mx::AO::analysis::clGainOpt< _realT > Struct Template Reference

template<typename _realT>
struct mx::AO::analysis::clGainOpt< _realT >

A class to manage optimizing closed-loop gains.

Template Parameters

_realT

the real floating point type in which to do all arithmetic. Is used to define the complex type as well.

Definition at line 68 of file clGainOpt.hpp.

#include <ao/analysis/clGainOpt.hpp>

Classes
struct	maxStableGainReport
	Diagnostic summary of a maximum-stable-gain search. More...
struct	optGainReport
	Diagnostic summary of an open-loop optimum-gain search. More...

Public Types
enum class	maxStableGainStatus { notRun , crossingFound , invalidInput , noCrossing }
	Termination state of a maximum-stable-gain search. More...
enum class	optGainStatus {   notRun , converged , boundaryLimited , invalidInput ,   stabilityFailure , iterationLimit , calculationFailure }
	Termination state of an open-loop optimum-gain search. More...
typedef _realT	realT
	The real data type.
typedef std::complex< _realT >	complexT
	The complex data type.

Public Member Functions
	clGainOpt ()
	Default c'tor.
	clGainOpt (realT Ti, realT tau)
	C'tor setting the loop timings.
void	init ()
	Initialize this instance.
int	N ()
	Get the number of integrations in the (optional) moving average.
void	N (int newN)
	Set the number of integrations in the moving average.
realT	Ti ()
	Get the loop sampling interval.
void	Ti (realT newTi)
	Set the loop sampling interval.
realT	tau ()
	Get the loop delay.
void	tau (realT newTau)
	Set the loop delay.
void	b (const std::vector< realT > &newB)
	Set the vector of FIR coefficients.
void	b (const Eigen::Array< realT, -1, -1 > &newB)
	Set the vector of FIR coefficients.
realT	b (size_t i)
	Get a single FIR coefficient.
const std::vector< realT > &	b ()
	Get the vector of FIR coefficients.
void	a (const std::vector< realT > &newA)
	Set the vector of IIR coefficients.
void	a (const Eigen::Array< realT, -1, -1 > &newA)
	Set the vector of IIR coefficients.
realT	a (size_t i)
	Get a single IIR coefficient.
const std::vector< realT > &	a ()
	Get the vector of IIR coefficients.
void	remember (const realT &rem)
	Set the remember factor for a leaky integrator.
realT	remember ()
	Get the remember factor.
void	setLeakyIntegrator (realT remember)
	Set the FIR and IIR coefficients so that the control law is a leaky integrator.
void	f (realT *newF, size_t nF)
	Set the vector of frequencies.
void	f (const std::vector< realT > &newF)
	Set the vector of frequencies.
size_t	f_size ()
	Get the size of the frequency vector.
realT	f (size_t i)
	Get the i-th value of frequency.
complexT	olXfer (int fi, complexT &H_dm, complexT &H_del, complexT &H_con)
	Calculate the open-loop transfer function.
complexT	olXfer (int fi)
	Calculate the open-loop transfer function.
complexT	clETF (int fi, realT g)
	Return the closed loop error transfer function (ETF) at frequency f for gain g.
realT	clETFPhase (int fi, realT g)
	Return the closed loop error transfer function (ETF) phase at frequency f for gain g.
realT	clETF2 (int fi, realT g)
	Return the norm of the closed loop error transfer function (ETF) at frequency f for gain g.
complexT	clNTF (int fi, realT g)
	Return the closed loop noise transfer function (NTF) at frequency f for gain g.
realT	clNTF2 (int fi, realT g)
	Return the norm of the closed loop noise transfer function (NTF) at frequency f for gain g.
void	clTF2 (realT &ETF, realT &NTF, int fi, realT g)
	Return the norm of the closed loop transfer functions at frequency f for gain g.
realT	clVariance (realT &varErr, realT &varNoise, const std::vector< realT > &PSDerr, const std::vector< realT > &PSDnoise, realT g)
	Calculate the closed loop variance given open-loop PSDs and gain.
realT	clVariance (const std::vector< realT > &PSDerr, const std::vector< realT > &PSDnoise, realT g)
	Calculate the closed loop variance given open-loop PSDs and gain.
mx::error_t	maxStableGain (realT &gain, maxStableGainReport *report=nullptr)
	Find the maximum stable gain for the loop parameters.
mx::error_t	optGainOpenLoop (realT &gain, realT &var, const std::vector< realT > &PSDerr, const std::vector< realT > &PSDnoise, bool gridSearch, optGainReport *report=nullptr)
	Return the optimum closed loop gain given an open loop PSD.
mx::error_t	optGainOpenLoop (realT &gain, realT &var, const std::vector< realT > &PSDerr, const std::vector< realT > &PSDnoise, realT maximumGain, bool gridSearch, optGainReport *report=nullptr)
	Return the optimum closed loop gain given an open loop PSD.
int	pseudoOpenLoop (std::vector< realT > &PSD, realT g)
	Calculate the pseudo open-loop PSD given a closed loop PSD.

