Abstract
Polar codes have become the channel coding scheme for control channel of enhanced mobile broadband in the fifth generation (5G) communication systems. Belief propagation (BP) decoding of polar codes has the advantage of low decoding latency and high parallelism but suffers from high complexity. In this paper, a low complexity BP decoder is proposed for polar codes. We reduce the computational complexity by two steps. First, the cyclic redundancy check is concatenated to the decoder in order to decrease the number iterations of the BP algorithm. Then, a threshold is proposed based on Gaussian approximation to save the computational complexity of BP nodes. If the log-likelihood ratio of a node in the tanner graph is larger than the threshold, this node is no longer updated during the rest of the decoding process. The simulation results show that the proposed scheme has a similar block error rate performance with the original BP decoder, while the computational complexity is reduced significantly.
Original language | English |
---|---|
Article number | 8688403 |
Pages (from-to) | 50710-50717 |
Number of pages | 8 |
Journal | IEEE Access |
Volume | 7 |
DOIs | |
Publication status | Published - 2019 |
Keywords
- Polar codes
- belief propagation
- eMBB
- low complexity decoding