calculate_tb_size#

sionna.phy.nr.utils.calculate_tb_size(modulation_order: int | torch.Tensor, target_coderate: float | torch.Tensor, target_tb_size: int | float | torch.Tensor | None = None, num_coded_bits: int | torch.Tensor | None = None, num_prbs: int | torch.Tensor | None = None, num_ofdm_symbols: int | torch.Tensor | None = None, num_dmrs_per_prb: int | torch.Tensor | None = None, num_layers: int | torch.Tensor = 1, num_ov: int | torch.Tensor = 0, tb_scaling: float | torch.Tensor = 1.0, return_cw_length: bool = True, verbose: bool = False, device: str | None = None) → Tuple[source]#

Calculates the transport block (TB) size for given system parameters.

This function follows the procedure defined in TS 38.214 Sec. 5.1.3.2 and Sec. 6.1.4.2 [3GPPTS38214].

Parameters:

  • modulation_order (int | torch.Tensor) – Modulation order, i.e., number of bits per QAM symbol.

  • target_coderate (float | torch.Tensor) – Target coderate.

  • target_tb_size (int | float | torch.Tensor | None) – Target transport block size, i.e., number of information bits that can be encoded into a slot for the given slot configuration.

  • num_coded_bits (int | torch.Tensor | None) – Number of coded bits that can be fit into a given slot. If provided, num_prbs, num_ofdm_symbols and num_dmrs_per_prb are ignored.

  • num_prbs (int | torch.Tensor | None) – Total number of allocated PRBs per OFDM symbol, where 1 PRB equals 12 subcarriers. Must not exceed 275.

  • num_ofdm_symbols (int | torch.Tensor | None) – Number of OFDM symbols allocated for transmission. Cannot be larger than 14.

  • num_dmrs_per_prb (int | torch.Tensor | None) – Number of DMRS (i.e., pilot) symbols per PRB that are not used for data transmission, across all num_ofdm_symbols OFDM symbols.

  • num_layers (int | torch.Tensor) – Number of MIMO layers.

  • num_ov (int | torch.Tensor) – Number of unused resource elements due to additional overhead as specified by higher layer.

  • tb_scaling (float | torch.Tensor) – TB scaling factor for PDSCH as defined in TS 38.214 Tab. 5.1.3.2-2.

  • return_cw_length (bool) – If True, the function returns tb_size, cb_size, num_cb, tb_crc_length, cb_crc_length, cw_length. Otherwise, it does not return cw_length to reduce computation time.

  • verbose (bool) – If True, additional information is printed.

  • device (str | None) – Device for computation. If None, device is used.

Outputs:

  • tb_size – […], torch.int32. Transport block (TB) size, i.e., how many information bits can be encoded into a slot for the given slot configuration.

  • cb_size – […], torch.int32. Code block (CB) size, i.e., the number of information bits per codeword, including the TB/CB CRC parity bits.

  • num_cb – […], torch.int32. Number of CBs that the TB is segmented into.

  • tb_crc_length – […], torch.int32. Length of the TB CRC.

  • cb_crc_length – […], torch.int32. Length of each CB CRC.

  • cw_length – […, N], torch.int32. Codeword length of each of the num_cbs codewords after LDPC encoding and rate-matching. Note that zeros are appended along the last axis to obtain a dense tensor. The total number of coded bits, num_coded_bits, is the sum of cw_length across its last axis. Only returned if return_cw_length is True.

Examples

from sionna.phy.nr.utils import calculate_tb_size

tb_size, cb_size, num_cb, tb_crc, cb_crc, cw_len = calculate_tb_size(
    modulation_order=4,
    target_coderate=0.5,
    num_coded_bits=4800,
    num_layers=1
)
print(f"TB size: {tb_size}, CB size: {cb_size}, Num CBs: {num_cb}")
On this page

This Page