Public Attributes
realT	m_maxFindMin
	The Minimum value for the maximum stable gain finding algorithm.
realT	m_minFindMin
	The Minimum value for the minimum finding algorithm.
realT	m_minFindMaxFact
	The maximum value, as a multiplicative factor of maximum gain.
int	m_minFindBits
uintmax_t	m_minFindMaxIter
	The maximum iterations allowed for minimization.

Protected Attributes
int	m_N
	Number of integrations in the (optional) moving average. Default is 1.
realT	m_Ti
	The loop sampling interval.
realT	m_tau
	The loop delay.
realT	m_remember { 1.0 }
	The leaky integrator forget factor.
std::vector< realT >	m_b
	Vector of FIR coefficients.
std::vector< realT >	m_a
	Vector of IIR coefficients.
std::vector< realT >	m_f
	Vector of frequencies.
bool	m_trigCacheChanged { true }
	True when frequency, sampling interval, or required controller tap count invalidates m_cs and m_ss.
bool	m_changed { true }
	True if any of the members which make up the basic transfer functions are changed.

Member Typedef Documentation

complexT

template<typename _realT>

typedef std::complex<_realT> mx::AO::analysis::clGainOpt< _realT >::complexT

The complex data type.

Definition at line 71 of file clGainOpt.hpp.

realT

template<typename _realT>

typedef _realT mx::AO::analysis::clGainOpt< _realT >::realT

The real data type.

Definition at line 70 of file clGainOpt.hpp.

Member Enumeration Documentation

maxStableGainStatus

template<typename _realT>

enum class mx::AO::analysis::clGainOpt::maxStableGainStatus

strong

Termination state of a maximum-stable-gain search.

Enumerator
notRun	No search has been attempted.
crossingFound	A qualifying Nyquist crossing was found.
invalidInput	The frequency grid or derived Nyquist values were invalid.
noCrossing	No qualifying Nyquist crossing was found.

Definition at line 74 of file clGainOpt.hpp.

optGainStatus

template<typename _realT>

enum class mx::AO::analysis::clGainOpt::optGainStatus

strong

Termination state of an open-loop optimum-gain search.

Enumerator
notRun	No search has been attempted.
converged	The minimizer converged inside the search interval.
boundaryLimited	The reported minimum lies on a search boundary.
invalidInput	The PSDs, search controls, or requested interval were invalid.
stabilityFailure	The automatic maximum-stable-gain search failed.
iterationLimit	The minimizer exhausted its iteration limit.
calculationFailure	The minimizer threw or returned invalid output.

Definition at line 96 of file clGainOpt.hpp.

Constructor & Destructor Documentation

clGainOpt() [1/2]

template<typename realT>

mx::AO::analysis::clGainOpt< realT >::clGainOpt

(

)

Default c'tor.

Definition at line 467 of file clGainOpt.hpp.

References init().

clGainOpt() [2/2]

template<typename realT>

mx::AO::analysis::clGainOpt< realT >::clGainOpt	(	realT	Ti,
		realT	tau )

C'tor setting the loop timings.

Parameters

[in]	Ti	the desired loop sampling interval.
[in]	tau	the desired loop delay.

Definition at line 473 of file clGainOpt.hpp.

References init(), m_tau, m_Ti, tau(), and Ti().

Member Function Documentation

a() [1/4]

template<typename _realT>

const std::vector< realT > & mx::AO::analysis::clGainOpt< _realT >::a ( )

inline

Get the vector of IIR coefficients.

