Syndrome-Enabled Unsupervised Learning for Neural Network-Based Polar Decoder and Jointly Optimized Blind Equalizer
This work addresses incremental improvements in communication systems by enabling unsupervised learning for polar codes and equalizers, reducing the need for training sequences.
The authors tackled the problem of adapting syndrome loss for unsupervised learning in polar decoders and blind equalizers, resulting in a BP polar decoder that outperforms supervised methods in block error rate and a blind equalizer achieving a 1.3 dB gain over non-blind MMSE equalizer.
Recently, the syndrome loss has been proposed to achieve "unsupervised learning" for neural network-based BCH/LDPC decoders. However, the design approach cannot be applied to polar codes directly and has not been evaluated under varying channels. In this work, we propose two modified syndrome losses to facilitate unsupervised learning in the receiver. Then, we first apply it to a neural network-based belief propagation (BP) polar decoder. With the aid of CRC-enabled syndrome loss, the BP decoder can even outperform conventional supervised learning methods in terms of block error rate. Secondly, we propose a jointly optimized syndrome-enabled blind equalizer, which can avoid the transmission of training sequences and achieve global optimum with 1.3 dB gain over non-blind minimum mean square error (MMSE) equalizer.