"""Collection of actions to perform analysis on squeezing."""
import numpy as np
from numpy import cosh, sinh, cos
from finesse.analysis.actions import FrequencyResponse
from finesse.solutions import BaseSolution
from finesse.analysis.actions.base import Action, names_to_nodes
[docs]class AntiSqueezingSolution(BaseSolution):
pass
# IMPORTANT: renaming this class impacts the katscript spec and should be avoided!
[docs]class AntiSqueezing(Action):
"""Computes the amount of anti-squeezing at an output from a squeezing element.
Notes
-----
This will only works in cases where the HG00 is squeezed and is used
after a filtering component, like a cavity to select only the HG00.
The calculation does not use the usual internal quantum noise solver in
Finesse. Instead it computes how much loss and rotation from a squeezer
happens by calculating the upper and lower sideband transfer functions
from the squeezer to some readout. From this the relevant squeezing outputs
can be calculated.
Parameters
----------
f : array_like
Signal frequencies to compute the anti-squeezing over
squeezer : str
Name of squeezing component
readout : str
Name of readout port to compute squeezing at
signal : str
Name of signal drive to calculate a the signal transfer function of.
This is returned in `sol.signal` and can be used to scale the noise
into equivalent units of some signal.
"""
def __init__(self, f, squeezer, readout, *, signal=None, name="antisqueezing"):
super().__init__(name)
self.squeezer = squeezer
self.readout = readout
self.signal = signal
inputs = [f"{self.squeezer}.upper", f"{self.squeezer}.lower"]
if self.signal:
inputs.append(self.signal)
self.frequency_response = FrequencyResponse(f, inputs, self.readout)
@property
def f(self):
return self.frequency_response.f
def _requests(self, model, memo, first=True):
self.frequency_response._requests(model, memo)
def _do(self, state):
# Here we grab the readout element and from that the optical
# port for getting the LO used
signal_node = names_to_nodes(
state.model, (self.readout,), default_hints=("output",)
)[0]
readout = signal_node.component
opt_node = readout.p1.i
rhs_idx = state.sim.carrier.field(opt_node, 0, 0)
sol = AntiSqueezingSolution(self.name)
# Power scaling needed here 1/sqrt(2)
E_LO = state.sim.carrier.M().rhs_view[0, rhs_idx] / np.sqrt(2)
P_carrier = abs(E_LO) ** 2
f = 299792458.0 / float(state.sim.model_settings.lambda0)
h = 6.62607015e-34
sol.shotASD = np.sqrt(2 * P_carrier * h * f) # pure shot noise at output
sol.f = self.f
sol_rot = state.apply(self.frequency_response)
sol.Hu = sol_rot[f"{self.squeezer}.upper"]
sol.Hl = sol_rot[f"{self.squeezer}.lower"]
if self.signal is not None:
sol.signal = abs(sol_rot[self.signal])
else:
sol.signal = np.ones_like(sol.Hu)
# From these we can calculate the relative rotation
# of each which tells us how much the squeezing is rotating
# and how much anti-squeezing will appear
db = float(state.model.elements[self.squeezer].db)
r = db / 8.685889638065037
# rotation of squeezed state
sol.angle = (np.angle(sol.Hu) - np.angle(sol.Hl)) / 2
ASD = sol.shotASD * np.sqrt(cosh(2 * r) + sinh(2 * r) * cos(2 * sol.angle))
# Noise if pure squeezing at all frequencies
sol.ideal_noise = sol.shotASD * np.exp(-r)
# Take into account any sideband loss rotation
sol.scale = (abs(sol.Hu) + abs(sol.Hl)) / 2 / np.sqrt(P_carrier)
sol.anti_squeezing_noise = sol.scale * np.sqrt(
np.abs(ASD**2 - sol.ideal_noise**2)
)
# Equations from https://doi.org/10.1103/PhysRevD.104.062006
sol.nu = (abs(sol.Hu) ** 2 + abs(sol.Hl) ** 2) / 2 # Efficiency Eq 52
sol.Xi = (abs(sol.Hu) - abs(sol.Hl)) ** 2 / (
4 * sol.nu
) # Intrinsic dephasing Eq 53
return sol