Quantization Analysis for the CRC Aided Successive Cancellation List Polar Decoder
Nikola Borović, Dragomir El Mezeni, Vladimir L. Petrović · 2025
Polar codes have recently found application as error correction codes for the control information in 5 G wireless communication. Stringent requirements for decoding latency, power consumption, and communication reliability demand efficient hardware implementation of communication modules. One of the key aspects of efficient hardware implementation is a robust quantization scheme. In this paper, we present an empirical approach to analyzing quantization for the Cyclic Redundancy Check-Aided Successive Cancellation List (CA-SCL) decoding algorithm, which is used in the 5 G standard. The paper gives detailed quantization analysis of the CA-SCL algorithm for 11-bit CRC and 4 lists. We proposed a quantization scheme with decoding performance within 0.1 dB of the floating point performance for positive Signal-to-noise ratios (SNRs). It is shown that a decoder with one extra bit achieves performance within 0.1 dB of the floating-point model over a wide range of SNRs and code rates.