DEEP LEARNING-BASED SYMBOL DETECTION TECHNOLOGY FOR MULTI-USER OF MIMO-OFDM-IM SYSTEM

MIMO-OFDM-IM is a novel multi-carrier modulation technique combining MIMO-OFDM (multiple-input multiple-output orthogonal frequency division multiplexing) and IM (index modulation). In order to solve the problem that the symbol detection technology of multi-user MIMO-OFDM index modulation still has high complexity and the bit error performance is seriously affected by the number of users, a symbol detection framework based on deep learning is proposed. Both encoders and decoders were constructed by DNN (deep neural network), using supervised learning training data. The experimental results show that the above problems are effectively solved.