Efficient Operation Scheduling in Successive-Cancellation-based polar decoders

摘要

Polar codes are a class of error correcting codes that can achieve channel capacity, and that have been selected for the next generation of wireless communication standards. Successive-cancellation (SC) is the first proposed decoding algorithm for polar codes, suffering from mediocre error-correction performance at medium code length and relatively long decoding latency. Various evolutions of SC are present in literature, attempting to overcome said limitations. Decoder architectures implementing SC-based algorithms are present in literature, spanning a wide variety of decoding algorithms and implementing different architectural solutions to decrease the area occupation, power and energy consumption, and increase speed and throughput. In this work, we propose two techniques that aim at reducing the decoding latency of SC-based decoders: they rely on the optimization of the scheduling of operations in SC. They are complementary to each other, each one being applicable to the part of the algorithm where the other can not. Depending on the decoder architecture, latency analysis shows improvements ranging between 15.79% and 33.34%, when both techniques are combined.