Root Mean Square Error

class RootMeanSquareError(receiving_radar, transmitting_device, receiving_device, radar_channel)

Bases: RadarEvaluator, Serializable

Root mean square error of estimated points to ground truth.

Root mean square error (RMSE) is a widely used metric for evaluating the performance of a radar detector. It estimates the average distance between the estimated and the ground truth position of a target.

A minimal example within the context of a Simulation evaluating the probability of detection for a single radar target illuminated by an FMCW radar would be:

from hermespy.core import dB
from hermespy.radar import Radar, FMCW, MaxDetector, RootMeanSquareError
from hermespy.simulation import Simulation
from hermespy.channel import SingleTargetRadarChannel

# Create a new simulated scenario featuring a single device
simulation = Simulation(num_samples=1000)
device = simulation.new_device(carrier_frequency=60e9, bandwidth=3e9, oversampling_factor=1)

# Configure the device to transmit and reveive radar waveforms
radar = Radar(waveform=FMCW())
radar.detector = MaxDetector()
device.add_dsp(radar)

# Create a new radar channel with a single illuminated target
target = SingleTargetRadarChannel((1, radar.max_range(device.bandwidth)), 1., attenuate=False)
simulation.scenario.set_channel(device, device, target)

# Create a new detection probability evaluator
simulation.add_evaluator(RootMeanSquareError(radar, device, device, target))

# Sweep over the target's SNR during the simulation
simulation.new_dimension('noise_level', dB(0, -5, -10, -20, -30), device)

# Run the simulation
result = simulation.run()
result.plot()

Parameters:

• receiving_radar (Radar) – Radar detector to be evaluated.

• transmitting_device (SimulatedDevice) – Device transmitting into the evaluated channel.

• receiving_device (SimulatedDevice) – Device receiving from the evaluated channel. The receiving_radar must be a receive DSP algorithm of this device.

• radar_channel (RadarChannelBase) – The radar channel containing the ground truth to be evaluated.

Raises:

ValueError – If the receiving radar is not an operator of the radar_channel receiver.

evaluate()

Evaluate the state of an investigated object.

Implements the process of extracting an arbitrary performance indicator, represented by the returned Artifact \(X_m\).

Returns: Artifact \(X_m\) resulting from the evaluation.

Return type:

RootMeanSquareEvaluation

initialize_result(grid)

Initialize the respective result object for this evaluator.

Parameters:

grid (Sequence[GridDimensionInfo]) – The parameter grid over which the simulation iterates.

Return type:

RootMeanSquareErrorResult

Returns: The initialized evaluation result.

property abbreviation: str

Short string representation of this evaluator.

Used as a label for console output and plot axes annotations.

property title: str

Long string representation of this evaluator.

Used as plot title.

class RootMeanSquareEvaluation(pcl, ground_truth)

Bases: Evaluation[ScatterVisualization]

Result of a single root mean squre evaluation.

artifact()

Generate an artifact from this evaluation.

Returns: The evaluation artifact.

Return type:

RootMeanSquareArtifact

class RootMeanSquareArtifact(num_errors, cummulation)

Bases: Artifact

Artifact of a root mean square evaluation

Parameters:

• num_errors (int) – Number of errros.

• cummulation (float) – Sum of squared errors distances.

to_scalar()

Scalar representation of this artifact’s content.

Used to evaluate premature stopping criteria for the underlying evaluation.

Return type:

float

Returns:

Scalar floating-point representation. None if a conversion to scalar is impossible.

property cummulation: float

Cummulated squared error

property num_errors: int

Number of cummulated errors

class RootMeanSquareErrorResult(grid, evaluator)

Bases: EvaluationResult[RootMeanSquareArtifact]

Result of a root mean square error evaluation.

Parameters:

• grid (Sequence[GridDimensionInfo]) – Simulation grid.

• evaluator (RootMeanSquareError) – Evaluator associated with this result.

add_artifact(coordinates, artifact, compute_confidence=True)

Add an artifact to this evaluation result.

Parameters:

• coordinates (tuple[int, ...]) – Coordinates of the grid section to which the artifact belongs.

• artifact (RootMeanSquareArtifact) – Artifact to be added.

• compute_confidence (bool) – Whether to compute the confidence level of the evaluation result for the given section coordinates.

Return type:

bool

Returns:

Can the result be trusted? Always False if compute_confidence is set to False.

runtime_estimates()

Extract a runtime estimate for this evaluation result.

Returns: A numpy array containing the runtime estimates for each grid section. If no estimates are available, None is returned.

Return type:

ndarray

to_array()

Convert the evaluation result raw data to an array representation.

Used to store the results in arbitrary binary file formats after simulation execution.

Returns: The array result representation.

Return type:

ndarray