REST-for-Physics  v2.3
Rare Event Searches ToolKit for Physics
Loading...
Searching...
No Matches
TRestRawSignalAnalysisProcess Class Reference

Detailed Description

An analysis process to extract valuable information from a TRestRawSignalEvent.

THIS DOCUMENTATION NEEDS REVISION. TODO If possible a figure with the pulse ilustrating the different observables extracted.

The following metadata parameters might be defined at the RML to control the rawsignal analysis is performed:

  • baseLineRange: The bins from the rawdata samples that will be used to calculate the baseline average and fluctuation.
  • baseLineOption: An optional parameter. When set to "ROBUST", the baseline parameters will be calculated using TRestRawSignal's robust methods, instead of mean and standard deviation, the median and IQR sigma are used.
  • integralRange: The calculated observables will only consider points found inside this range.

There are 3 additional parameters that are used in combination to identify the points over threshold at each signal. Those parameters are used, for example, to calculate the number of good signals observable.

  • pointThreshold: The number of sigmas over baseline fluctuations to identify a point overthreshold
  • signalThreshold: A parameter to define a minimum signal fluctuation. Measured in sigmas.
  • pointsOverThreshold: The minimum number of points over threshold to identify a signal as such

Additionaly, there is a metadata parameter,signalsRange that allows to define the signal ids over which this process will have effect. This parameter may allow to define different TRestRawSignalAnalysisProcess instances at the RML operating over different signal ranges. For example:

<TRestRawSignalAnalysisProcess name="rawMM" signalsRange="(0,512)"
baseLineRange="(5,100)" observables="all" />
<TRestRawSignalAnalysisProcess name="rawVETO" signalsRange="(513,1024)" >
<parameter name="pointThreshold" value="5" />
<parameter name="signalThreshold" value="3.5" />
<parameter name="pointsOverThreshold" value="15" />
<observable name="rawAna_max_amplitude_map" type="map<int,double>" />

Observables

Number of signals and base line:

  • NumberOfSignals: Number of pulses recorded in an event.
  • NumberOfGoodSignals: Number of pulses recorded in an event that gets over the threshold. It counts the signal if GetIntegralWithThreshold (fSignal) > 0.
  • BaseLineMean: Average of the base line of the pulses of the event.
  • BaseLineSigmaMean: Average of the standard deviation from the base line in the pulses of the event.

Integrals and energy estimations:

  • FullIntegral: Add the integral of all pulses of the event. The integral of a pulse is the sum of all samples.
  • TripleMaxIntegral: Add the highest value of each pulse with the previous sample and the next, and then add this amount for all pulses in the event. It is an estimation of the deposited energy.
  • ThresholdIntegral: Add the integral of all the pulses in the event that pass the threshold.
  • SlopeIntegral: Add the integral of the rising part of each pulse of the event.
  • RiseSlopeAvg: Add the SlopeIntegral of all pulses in the event that pass the GetThresholdIntegralValue() > 0 condition and divides it by the number of signals that pass the cut.
  • IntegralBalance: (fullIntegral - thrIntegral) / (fullIntegral + thrIntegral) Balance between FullIntegral and ThresholdIntegral.
  • RateOfChangeAvg: RiseSlopeAvg/SlopeIntegral. The first value takes only the events that pass the threshold cut, SopeIntegral takes all the events.
  • xEnergySum: Add the ThresholdIntegral of all signals recorded in X direction.
  • yEnergySum: Add the ThresholdIntegral of all signals recorded in Y direction.

Time observables:

  • TimeBinsLength: MaxTime of the event - MinTime of the event. The functions GetMaxTime() and GetMinTime() take the number of points of the signal (MinTime=0 MaxTime=fSignal[0].GetNumberOfPoints()).
  • RiseTimeAvg: Add GetRiseTime(fSignal) for all signals that satisfy the GetThresholdIntegralValue() > 0 condition and divide it by the number of signals that pass this cut. GetRiseTime(fSignal) provides the number of bins between the fist sample that pass the threshold and the maximum of the peak.

Peak amplitude observables:

  • AmplitudeIntegralRatio: ThresholdIntegral/maxValueIntegral. This is the sum of the integral of all the pulses in the event that pass the threshold divided by the sum of the maximum point of these pulses.
  • MinPeakAmplitude: Minimum value between the maximum points of the pulses that pass the threshold in the event.
  • MaxPeakAmplitude: Maximum value between the maximum points of the pulses that pass the threshold in the event.
  • PeakAmplitudeIntegral: Sum of all maximum points of the pulses that pass the threshold in the event.
  • AmplitudeRatio: PeakAmplitudeIntegral/ MaxPeakAmplitude. Sum of all maximum points divided by the maximum between the maximum points of the pulses.
  • MinEventValue: Minimum value reached by a pulse in the event.

