geometry_sir_db#

sionna.sys.geometry_sir_db(coupling_loss, serving: torch.Tensor | None = None, precision: Literal['single', 'double'] | None = None, device: str | None = None) → torch.Tensor[source]#

Computes geometry SIR from coupling losses.

Geometry SIR is the interference-limited signal-to-interference ratio

\[\mathrm{SIR}_{u} = \frac{P_{r,b^\star,u}} {\sum_{b\ne b^\star} P_{r,b,u}},\]

where \(b^\star\) is the serving BS and received powers are obtained from coupling losses. Transmit powers are assumed equal for all base stations and therefore cancel.

Geometry SIR is a 3GPP TR 38.901 system calibration metric from Section 7.8 of [TR38901V160100]. It is used to verify interference geometry independently of thermal noise.

The ratio is evaluated from relative gains in the log domain, making the result invariant to large common coupling-loss offsets without numerical underflow or overflow.

Parameters:

  • coupling_loss – Coupling loss [dB] with shape [..., num_bs].

  • serving (torch.Tensor | None) – Serving BS indices with shape [...]. If None, the serving BS is selected by serving_indices().

  • precision (Literal['single', 'double'] | None) – Precision used for internal calculations and outputs. If set to None, precision is used.

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

Outputs:

sir – Geometry SIR [dB] with shape [...].