AMC With a BP-ANN Scheme for 5G Enhanced Mobile Broadband

In 5G enhanced mobile broadband (eMBB), applications that require high spectral efficiency and transmission speeds employ adaptive modulation and coding (AMC) technology. Such technology enables various levels of modulation and coding. AMC technologies use channel state information to select the optimal modulation or coding scheme as well as transmission parameters to improve the transmission speed, transmission quality, and spectrum efficiency of links. Because of equipment limitations at the receiving end, channel estimation algorithms cannot be used to acquire ideal solutions, and thus estimation errors are inevitable. Consequently, parameter information returned from the receiving end may be inaccurate and the radio link is affected by interference, which may impede the reliability of data transmission and reduce spectrum efficiency. This study proposed an AMC with a backpropagation artificial neural network (BP-ANN) scheme (AMC-BP-ANN). This study addressed problems regarding reduced efficiency in conventional AMC technology caused by channel, interference and noise estimation errors. In particular, this study used channel estimation, interference and noise estimation errors as features for machine learning, which allowed eMBB systems to accurately estimate the signal-to-interference plus noise ration and ensure that the estimated value approximated the value was determined through ideal channel estimation. The simulation results indicated that the AMC-BP-ANN scheme could achieve ideal performance and a superior bit error rate compared with the AMC technologies using lookup table and genetic algorithm, and furthermore, a lower bit error rate could improve the transmission reliability and spectral efficiency in 5G eMBB.