Polar5GDecoder#

class sionna.phy.fec.polar.Polar5GDecoder(enc_polar: sionna.phy.fec.polar.encoding.Polar5GEncoder, dec_type: str = 'SC', list_size: int = 8, num_iter: int = 20, return_crc_status: bool = False, *, precision: str | None = None, device: str | None = None, **kwargs)[source]#

Bases: sionna.phy.block.Block

Wrapper for 5G compliant decoding including rate-recovery and CRC removal.

Parameters:

  • enc_polar (sionna.phy.fec.polar.encoding.Polar5GEncoder) – Instance of the Polar5GEncoder used for encoding including rate-matching.

  • dec_type (str) – Defining the decoder to be used. Must be one of {“SC”, “SCL”, “hybSCL”, “BP”}.

  • list_size (int) – Defining the list size iff list-decoding is used. Only required for dec_types {“SCL”, “hybSCL”}.

  • num_iter (int) – Defining the number of BP iterations. Only required for dec_type “BP”.

  • return_crc_status (bool) – If True, the decoder additionally returns the CRC status indicating if a codeword was (most likely) correctly recovered.

  • precision (str | None) – Precision used for internal calculations and outputs. If None, precision is used.

  • device (str | None) – Device for computation (e.g., ‘cpu’, ‘cuda:0’). If None, device is used.

Inputs:

llr_ch – […, n], torch.float. Tensor containing the channel logits/llr values.

Outputs:

  • b_hat – […, k], torch.float. Binary tensor containing hard-decided estimations of all k information bits.

  • crc_status – […], torch.bool. CRC status indicating if a codeword was (most likely) correctly recovered. This is only returned if return_crc_status is True. Note that false positives are possible.

Notes

This block supports the uplink and downlink Polar rate-matching scheme without codeword segmentation.

Although the decoding list size is not provided by 3GPP [3GPPTS38212], the consortium has agreed on a list size of 8 for the 5G decoding reference curves [Bioglio_Design].

All list-decoders apply CRC-aided decoding, however, the non-list decoders (“SC” and “BP”) cannot materialize the CRC leading to an effective rate-loss.

Examples

import torch
from sionna.phy.fec.polar import Polar5GEncoder, Polar5GDecoder

k, n = 100, 200
encoder = Polar5GEncoder(k, n)
decoder = Polar5GDecoder(encoder, dec_type="SCL", list_size=8)

bits = torch.randint(0, 2, (10, k), dtype=torch.float32)
codewords = encoder(bits)
llr_ch = 20.0 * (2.0 * codewords - 1)  # BPSK without noise
decoded = decoder(llr_ch)
print(torch.equal(bits, decoded))
# True

Attributes

property k_target: int#

Number of information bits including rate-matching.

property n_target: int#

Codeword length including rate-matching.

property k_polar: int#

Number of information bits of mother Polar code.

property n_polar: int#

Codeword length of mother Polar code.

property llr_max: float#

Maximum LLR value for internal calculations.

property dec_type: str#

Decoder type used for decoding as str.

property polar_dec#

Decoder instance used for decoding.

Methods

build(input_shape: Tuple[int, ...]) → None[source]#

Build and check if shape of input is invalid.

Parameters:

input_shape (Tuple[int, ...])