Public Member Functions | Private Member Functions | Private Attributes

ReduOptAmplifier Class Reference

#include <ocsReduOptAmplifier.hh>

List of all members.

Public Member Functions

 ReduOptAmplifier (string InFileName, ReduOptSignal *oReduOptSignalTmp)
 ~ReduOptAmplifier ()
 Destructor, calls void ReleaseMemory(void).
void SetFixedOutputPowerFromCurrentOptSignal (void)
 Used to initialize OutputPowerLinear.
void SetFixedTotalOutputPowerFromCurrentOptSignal (void)
void AmplifyOptSignal (void)
 Amplifies the reduced optical signal and adds PDG.
void SetTypeAmplifier (typeAmplifier TypeAmplifier2)
typeAmplifier GetTypeAmplifier (void)
void SetTypeAmplifierNoise (typeAmplifierNoise TypeAmplifierNoise2)
void SetPolDepGain (double PolDepGainOptAmplifier2)
void SetGain (double dBGain2)
double SteadyStateSaturatedGain (double InputPower)
void SetGainCompressionSaturatedAmplifier (double GainTarget_dB, double GainCompression_dB, double OutputPowerTarget)
void SyncNumChannels (void)
 Sets NumChannels = oReduOptSignal::NumChannels.
string WriteTypeAmplifier (void)
string WriteTypeAmplifierNoise (void)
void WriteSteadyStateGainVersusPower (string OutFileName, double InputPowerMaxLinear, int NumPowerSteps)
 NOT USED by reduced model.
void WriteSteadyInputPowerVersusOutputPower (string OutFileName, double InputPowerMaxLinear, int NumPowerSteps)
double GetSaturatingPowerLinear (void)
 NOT USED by reduced model.
double GetPolDepGain (void)
double GetOutputPowerLinear (void)
double GetLinearGainOptAmplif (void)

Private Member Functions

void AllocateNoiseAmplitudeChannels (void)
void ReleaseMemory (void)
 Releases memory allocated by void AllocateNoiseAmplitudeChannels(void).
void AddPolDepGainEffect (void)
 Adds the PDG. Called by void AmplifyOptSignal(void).
double GetGainFixedOutputPower (double InputPower)
 This method computes the linear gain of the amplifier.

Private Attributes

 Pointer to the reduced optical signal object that is being amplified.
double StokesSignalTmp [4]
double StokesNoiseTmp [4]
double CenterFreq
double ChannSpacing
int NumChannels
double dBGain
double OutputPowerLinear
double PolDepGainOptAmplifier
 The polarization dependend gain in dB, same as x_PDG in [WM 2001] p 489.
double LinearGainOptAmplif
 The linear gain of the amplifier.
double NoiseFigOptAmplifier
double LinearNoiseFigOptAmplif
double SpontaneousEmissionFactor
double AmpLength
 NOT USED by reduced model.
int NumZSteps
 NOT USED by reduced model.
double SaturatingPower
 NOT USED by reduced model.
double SaturatingPowerLinear
 NOT USED by reduced model.
double SaturatingPowerdBm
 NOT USED by reduced model.
double UnsaturatedGaindB
 NOT USED by reduced model.
double UnsaturatedGainLinear
 NOT USED by reduced model.
double UnsaturatedDifferentialGainLinear
 NOT USED by reduced model.
double * NormalizedNoiseAmplitude
typeAmplifier TypeAmplifier
 The type of amplifier.
typeAmplifierNoise TypeAmplifierNoise
 The type of amplifier noise.
int DebugLevel

Detailed Description

The reduced optical amplifier.
The model includes a simple model for gain saturated amplifiers as well as polarization dependent gain (PDG).

For a simple application you just need to call the constructor ReduOptAmplifier( string InFileName,ReduOptSignal *oReduOptSignalTmp) and the method void AmplifyOptSignal(void)

Constructor & Destructor Documentation

ReduOptAmplifier::ReduOptAmplifier ( string  InFileName,
ReduOptSignal oReduOptSignalTmp 


string InFileName: The name of the input parameter file
ReduOptSignal *oReduOptSignalTmp: Pointer to the reduced optical signal object that is to be amplified.

The constructor does the following:

  1. Initializes class parameters that it gets from the ReduOptSignal object
  2. Reads input parameters from file and initializes internal class parameters.
  3. Initializes a ReduPolDepAttenuator object that is used internally to do the PDG.
  4. Calls void AllocateNoiseAmplitudeChannels(void)

The parameters that need to be set in the input parameter file are as follows.

