An Efficient Channel Estimation Scheme based on Frozen Bits of Polar Codes over Fading Channel

Abstract

Polar code has emerged as a candidate of channel coding scheme for 5G as it can achieve the Shannon limit with low encoding and decoding complexity. Channel polarization is the main principle of polar code based on channel state information estimation, and then the encoded bits can separate as information bits and frozen bits, where the frozen bits in polar code are constant bits which are known at both the transmitting and receiving side. Hence, an efficient signal-to-noise-ratio (SNR) estimation scheme based on frozen bits over fading channel is proposed in this paper. In this channel estimation scheme, we first analyze the theoretically SNR with infinite code length according to the mapping between SNR and frozen bit error rate. Then we propose an approximate estimation method based on frozen bits (FBES) for finite code length polar code over fading channel. In addition, we compare the estimation performance of the proposed scheme with the existed estimation scheme. Simulation results validate our analytical results and show that the proposed estimation scheme has better estimation performance with a relatively lower complexity.