Peak time observables:

  • MaxPeakTime: Highest bin for the maximum point of a pulse that pass the threshold in the event.
  • MinPeakTime: Smallest bin for the maximum point of a pulse that pass the threshold in the event.
  • MaxPeakTimeDelay: MaxPeakTime-MinPeakTime. Time between the earliest peak and the latest peak between the pulses that pass the threshold in the event.
  • AveragePeakTime: For all pulses that pass the threshold, add the bin of their maximums and divide this amount by the number of signals that pass the threshold.

Observables for individual signal info:

  • risetime: Map the ID of each signal in the event with its rise time. GetRiseTime() gives the number of bins between the first sample that pass the threshold and the maximum point of the peak.
  • risetimemean: Add the rise time of all pulses in the event and divide it by the number of pulses.
  • baseline: Map the ID of each signal in the event with its base line. It computes the base line adding the samples in a certain range and dividing it by the number of samples.
  • baselinemean: Add the base line of all pulses in the event and divide it by the number of pulses.
  • baselinesigma: Map the ID of each signal in the event with its base line's standard deviation. Standard deviation computed as the squared root of the sum of the squared differences between the samples and the base line, divided by the number of samples in the range used to compute the base line.
  • baselinesigmamean: Add the base line's standard deviation of all pulses in the event and divide it by the number of pulses.
  • max_amplitude_map: Map the ID of each signal in the event with the amplitude of its peak. Amplitude = MaxPeakValue-BaseLine.
  • ampeve_maxmethod: Add the amplitude of all pulses in the event. Amplitude = MaxPeakValue-BaseLine.
  • thr_integral_map: Map the ID of each signal in the event with the threshold integral of its peak. GetIntegralWithThreshold() adds the samples of a pulse that get over the threshold. A certain number of samples must pass the threshold to be taken into account.
  • ampeve_intmethod: Add the threshold integral of all pulses in the event. GetIntegralWithThreshold () adds the samples of a pulse that get over the threshold. A certain number of samples must pass the threshold to be taken into account.

You may add filters to any observable inside the analysis tree. To add a cut, write "cut" sections in your rml file:

<cut name="MeanBaseLineCut" value="(0,4096)" />

Warning
⚠ REST is under continous development. This documentation is offered to you by the REST community. Your HELP is needed to keep this code up to date. Your feedback will be worth to support this software, please report any problems/suggestions you may find will using it at The REST Framework forum. You are welcome to contribute fixing typos, updating information or adding/proposing new contributions. See also our Contribution Guide.

RESTsoft - Software for Rare Event Searches with TPCs

History of developments:

2017-February: First implementation of raw signal analysis process into REST_v2. Created from TRestDetectorSignalAnalysisProcess

Author
Javier Galan

Definition at line 31 of file TRestRawSignalAnalysisProcess.h.

#include <TRestRawSignalAnalysisProcess.h>

Inheritance diagram for TRestRawSignalAnalysisProcess:
TRestEventProcess TRestMetadata

Public Member Functions

 ClassDefOverride (TRestRawSignalAnalysisProcess, 5)
RESTValue GetInputEvent () const override
 Get pointer to input event. Must be implemented in the derived class.
RESTValue GetOutputEvent () const override
 Get pointer to output event. Must be implemented in the derived class.
const char * GetProcessName () const override
void InitFromConfigFile () override
 To make settings from rml file. This method must be implemented in the derived class.
void InitProcess () override
 Process initialization.
void PrintMetadata () override
 Implemented it in the derived metadata class to print out specific metadata information.
TRestEvent * ProcessEvent (TRestEvent *inputEvent) override
 The main processing event function.
 TRestRawSignalAnalysisProcess ()
 Default constructor.
 ~TRestRawSignalAnalysisProcess ()
 Default destructor.
Public Member Functions inherited from TRestEventProcess
virtual Bool_t AddInputFile (const std::string &file)
bool ApplyCut ()
 Apply cut according to the cut conditions saved in fCut.
virtual void BeginOfEventProcess (TRestEvent *inputEvent=nullptr)
 Begin of event process, preparation work. Called right before ProcessEvent().
 ClassDefOverride (TRestEventProcess, 3)
virtual void EndOfEventProcess (TRestEvent *inputEvent=nullptr)
 End of event process. Nothing to do. Called directly after ProcessEvent().
virtual void EndProcess ()
 To be executed at the end of the run (outside event loop).
TRestAnalysisTree * GetAnalysisTree () const
 Return the local analysis tree (dummy).
TCanvas * GetCanvas () const
 Get canvas.
TRestAnalysisTree * GetFullAnalysisTree ()
 Get the full analysis tree from TRestRun, with non-empty entries.
std::vector< std::string > GetListOfAddedObservables ()
 Get list of observables, convert map to vector.
TRestRun * GetRunInfo () const
 Return the pointer of the hosting TRestRun object.
virtual Long64_t GetTotalBytes () const
virtual Long64_t GetTotalBytesRead () const
 Interface to external file reading, get the read bytes. To be implemented in external processes.
Bool_t isExternal () const
 Return whether this process is external process.
Int_t LoadSectionMetadata () override
 Load extra section metadata: outputlevel after calling TRestMetadata::LoadSectionMetadata().
virtual void NotifyAnalysisTreeReset ()
 In case the analysis tree is reset(switched to new file), some process needs to have action.
virtual Bool_t OpenInputFiles (const std::vector< std::string > &files)
 Interface to external file reading, open input file. To be implemented in external processes.
std::vector< std::string > ReadObservables ()
void RegisterAllObservables (Bool_t value=true)
virtual Bool_t ResetEntry ()
void SetAnalysisTree (TRestAnalysisTree *tree)
 Set analysis tree of this process, then add observables to it.
void SetCanvasSize (Int_t x, Int_t y)
 Set canvas size.
void SetFriendProcess (TRestEventProcess *p)
 Add friendly process to this process.
void SetObservableValidation (bool validate)
void SetParallelProcess (TRestEventProcess *p)
 Add parallel process to this process.
void SetRunInfo (TRestRun *r)
 Set TRestRun for this process.
Bool_t singleThreadOnly () const
 Return whether this process is single std::thread only.
 TRestEventProcess ()
 ~TRestEventProcess ()