(Note that in some of the example input files we also include many other parameters. These are for our full time and frequency domain amplifier and are NOT required by the reduced amplifier. Consequently this constructor could be cleaned up quite a bit...)

  1. TypeAmplifier: Should be set to 6; See ReduOptAmplifier::TypeAmplifier
  2. TypeAmplifierNoise: Should be set to 4; See ReduOptAmplifier::TypeAmplifierNoise
  3. NoiseFigOptAmplifier: The noise figure of the amplifier in dB
  4. SpontEmissionFactor: The spontaneous emission factor for the noise, n_sp, on a linear scale.
  5. PolDepGainOptAmplifier: The PDG in dB.

We only need to set one of NoiseFigOptAmplifier and SpontEmissionFactor since SpontaneousEmissionFactor = LinearNoiseFigOptAmplif/2.0 where LinearNoiseFigOptAmplif = dB2Linear(NoiseFigOptAmplifier)

The code knows which of these two parameters to use provided that the user sets the other one to zero in the input file.

So exactly one of NoiseFigOptAmplifier and SpontEmissionFactor must be zero in the input file.

References AllocateNoiseAmplitudeChannels(), AmpLength, CenterFreq, ChannSpacing, dB2Linear(), dBGain, dBm2Linear(), DebugLevel, ReduOptSignal::GetCenterFreq(), ReduOptSignal::GetChannSpacing(), ReduOptSignal::GetNumChannels(), LinearGainOptAmplif, LinearNoiseFigOptAmplif, log(), LogFile, LogFileSeparator(), LOWER_AND_UPPER, LOWER_ONLY, MANDATORY, NO_BOUNDS, NOISE_ON_SEMI_ANALYTICAL, NoiseFigOptAmplifier, NumChannels, NumZSteps, OPTIONAL_NO_WARNING, OPTIONAL_WARNING, oReduOptSignal, oReduPolDepAttenuator, OutputPowerLinear, PolDepGainOptAmplifier, ReadDouble(), ReadInt(), SaturatingPowerdBm, SaturatingPowerLinear, ReduPolDepAttenuator::SetPolDepAttenuation(), SpontaneousEmissionFactor, TypeAmplifier, TypeAmplifierNoise, UnsaturatedDifferentialGainLinear, UnsaturatedGaindB, UnsaturatedGainLinear, VECTOR_FIXED_OUTPUT_POWER, VECTOR_FIXED_TOTAL_OUTPUT_POWER, WriteTypeAmplifier(), and WriteTypeAmplifierNoise().

ReduOptAmplifier::~ReduOptAmplifier (  ) 

Destructor, calls void ReleaseMemory(void).

References ReleaseMemory().

Member Function Documentation

void ReduOptAmplifier::AddPolDepGainEffect ( void   )  [private]

Adds the PDG. Called by void AmplifyOptSignal(void).

The PDG is implemented somewhat differently from the formulae given on page 489 of [WM 2001] but the final result will be identical.

In this method the idea is that PDG can be modeled by a negative-loss (i.e. gain!) PDL element, provided that the signal is rotated so that the high loss axis of the (negative loss) PDL element is lined up with the high gain direction for the PDG. The high gain direction for the PDG is the direction that is orthogonal to the direction determined by the total Stokes parameters.

The amount of gain in the high gain direction is given by
PolDepGainOptAmplifier*GetDegreeOfPolarization(StokesTotal) in dB.

References GetDegreeOfPolarization(), NumChannels, oReduOptSignal, oReduPolDepAttenuator, ReduPolDepAttenuator::PolDepAttenuate(), PolDepGainOptAmplifier, RotatesAboutY(), RotatesAboutZ(), ReduPolDepAttenuator::SetPolDepAttenuationX(), ReduOptSignal::StokesNoise, ReduOptSignal::StokesSignal, and TransRzRyV3to_Xhat().

Referenced by AmplifyOptSignal().

void ReduOptAmplifier::AllocateNoiseAmplitudeChannels ( void   )  [private]
void ReduOptAmplifier::AmplifyOptSignal ( void   ) 

Amplifies the reduced optical signal and adds PDG.

In the reduced model we use ReduOptAmplifier::TypeAmplifier = 6 = VECTOR_FIXED_TOTAL_OUTPUT_POWER

So you should ignore most the the options in the switch statement in this method.

