Learning from the Syndrome
This work addresses the challenge of enhancing decoder performance in communication systems, particularly for online adaptation to changing channels, though it is incremental as it builds on existing neural decoder methods.
The paper tackles the problem of improving neural error-correcting decoders by introducing the syndrome loss, which penalizes invalid codewords, resulting in consistently lower frame error rates for short block codes with minimal training cost and no inference overhead.
In this paper, we introduce the syndrome loss, an alternative loss function for neural error-correcting decoders based on a relaxation of the syndrome. The syndrome loss penalizes the decoder for producing outputs that do not correspond to valid codewords. We show that training with the syndrome loss yields decoders with consistently lower frame error rate for a number of short block codes, at little additional cost during training and no additional cost during inference. The proposed method does not depend on knowledge of the transmitted codeword, making it a promising tool for online adaptation to changing channel conditions.