Improved Logical Error Rate via List Decoding of Quantum Polar Codes

Abstract

The successive cancellation list decoder (SCL) is an efficient decoder for classical polar codes with low decoding error, approximating the maximum likelihood decoder (MLD) for small list sizes. Here we adapt the SCL to the task of decoding quantum polar codes and show that it inherits the high performance and low complexity of the classical case, and can approximate the quantum MLD for certain channels. We apply SCL decoding to a novel version of quantum polar codes based on the polarization weight (PW) method, which entirely avoids the need for small amounts of entanglement assistance apparent in previous quantum polar code constructions. When used to find the precise error pattern, the quantum SCL decoder (SCL-E) shows competitive performance with surface codes of similar size and low-density parity check codes of similar size and rate. The SCL decoder may instead be used to approximate the probability of each equivalence class of errors, and then choose the most likely class. We benchmark this class-oriented decoder (SCL-C) against the SCL-E decoder and find a noticeable improvement in the logical error rate. This improvement stems from the fact that the contributions from just the low-weight errors give a reasonable approximation to the error class probabilities. Both SCL-E and SCL-C maintain the complexity O(LN logN) of SCL for code size N and list size L. We also show that the list decoder can be used to gain insight into the weight distribution of the codes and how this impacts the effect of degenerate errors.

arXiv comment14 pages, 7 figures

References in corpus (12)

• Channel polarization: A method for constructing capacity-achieving codes for symmetric binary-input memoryless channels
• Good Quantum Error-Correcting Codes Exist
• Multiple Particle Interference and Quantum Error Correction
• LLR-based Successive Cancellation List Decoding of Polar Codes
• Active stabilisation, quantum computation and quantum state synthesis
• Efficient Algorithms for Maximum Likelihood Decoding in the Surface Code
• Decoding Across the Quantum LDPC Code Landscape
• Algebraic Properties of Polar Codes From a New Polynomial Formalism
• Tensor Networks and Quantum Error Correction
• Efficient Quantum Polar Coding
• Alignment of Polarized Sets
• Fault-Tolerant Preparation of Quantum Polar Codes Encoding One Logical Qubit