The algorithm is as follows:

  1. Compute LinearGainOptAmplifier = GetGainFixedOutputPower(oReduOptSignal->GetTotalPower())
    The idea here is to chose the gain that will ensure that the total output power is given by the pre-specified value OutputPowerLinear
  2. Amplify with gain given by LinearGainOptAmplifier. The amplification involves multiplying the Stokes parameters of the signal and noise in each channel as well as the oReduOptSignal->StokesExtraUnpolChannNoise by LinearGainOptAmplifier.
  3. Add the noise in the oReduOptSignal->StokesExtraUnpolChannNoise where the noise figure is computed using the frequency of the central channel, which is assumed have index NumChannels/2 which it will if you use set the flag EvenlySpacedChannels = 1 in the reduced optical signal input parameter file.
  4. Add the noise to the Noise Power in each of the channels.
  5. Now that we have added noise the total power will be higher than the value we wnted it to be. So we decrease the power of the signal in each channel to end up with the required total output power. Note that we don't change the power of the noise in this step.
  6. The consequence of the last step is that after many amplifications the noise power grows and the signal power drops. This is what happens in realistic gain saturated amplifiers.
See also:
void SetFixedOutputPowerFromCurrentOptSignal(void),
double GetGainFixedOutputPower(double InputPower),
double OutputPowerLinear,
void AddPolDepGainEffect(void)

References AddPolDepGainEffect(), ChannSpacing, dB2Linear(), dBGain, DebugLevel, GetGainFixedOutputPower(), ReduOptSignal::GetSignalPower(), ReduOptSignal::GetTotalPower(), Linear2dBm(), LinearGainOptAmplif, LogFile, LogFileSeparator(), NO_AMPLIFICATION, NOISE_ON_SEMI_ANALYTICAL, NormalizedNoiseAmplitude, NumChannels, oReduOptSignal, PolDepGainOptAmplifier, SCALAR_SATURATED, SCALAR_SIMPLE, SteadyStateSaturatedGain(), ReduOptSignal::StokesNoise, ReduOptSignal::StokesSignal, TypeAmplifier, TypeAmplifierNoise, VECTOR_FIXED_OUTPUT_POWER, VECTOR_FIXED_TOTAL_OUTPUT_POWER, VECTOR_SATURATED, and VECTOR_SIMPLE.

Referenced by OptAmplifier::AmplifyOptSignal().

double ReduOptAmplifier::GetGainFixedOutputPower ( double  InputPower  )  [private]

This method computes the linear gain of the amplifier.

The method returns the linear gain required to change the total power from an input value given by the parameter InputPower to the required TotalOutputPower = NumChannels*OutputPowerLinear. Note that the gain may actually be a loss in some circumstances.

See also:

NewNormalizedNoiseAmplitudePerChannel /= oReduOptSignal->GetNumChannels();

References ChannSpacing, LinearGainOptAmplif, NormalizedNoiseAmplitude, NumChannels, and OutputPowerLinear.

Referenced by AmplifyOptSignal().

double ReduOptAmplifier::GetLinearGainOptAmplif ( void   )  [inline]

References LinearGainOptAmplif.

double ReduOptAmplifier::GetOutputPowerLinear ( void   )  [inline]

References OutputPowerLinear.

double ReduOptAmplifier::GetPolDepGain ( void   )  [inline]
double ReduOptAmplifier::GetSaturatingPowerLinear ( void   )  [inline]

NOT USED by reduced model.

References SaturatingPowerLinear.

typeAmplifier ReduOptAmplifier::GetTypeAmplifier ( void   )  [inline]

References TypeAmplifier.

void ReduOptAmplifier::ReleaseMemory ( void   )  [private]

Releases memory allocated by void AllocateNoiseAmplitudeChannels(void).

References NormalizedNoiseAmplitude, and oReduPolDepAttenuator.

Referenced by SyncNumChannels(), and ~ReduOptAmplifier().

void ReduOptAmplifier::SetFixedOutputPowerFromCurrentOptSignal ( void   ) 

Used to initialize OutputPowerLinear.

Our simple amplifier gain saturation model keeps the total power in the optical signal constant. To determine the constant output power level per channel the user should call this method in an application once just after the (noiseless) signal has been generated. It sets OutputPowerLinear to be the average power per channel on a linear scale of the current signal.

See also:
void AmplifyOptSignal(void)

References ReduOptSignal::GetTotalPower(), oReduOptSignal, and OutputPowerLinear.