Definition at line 268 of file clGainOpt.hpp.

References m_a.

a() [2/4]

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::a

(

const Eigen::Array< realT, -1, -1 > &

newA

)

Set the vector of IIR coefficients.

Parameters

[in]

newA

a column-vector Eigen::Array of coefficients, which is copied to m_a.

Definition at line 611 of file clGainOpt.hpp.

References m_a, m_changed, and m_trigCacheChanged.

a() [3/4]

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::a

(

const std::vector< realT > &

newA

)

Set the vector of IIR coefficients.

Parameters

[in]

newA

a vector of coefficients, which is copied to m_a.

Definition at line 599 of file clGainOpt.hpp.

References m_a, m_changed, and m_trigCacheChanged.

Referenced by mx::AO::analysis::clAOLinearPredictor< _realT >::_regularizeCoefficients(), mx::AO::analysis::fourierTemporalPSD< _realT, aosysT >::analyzePSDGrid(), and mx::AO::analysis::clAOLinearPredictor< _realT >::regularizeCoefficients().

a() [4/4]

template<typename _realT>

realT mx::AO::analysis::clGainOpt< _realT >::a ( size_t i )

inline

Get a single IIR coefficient.

Returns

a single IIR coefficient

Definition at line 260 of file clGainOpt.hpp.

References m_a.

b() [1/4]

template<typename _realT>

const std::vector< realT > & mx::AO::analysis::clGainOpt< _realT >::b ( )

inline

Get the vector of FIR coefficients.

Definition at line 239 of file clGainOpt.hpp.

References m_b.

b() [2/4]

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::b

(

const Eigen::Array< realT, -1, -1 > &

newB

)

Set the vector of FIR coefficients.

Parameters

[in]

newB

a column-vector Eigen::Array of coefficients, which is copied to m_b.

Definition at line 570 of file clGainOpt.hpp.

References m_b, m_changed, and m_trigCacheChanged.

b() [3/4]

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::b

(

const std::vector< realT > &

newB

)

Set the vector of FIR coefficients.

Parameters

[in]

newB

a vector of coefficients, which is copied to m_b.

Definition at line 558 of file clGainOpt.hpp.

References m_b, m_changed, and m_trigCacheChanged.

Referenced by mx::AO::analysis::clAOLinearPredictor< _realT >::_regularizeCoefficients(), mx::AO::analysis::fourierTemporalPSD< _realT, aosysT >::analyzePSDGrid(), and mx::AO::analysis::clAOLinearPredictor< _realT >::regularizeCoefficients().

b() [4/4]

template<typename _realT>

realT mx::AO::analysis::clGainOpt< _realT >::b ( size_t i )

inline

Get a single FIR coefficient.

Returns

a single FIR coefficient

Parameters

[in]

i

the index of the FIR coefficient

Definition at line 231 of file clGainOpt.hpp.

References m_b.

clETF()

template<typename realT>

std::complex< realT > mx::AO::analysis::clGainOpt< realT >::clETF	(	int	fi,
		realT	g )

Return the closed loop error transfer function (ETF) at frequency f for gain g.

Returns

the closed loop ETF at f and g.

Parameters

[in]	fi	the index of the frequency at which to calculate the ETF
[in]	g	the loop gain.

Definition at line 877 of file clGainOpt.hpp.

References m_f, and olXfer().

Referenced by mx::AO::analysis::fourierTemporalPSD< _realT, aosysT >::analyzePSDGrid().

clETF2()

template<typename realT>

realT mx::AO::analysis::clGainOpt< realT >::clETF2	(	int	fi,
		realT	g )

Return the norm of the closed loop error transfer function (ETF) at frequency f for gain g.

Returns

the norm of the closed loop ETF at f and g.

Parameters

[in]	fi	the index of the frequency at which to calculate the ETF.
[in]	g	the loop gain.

Definition at line 905 of file clGainOpt.hpp.

References m_f, and olXfer().

Referenced by pseudoOpenLoop().

clETFPhase()

template<typename realT>

realT mx::AO::analysis::clGainOpt< realT >::clETFPhase	(	int	fi,
		realT	g )

Return the closed loop error transfer function (ETF) phase at frequency f for gain g.

