KroneckerModel#

class sionna.phy.channel.KroneckerModel(r_tx: torch.Tensor | None = None, r_rx: torch.Tensor | None = None, precision: str | None = None, device: str | None = None)[source]#

Bases: sionna.phy.channel.spatial_correlation.SpatialCorrelation

Kronecker model for spatial correlation

Given a batch of matrices \(\mathbf{H}\in\mathbb{C}^{M\times K}\), \(\mathbf{R}_\text{tx}\in\mathbb{C}^{K\times K}\), and \(\mathbf{R}_\text{rx}\in\mathbb{C}^{M\times M}\), this function will generate the following output:

\[\mathbf{H}_\text{corr} = \mathbf{R}^{\frac12}_\text{rx} \mathbf{H} \mathbf{R}^{\frac12}_\text{tx}\]

Note that \(\mathbf{R}_\text{tx}\in\mathbb{C}^{K\times K}\) and \(\mathbf{R}_\text{rx}\in\mathbb{C}^{M\times M}\) must be positive semi-definite, such as the ones generated by exp_corr_mat().

Parameters:

  • r_tx (torch.Tensor | None) – Transmit correlation matrices of shape […, K, K]. If the rank of r_tx is smaller than that of the input h, it will be broadcast.

  • r_rx (torch.Tensor | None) – Receive correlation matrices of shape […, M, M]. If the rank of r_rx is smaller than that of the input h, it will be broadcast.

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

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

Inputs:

h – […, M, K], torch.complex. Spatially uncorrelated channel coefficients.

Outputs:

h_corr – […, M, K], torch.complex. Spatially correlated channel coefficients.

Examples

import torch
from sionna.phy.channel import exp_corr_mat, KroneckerModel

# Create correlation matrices
r_tx = exp_corr_mat(0.4, 4)  # 4x4 TX correlation
r_rx = exp_corr_mat(0.9, 16)  # 16x16 RX correlation

# Create model
kron = KroneckerModel(r_tx, r_rx)

# Apply to channel matrix
h = torch.randn(32, 16, 4, dtype=torch.complex64)
h_corr = kron(h)
print(h_corr.shape)
# torch.Size([32, 16, 4])

# For use with torch.compile, update matrices in-place:
new_r_tx = exp_corr_mat(0.5, 4)
kron.r_tx.copy_(new_r_tx)  # In-place update avoids graph breaks

Attributes

property r_tx: torch.Tensor | None#

Get/set transmit correlation matrices.

For use within torch.compile, use in-place updates via model.r_tx.copy_(new_value) to avoid graph breaks.

property r_rx: torch.Tensor | None#

Get/set receive correlation matrices.

For use within torch.compile, use in-place updates via model.r_rx.copy_(new_value) to avoid graph breaks.