pytwinnet.accelerate

Accelerated (vectorized) utilities for PyTwinNet.

pytwinnet.accelerate.fspl_matrix_db(tx_xyz, rx_xyz, f_hz)

Vectorized Free-Space Path Loss in dB: (T,R) If f_hz is scalar and Numba is available -> JIT kernel.

Return type:

ndarray

Parameters:

• tx_xyz (ndarray)

• rx_xyz (ndarray)

• f_hz (ndarray | float)

pytwinnet.accelerate.max_rsrp_association_vectorized(twin, tx_ids, ue_ids)

Vectorized max-RSRP association. Much faster than per-pair loops for large sets.

Return type:

Dict[str, str]

Parameters:

• twin (DigitalTwin)

• tx_ids (List[str])

• ue_ids (List[str])

pytwinnet.accelerate.noise_dbm_vector(bandwidth_hz, temperature_k=290.0, noise_figure_db=0.0)

Vectorized thermal noise power (dBm). bandwidth_hz: scalar or (R,), noise_figure_db: scalar or (R,) -> (R,)

Return type:

ndarray

Parameters:

• bandwidth_hz (ndarray | float)

• temperature_k (float)

• noise_figure_db (ndarray | float)

pytwinnet.accelerate.rsrp_matrix_dbm(tx_dbm, gt_dbi, gr_dbi, pl_db)

RSRP (received power) in dBm for all TX-RX pairs. tx_dbm, gt_dbi: (T,), gr_dbi: (R,), pl_db: (T,R) -> (T,R)

Return type:

ndarray

Parameters:

• tx_dbm (ndarray)

• gt_dbi (ndarray)

• gr_dbi (ndarray)

• pl_db (ndarray)

pytwinnet.accelerate.shannon_throughput_bps_vector(bandwidth_hz, sinr_db, efficiency=1.0)

Vectorized Shannon throughput per RX (bps).

Return type:

ndarray

Parameters:

• bandwidth_hz (ndarray | float)

• sinr_db (ndarray)

• efficiency (float)

pytwinnet.accelerate.sinr_db_from_rsrp_matrix(rsrp_dbm, serving_tx_idx, noise_dbm)

Return type:

ndarray

Parameters:

• rsrp_dbm (ndarray)

• serving_tx_idx (ndarray)

• noise_dbm (ndarray)

Modules

association
numba_kernels
vectorized