Returns

the phase of the closed loop ETF at f and g.

Parameters

[in]	fi	the index of the frequency at which to calculate the ETF
[in]	g	the loop gain.

Definition at line 891 of file clGainOpt.hpp.

References m_f, and olXfer().

clNTF()

template<typename realT>

std::complex< realT > mx::AO::analysis::clGainOpt< realT >::clNTF	(	int	fi,
		realT	g )

Return the closed loop noise transfer function (NTF) at frequency f for gain g.

Returns

the closed loop NTF at f and g.

Parameters

[in]	fi	the index of the frequency at which to calculate the NTF
[in]	g	the loop gain.

Definition at line 919 of file clGainOpt.hpp.

References m_f, and olXfer().

Referenced by mx::AO::analysis::fourierTemporalPSD< _realT, aosysT >::analyzePSDGrid().

clNTF2()

template<typename realT>

realT mx::AO::analysis::clGainOpt< realT >::clNTF2	(	int	fi,
		realT	g )

Return the norm of the closed loop noise transfer function (NTF) at frequency f for gain g.

Returns

the value of the closed loop NTF at f and g.

Parameters

[in]	fi	the index of the frequency at which to calculate the NTF
[in]	g	the loop gain.

Definition at line 937 of file clGainOpt.hpp.

References m_f, and olXfer().

clTF2()

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::clTF2	(	realT &	ETF,
		realT &	NTF,
		int	fi,
		realT	g )

Return the norm of the closed loop transfer functions at frequency f for gain g.

Calculates both the error transfer function (ETF) and the noise transfer function (NTF). This minimizes the various complex number operations, compared to calling both clETF2 and clNTF2.

Parameters

[out]	ETF	is set to the ETF at f and g
[out]	NTF	is set to the NTF at f and g
[in]	fi	is the index of the frequency at which to calculate the ETF
[in]	g	is the loop gain.

Definition at line 957 of file clGainOpt.hpp.

References m_f, and olXfer().

Referenced by mx::AO::analysis::fourierTemporalPSD< _realT, aosysT >::analyzePSDGrid(), and clVariance().

clVariance() [1/2]

template<typename realT>

realT mx::AO::analysis::clGainOpt< realT >::clVariance	(	const std::vector< realT > &	PSDerr,
		const std::vector< realT > &	PSDnoise,
		realT	g )

Calculate the closed loop variance given open-loop PSDs and gain.

Overload of clVariance without the varErr and varNoise output parameters.

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Returns

the total variance (error + noise) in closed loop

Parameters

[in]	PSDerr	the open-loop process error PSD.
[in]	PSDnoise	the open-loop measurement noise PSD.
[in]	g	the gain.

Definition at line 1023 of file clGainOpt.hpp.

References clVariance().

clVariance() [2/2]

template<typename realT>

realT mx::AO::analysis::clGainOpt< realT >::clVariance	(	realT &	varErr,
		realT &	varNoise,
		const std::vector< realT > &	PSDerr,
		const std::vector< realT > &	PSDnoise,
		realT	g )

Calculate the closed loop variance given open-loop PSDs and gain.

Calculates the following quantities.

\[\sigma_{err}^2 = \sum_i \left| ETF(f_i) \right|^2 PSD_{err}(fi) \Delta f\\ \sigma_{noise}^2 = \sum_i \left| NTF(f_i) \right|^2 PSD_{noise}(fi) \Delta f\\ \sigma^2 = \sigma_{err}^2 + \sigma_{noise}^2 \]

\( \sigma^2 \) is returned, and \( \sigma_{err}^2 \) and \( \sigma_{noise}^2 \) are available as the optional arguments varErr and varNoise.

Returns

the total variance (error + noise) in closed loop

Parameters

[out]	varErr	the variance in the residual process error.
[out]	varNoise	the variance in the residual measurement noise.
[in]	PSDerr	the open-loop process error PSD.
[in]	PSDnoise	the open-loop measurement noise PSD.
[in]	g	the gain.

Definition at line 988 of file clGainOpt.hpp.

References clTF2(), and m_f.

