Wireless communications' spectral efficiency (SE) and throughput can be significantly increased with 5G technology. Multiple input multiple output (MIMO) technology, which makes use of large antennas and user equipment(UE), has emerged as most significant technology in this respect. However, the 5G wireless technology's use of MIMO would result in higher circuit power consumption as well as lower energy efficiency (EE). This research proposed novel technique in enhancing the wireless communication network spectral efficiency with secure data transmission. here the spectral efficiency of the network has been enhanced using massive multiple user based multiple input- multiple output (MMU-MIMO). Data transmission of network has been carried out based on orthogonal frequency division multiplexing (OFDM). Base station essentially serves several users simultaneously utilising the same frequency in MMU-MIMO. The spectral performance of downlink MMU-MIMO is the subject under discussion in this work. Secure data transmission of the network has been carried out using Federated deep Q- neural networks. Our simulation findings show how, for various QoS constraint levels, the ideal number of scheduled users varies