Referenced by SetFixedTotalOutputPowerFromCurrentOptSignal().

void ReduOptAmplifier::SetFixedTotalOutputPowerFromCurrentOptSignal ( void   )  [inline]
void ReduOptAmplifier::SetGain ( double  dBGain2  ) 
void ReduOptAmplifier::SetGainCompressionSaturatedAmplifier ( double  GainTarget_dB,
double  GainCompression_dB,
double  OutputPowerTarget 
void ReduOptAmplifier::SetPolDepGain ( double  PolDepGainOptAmplifier2  )  [inline]
void ReduOptAmplifier::SetTypeAmplifier ( typeAmplifier  TypeAmplifier2  ) 
void ReduOptAmplifier::SetTypeAmplifierNoise ( typeAmplifierNoise  TypeAmplifierNoise2  ) 
double ReduOptAmplifier::SteadyStateSaturatedGain ( double  InputPower  ) 
void ReduOptAmplifier::SyncNumChannels ( void   ) 
void ReduOptAmplifier::WriteSteadyInputPowerVersusOutputPower ( string  OutFileName,
double  InputPowerMaxLinear,
int  NumPowerSteps 
void ReduOptAmplifier::WriteSteadyStateGainVersusPower ( string  OutFileName,
double  InputPowerMaxLinear,
int  NumPowerSteps 

NOT USED by reduced model.

References LogFile, LogFileSeparator(), and SteadyStateSaturatedGain().

string ReduOptAmplifier::WriteTypeAmplifier ( void   ) 
string ReduOptAmplifier::WriteTypeAmplifierNoise ( void   ) 

Member Data Documentation

double ReduOptAmplifier::AmpLength [private]
double ReduOptAmplifier::CenterFreq [private]

Referenced by ReduOptAmplifier().

double ReduOptAmplifier::dBGain [private]

The linear gain of the amplifier.

Referenced by AmplifyOptSignal(), GetGainFixedOutputPower(), GetLinearGainOptAmplif(), ReduOptAmplifier(), and SetGain().

Referenced by ReduOptAmplifier().

Referenced by ReduOptAmplifier().

This double array stores the NormalizedNoiseAmplitude factor for each channel.

This factor tells us how much noise is added by the amplifier and depends on the frequency of the channel. Note that we are using an amplifier with flat gain. The frequency dependence of the amount of noise that is added is explained in both of Agrawal's books.

Referenced by AllocateNoiseAmplitudeChannels(), AmplifyOptSignal(), GetGainFixedOutputPower(), and ReleaseMemory().

NOT USED by reduced model.

Referenced by ReduOptAmplifier(), and SteadyStateSaturatedGain().

Pointer to the reduced optical signal object that is being amplified.

Referenced by AddPolDepGainEffect(), AllocateNoiseAmplitudeChannels(), AmplifyOptSignal(), ReduOptAmplifier(), SetFixedOutputPowerFromCurrentOptSignal(), and SyncNumChannels().

A ReduPolDepAttenuator object that is used internally to do the PDG

Referenced by AddPolDepGainEffect(), ReduOptAmplifier(), ReleaseMemory(), and SyncNumChannels().

The output power per channel we want in the case of a noiseless amplifier.

With ReduOptAmplifier::TypeAmplifier = VECTOR_FIXED_TOTAL_OUTPUT_POWER and when noise is added by the amplifier we keep the total power TotalPower = NumChannels*OutputPowerLinear constant.

See also:
void AmplifyOptSignal(void)

Referenced by GetGainFixedOutputPower(), GetOutputPowerLinear(), ReduOptAmplifier(), and SetFixedOutputPowerFromCurrentOptSignal().

The polarization dependend gain in dB, same as x_PDG in [WM 2001] p 489.

Referenced by AddPolDepGainEffect(), AmplifyOptSignal(), GetPolDepGain(), ReduOptAmplifier(), and SetPolDepGain().

NOT USED by reduced model.

NOT USED by reduced model.

Referenced by ReduOptAmplifier().

double ReduOptAmplifier::StokesNoiseTmp[4] [private]
double ReduOptAmplifier::StokesSignalTmp[4] [private]

The bandwidth of the polarized noise in a channel

See also:

The type of amplifier.

In the reduced model we use TypeAmplifier = 6 = VECTOR_FIXED_TOTAL_OUTPUT_POWER

Referenced by AmplifyOptSignal(), GetTypeAmplifier(), ReduOptAmplifier(), SetTypeAmplifier(), and WriteTypeAmplifier().

The type of amplifier noise.

In the reduced model we use TypeAmplifierNoise = ON which means that we include noise in the system

Referenced by AmplifyOptSignal(), ReduOptAmplifier(), SetTypeAmplifierNoise(), and WriteTypeAmplifierNoise().

NOT USED by reduced model.

Referenced by ReduOptAmplifier().

NOT USED by reduced model.

Referenced by ReduOptAmplifier(), and SetGainCompressionSaturatedAmplifier().

The documentation for this class was generated from the following files: