Research on Polar Code Construction Algorithms Under Gaussian Channel

Abstract

Polar code is widely accepted due to its excellent error correction performance, low complexity coding and decoding structure. However, the imperfect performance in the short code length of polar codes limits its further application. A significant way to improve the performance of short polar code is to choose fit subchannels for information bits accurately and effectively. After analyzing some construction algorithms for polar codes under Gaussian channel, comparing their complexity and their BER performance, this paper proposes an improved polarization weight (PW) algorithm, which can reduce the bit error rate.