Referenced by mx::AO::analysis::fourierTemporalPSD< _realT, aosysT >::analyzePSDGrid(), and clVariance().

f() [1/3]

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::f

(

const std::vector< realT > &

newF

)

Set the vector of frequencies.

Parameters

[in]

newF

a vector containing the new frequencies

Definition at line 697 of file clGainOpt.hpp.

References m_changed, m_f, and m_trigCacheChanged.

f() [2/3]

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::f	(	realT *	newF,
		size_t	nF )

Set the vector of frequencies.

Parameters

[in]	newF	a pointer to an array containing the new frequencies
[in]	nF	the number of elements of size(realT) in newF.

Definition at line 684 of file clGainOpt.hpp.

References m_changed, m_f, and m_trigCacheChanged.

Referenced by mx::AO::analysis::fourierTemporalPSD< _realT, aosysT >::analyzePSDGrid(), and pseudoOpenLoop().

f() [3/3]

template<typename realT>

realT mx::AO::analysis::clGainOpt< realT >::f

(

size_t

i

)

Get the i-th value of frequency.

No range checks are conducted.

Returns

the value of m_f[i]

Parameters

[in]

i

the index of the frequency to return

Definition at line 706 of file clGainOpt.hpp.

References m_f.

f_size()

template<typename _realT>

size_t mx::AO::analysis::clGainOpt< _realT >::f_size ( )

inline

Get the size of the frequency vector.

Returns

m_f.size()

Definition at line 304 of file clGainOpt.hpp.

References m_f.

init()

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::init

(

)

Initialize this instance.

Definition at line 482 of file clGainOpt.hpp.

References m_changed, m_maxFindMin, m_minFindBits, m_minFindMaxFact, m_minFindMaxIter, m_minFindMin, m_N, m_tau, m_Ti, m_trigCacheChanged, and setLeakyIntegrator().

Referenced by clGainOpt(), and clGainOpt().

maxStableGain()

template<typename realT>

mx::error_t mx::AO::analysis::clGainOpt< realT >::maxStableGain	(	realT &	gain,
		maxStableGainReport *	report = nullptr )

Find the maximum stable gain for the loop parameters.

Conducts a search along the Nyquist contour of the open-loop transfer function to find the most-negative crossing of the real axis.

Crossings below m_maxFindMin are ignored.

Returns

error_t::noerror when a crossing is found, error_t::notfound when none is found, or an input error.

Parameters

[out]	gain	maximum stable gain; NaN on failure
[out]	report	optional search diagnostics

Definition at line 1032 of file clGainOpt.hpp.

References crossingFound, mx::AO::analysis::clGainOpt< _realT >::maxStableGainReport::crossingFrequency, mx::AO::analysis::clGainOpt< _realT >::maxStableGainReport::crossingReal, mx::error, mx::AO::analysis::clGainOpt< _realT >::maxStableGainReport::gain, mx::invalidarg, invalidInput, mx::math::isFinite(), mx::AO::analysis::clGainOpt< _realT >::maxStableGainReport::lowerFrequency, mx::AO::analysis::clGainOpt< _realT >::maxStableGainReport::lowerIndex, m_f, m_maxFindMin, noCrossing, mx::noerror, mx::notfound, mx::sizeerr, mx::AO::analysis::clGainOpt< _realT >::maxStableGainReport::status, mx::AO::analysis::clGainOpt< _realT >::maxStableGainReport::upperFrequency, and mx::AO::analysis::clGainOpt< _realT >::maxStableGainReport::upperIndex.

Referenced by mx::AO::analysis::clAOLinearPredictor< _realT >::_regularizeCoefficients(), optGainOpenLoop(), and mx::AO::analysis::clAOLinearPredictor< _realT >::regularizeCoefficients().

N() [1/2]

template<typename realT>

int mx::AO::analysis::clGainOpt< realT >::N

(

)

Get the number of integrations in the (optional) moving average.

Returns

the current value of m_N.

Definition at line 503 of file clGainOpt.hpp.

References m_N.

N() [2/2]

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::N

(

int

newN

)

Set the number of integrations in the moving average.

Parameters

[in]

newN

the value of m_N.

Definition at line 509 of file clGainOpt.hpp.

