LAUSR

Sparse code multiple access (SCMA) and polar codes are key techniques for 5G systems. Towards to the realization of massive machine type communications (mMTC) in Internet of Things (IoT), we propose a joint channel estimation and decoding (JCD) scheme to enhance the performance of polar coded SCMA system. There are three parts in our JCD scheme. First, a sparse Bayesian learning (SBL) algorithm is proposed to measure the initial channel state information (CSI) for the JCD scheme. Then, the JCD receiver performs an iterative channel estimation and detection and decoding (ICED) scheme between the SCMA detector and the polar decoder, where the log-likelihood ratio (LLR) from SCMA detector and polar decoder can iteratively update, and the channel gain is also adjusted by the frozen bit error rate (FBER) from the polar decoder in each ICED iteration. In addition, we propose a soft successive cancellation list (soft-SCL) decoding algorithm by calculating the path metric from an LLR based logarithm method, then the soft-SCL decoder outputs the LLR information and the SCMA detector can update the symbol probability in this ICED scheme. Simulation results demonstrate the superior performance of our JCD scheme over fading channels.