constitutes a sparse transmit vector for each user in MBM-mMIMO. The sparse nature of transmitted symbols from multiple users and inter-user interference makes the symbol detection in uplink MBM-mMIMO challenging. Therefore, in this article, the problem of symbol detection in MBM-mMIMO is analyzed from a graph-theoretical point of view, and a graph-traversal aided low-complexity symbol detection algorithm is proposed inspired by socio-cognitive learning of swarm optimization. Also, the convergence characteristic of the proposed technique is investigated theoretically. Further, an analytical expression of upper bound on bit error rate performance is derived and corroborated through simulations. Viability and robustness of the proposed technique are also justified through simulations over state-of-art detection techniques, under both perfect and imperfect channel state information scenarios.