References m_changed, and m_N.

olXfer() [1/2]

template<typename realT>

std::complex< realT > mx::AO::analysis::clGainOpt< realT >::olXfer

(

int

fi

)

Calculate the open-loop transfer function.

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Returns

the complex value of the open-loop transfer function at f[fi].

Parameters

[in]

fi

the index of the frequency

Definition at line 713 of file clGainOpt.hpp.

References olXfer().

olXfer() [2/2]

template<typename realT>

std::complex< realT > mx::AO::analysis::clGainOpt< realT >::olXfer	(	int	fi,
		complexT &	H_dm,
		complexT &	H_del,
		complexT &	H_con )

Calculate the open-loop transfer function.

Returns

the complex value of the open-loop transfer function at f.

Parameters

[in]	fi	the index of the frequency
[out]	H_dm	the transfer function of the DM
[out]	H_del	the delay transfer function
[out]	H_con	the controller transfer function.

Definition at line 727 of file clGainOpt.hpp.

References m_a, m_b, m_changed, m_f, m_remember, m_tau, m_Ti, m_trigCacheChanged, and mx::math::two_pi().

Referenced by clETF(), clETF2(), clETFPhase(), clNTF(), clNTF2(), clTF2(), and olXfer().

optGainOpenLoop() [1/2]

template<typename realT>

mx::error_t mx::AO::analysis::clGainOpt< realT >::optGainOpenLoop	(	realT &	gain,
		realT &	var,
		const std::vector< realT > &	PSDerr,
		const std::vector< realT > &	PSDnoise,
		bool	gridSearch,
		optGainReport *	report = nullptr )

Return the optimum closed loop gain given an open loop PSD.

Determines the maximum stable gain before minimizing the variance.

Returns

error_t::noerror on convergence or a boundary-limited result, otherwise an explicit failure status.

Parameters

[out]	gain	optimum gain; NaN on failure
[out]	var	variance at the optimum gain; NaN on failure
[in]	PSDerr	open-loop error PSD
[in]	PSDnoise	open-loop measurement-noise PSD
[in]	gridSearch	whether to perform a coarse initial search
[out]	report	optional search diagnostics

Definition at line 1215 of file clGainOpt.hpp.

References maxStableGain(), mx::noerror, optGainOpenLoop(), mx::AO::analysis::clGainOpt< _realT >::optGainReport::stability, stabilityFailure, and mx::AO::analysis::clGainOpt< _realT >::optGainReport::status.

Referenced by mx::AO::analysis::clAOLinearPredictor< _realT >::_regularizeCoefficients(), mx::AO::analysis::fourierTemporalPSD< _realT, aosysT >::analyzePSDGrid(), optGainOpenLoop(), and mx::AO::analysis::clAOLinearPredictor< _realT >::regularizeCoefficients().

optGainOpenLoop() [2/2]

template<typename realT>

mx::error_t mx::AO::analysis::clGainOpt< realT >::optGainOpenLoop	(	realT &	gain,
		realT &	var,
		const std::vector< realT > &	PSDerr,
		const std::vector< realT > &	PSDnoise,
		realT	maximumGain,
		bool	gridSearch,
		optGainReport *	report = nullptr )

Return the optimum closed loop gain given an open loop PSD.

Returns

error_t::noerror on convergence or a boundary-limited result, otherwise an explicit failure status.

Parameters

[out]	gain	optimum gain; best estimate on timeout
[out]	var	variance at the optimum gain
[in]	PSDerr	open-loop error PSD
[in]	PSDnoise	open-loop measurement-noise PSD
[in]	maximumGain	maximum stable gain bounding the search
[in]	gridSearch	whether to perform a coarse initial search
[out]	report	optional search diagnostics

Definition at line 1249 of file clGainOpt.hpp.

References boundaryLimited, calculationFailure, converged, mx::error, mx::AO::analysis::clGainOpt< _realT >::optGainReport::evaluations, mx::AO::analysis::clGainOptOptGain_OL< realT >::evaluations, mx::AO::analysis::clGainOpt< _realT >::optGainReport::gain, mx::AO::analysis::clGainOptOptGain_OL< realT >::go, mx::invalidconfig, invalidInput, mx::math::isFinite(), iterationLimit, mx::AO::analysis::clGainOpt< _realT >::optGainReport::iterations, m_f, m_minFindBits, m_minFindMaxFact, m_minFindMaxIter, m_minFindMin, mx::AO::analysis::clGainOpt< _realT >::optGainReport::maximumEvaluatedGain, mx::AO::analysis::clGainOptOptGain_OL< realT >::maximumEvaluatedGain, mx::AO::analysis::clGainOpt< _realT >::optGainReport::minimumEvaluatedGain, mx::AO::analysis::clGainOptOptGain_OL< realT >::minimumEvaluatedGain, mx::noerror, mx::AO::analysis::clGainOptOptGain_OL< realT >::PSDerr, mx::AO::analysis::clGainOptOptGain_OL< realT >::PSDnoise, mx::AO::analysis::clGainOpt< _realT >::optGainReport::requestedMaximumGain, mx::AO::analysis::clGainOpt< _realT >::optGainReport::searchMaximumGain, mx::AO::analysis::clGainOpt< _realT >::optGainReport::searchMinimumGain, mx::sizeerr, mx::AO::analysis::clGainOpt< _realT >::optGainReport::status, mx::timeout, and mx::AO::analysis::clGainOpt< _realT >::optGainReport::variance.

pseudoOpenLoop()

template<typename realT>

int mx::AO::analysis::clGainOpt< realT >::pseudoOpenLoop	(	std::vector< realT > &	PSD,
		realT	g )

Calculate the pseudo open-loop PSD given a closed loop PSD.

Returns

0 on success

Parameters

	PSD	[in.out] input closed loop PSD, on output contains the pseudo open loop error PSD
[in]	g	the loop gain when PSD was measured.

Definition at line 1374 of file clGainOpt.hpp.

References clETF2(), f(), and m_f.

remember() [1/2]

template<typename realT>

realT mx::AO::analysis::clGainOpt< realT >::remember

(

)

Get the remember factor.

Definition at line 651 of file clGainOpt.hpp.

References m_remember.

remember() [2/2]

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::remember

(

const realT &

rem

)

Set the remember factor for a leaky integrator.

Definition at line 640 of file clGainOpt.hpp.

References m_changed, and m_remember.

Referenced by setLeakyIntegrator().

setLeakyIntegrator()

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::setLeakyIntegrator

(

realT

remember

)

Set the FIR and IIR coefficients so that the control law is a leaky integrator.

Set remember to 1.0 for a pure integrator control law.

Parameters

[in]

remember

a number usually close to 1 setting the amount "remembered" from previous iterations.

Definition at line 657 of file clGainOpt.hpp.

References m_a, m_b, m_changed, m_remember, m_trigCacheChanged, and remember().

Referenced by init().

tau() [1/2]

template<typename realT>

realT mx::AO::analysis::clGainOpt< realT >::tau

(

)

Get the loop delay.

Returns

the current value of m_tau.

Definition at line 540 of file clGainOpt.hpp.

References m_tau.

Referenced by clGainOpt().

tau() [2/2]

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::tau

(

realT

newTau

)

Set the loop delay.

Parameters

[in]

newTau

the new value of m_tau.

Definition at line 546 of file clGainOpt.hpp.

References m_changed, and m_tau.

Ti() [1/2]

template<typename realT>

realT mx::AO::analysis::clGainOpt< realT >::Ti

(

)

Get the loop sampling interval.

Returns

the current value of m_Ti.

Definition at line 521 of file clGainOpt.hpp.

References m_Ti.

Referenced by clGainOpt().

Ti() [2/2]

template<typename realT>

void mx::AO::analysis::clGainOpt< realT >::Ti

(

realT

newTi

)

Set the loop sampling interval.

Parameters

[in]

newTi

the new value of m_Ti.

Definition at line 527 of file clGainOpt.hpp.

References m_changed, m_Ti, and m_trigCacheChanged.

Member Data Documentation

m_a

template<typename _realT>

std::vector<realT> mx::AO::analysis::clGainOpt< _realT >::m_a

protected

Vector of IIR coefficients.

Definition at line 130 of file clGainOpt.hpp.

Referenced by a(), a(), a(), a(), olXfer(), and setLeakyIntegrator().

m_b

template<typename _realT>

std::vector<realT> mx::AO::analysis::clGainOpt< _realT >::m_b

protected

Vector of FIR coefficients.

Definition at line 129 of file clGainOpt.hpp.

Referenced by b(), b(), b(), b(), olXfer(), and setLeakyIntegrator().

m_changed

template<typename _realT>

bool mx::AO::analysis::clGainOpt< _realT >::m_changed { true }

protected

True if any of the members which make up the basic transfer functions are changed.

Definition at line 137 of file clGainOpt.hpp.

Referenced by a(), a(), b(), b(), f(), f(), init(), N(), olXfer(), remember(), setLeakyIntegrator(), tau(), and Ti().

m_f

template<typename _realT>

std::vector<realT> mx::AO::analysis::clGainOpt< _realT >::m_f

protected

Vector of frequencies.

Definition at line 132 of file clGainOpt.hpp.

Referenced by clETF(), clETF2(), clETFPhase(), clNTF(), clNTF2(), clTF2(), clVariance(), f(), f(), f(), f_size(), maxStableGain(), olXfer(), optGainOpenLoop(), and pseudoOpenLoop().

m_maxFindMin

template<typename _realT>

realT mx::AO::analysis::clGainOpt< _realT >::m_maxFindMin

The Minimum value for the maximum stable gain finding algorithm.

Parameters for stability analysis

Definition at line 153 of file clGainOpt.hpp.

Referenced by init(), and maxStableGain().

m_minFindBits

template<typename _realT>

int mx::AO::analysis::clGainOpt< _realT >::m_minFindBits

The bits of precision to use for minimum finding. Defaults to std::numeric_limits<realT>::digits.

Definition at line 163 of file clGainOpt.hpp.

Referenced by init(), and optGainOpenLoop().

m_minFindMaxFact

template<typename _realT>

realT mx::AO::analysis::clGainOpt< _realT >::m_minFindMaxFact

The maximum value, as a multiplicative factor of maximum gain.

Definition at line 162 of file clGainOpt.hpp.

Referenced by init(), and optGainOpenLoop().

m_minFindMaxIter

template<typename _realT>

uintmax_t mx::AO::analysis::clGainOpt< _realT >::m_minFindMaxIter

The maximum iterations allowed for minimization.

Definition at line 165 of file clGainOpt.hpp.

Referenced by init(), and optGainOpenLoop().

m_minFindMin

template<typename _realT>

realT mx::AO::analysis::clGainOpt< _realT >::m_minFindMin

The Minimum value for the minimum finding algorithm.

Parameters for minimization finding

Definition at line 161 of file clGainOpt.hpp.

Referenced by init(), and optGainOpenLoop().

m_N

template<typename _realT>

int mx::AO::analysis::clGainOpt< _realT >::m_N

protected

Number of integrations in the (optional) moving average. Default is 1.

Definition at line 124 of file clGainOpt.hpp.

Referenced by init(), N(), and N().

m_remember

template<typename _realT>

realT mx::AO::analysis::clGainOpt< _realT >::m_remember { 1.0 }

protected

The leaky integrator forget factor.

Definition at line 128 of file clGainOpt.hpp.

Referenced by olXfer(), remember(), remember(), and setLeakyIntegrator().

m_tau

template<typename _realT>

realT mx::AO::analysis::clGainOpt< _realT >::m_tau

protected

The loop delay.

Definition at line 126 of file clGainOpt.hpp.

Referenced by clGainOpt(), init(), olXfer(), tau(), and tau().

m_Ti

template<typename _realT>

realT mx::AO::analysis::clGainOpt< _realT >::m_Ti

protected

The loop sampling interval.

Definition at line 125 of file clGainOpt.hpp.

Referenced by clGainOpt(), init(), olXfer(), Ti(), and Ti().

m_trigCacheChanged

template<typename _realT>

bool mx::AO::analysis::clGainOpt< _realT >::m_trigCacheChanged { true }

protected

True when frequency, sampling interval, or required controller tap count invalidates m_cs and m_ss.

Definition at line 135 of file clGainOpt.hpp.

Referenced by a(), a(), b(), b(), f(), f(), init(), olXfer(), setLeakyIntegrator(), and Ti().

---

The documentation for this struct was generated from the following file:

• ao/analysis/clGainOpt.hpp