Private Member Functions

void Initialize () override
 Function to initialize input/output event members and define the section name.

Private Attributes

std::string fBaseLineOption
 Option for calculation of baseline parameters, can be set to "ROBUST".
TVector2 fBaseLineRange = TVector2(5, 55)
 The range where the baseline range will be calculated.
std::string fChannelType
 avoid error when loading parameters (not used)
std::set< std::string > fChannelTypes = {}
TVector2 fIntegralRange = TVector2(10, 500)
 The range where the observables will be calculated.
Int_t fPointsOverThreshold = 5
 The minimum number of points over threshold to identify a signal as such.
Double_t fPointThreshold = 2
 The number of sigmas over baseline fluctuations to identify a point over threshold.
Bool_t fRangeEnabled = false
 Just a flag to quickly determine if we have to apply the range filter.
TRestRawReadoutMetadata * fReadoutMetadata = nullptr
TRestRawSignalEvent * fSignalEvent
 A pointer to the specific TRestRawSignalEvent input.
TVector2 fSignalsRange = TVector2(-1, -1)
 It defines the signals id range where analysis is applied.
Double_t fSignalThreshold = 5
 A parameter to define a minimum signal fluctuation. Measured in sigmas.

Additional Inherited Members

Protected Types inherited from TRestEventProcess
enum  REST_Process_Output { No_Output , Observable , Internal_Var , Full_Output }
Protected Member Functions inherited from TRestEventProcess
void BeginPrintProcess ()
 [name, cut range]
void CreateCanvas ()
 Create the canvas.
void EndPrintProcess ()
 Pre-defined printer, can be used at the ending in the implementation of PrintMetadata().
TRestEventProcess * GetFriend (const std::string &nameOrType)
 Get friendly TRestEventProcess object.
TRestEventProcess * GetFriendLive (const std::string &nameOrType)
 Get friendly TRestEventProcess object.
template<class T>
T * GetMetadata ()
 Get a metadata object from the host TRestRun.
TRestMetadata * GetMetadata (const std::string &nameOrType)
 Get a metadata object from the host TRestRun.
size_t GetNumberOfParallelProcesses () const
template<class T>
T GetObservableValue (const std::string &name)
TRestEventProcess * GetParallel (int i)
 Get a list of parallel processes from this process.
template<class T>
std::vector< T > GetParallelDataMembers (T *member_of_process)
 Get a list of data members from parallel processes which is same to this process's certain data member.
template<class T>
void SetObservableValue (const std::string &name, const T &value)
 Set observable value for AnalysisTree.
Protected Attributes inherited from TRestEventProcess
TRestAnalysisTree * fAnalysisTree = nullptr
TCanvas * fCanvas = nullptr
 < Canvas for some viewer event
TVector2 fCanvasSize
 Canvas size.
std::vector< std::pair< std::string, TVector2 > > fCuts
 Stores cut definitions. Any listed observables should be in the range.
bool fDynamicObs = false
 It defines whether to use added observables only or all the observables appear in the code.
bool fIsExternal = false
 It defines if the process reads event data from an external source.
std::map< std::string, int > fObservablesDefined
 Stores the list of all the appeared process observables in the code.
std::map< std::string, int > fObservablesUpdated
 Stores the list of process observables updated when processing this event.
bool fReadOnly = false
 not used, keep for compatibility
TRestRun * fRunInfo = nullptr
 < Pointer to TRestRun object where to find metadata.
bool fSingleThreadOnly = false
bool fValidateObservables = false
 It defines if observable names should be added to the validation list.

Constructor & Destructor Documentation

◆ TRestRawSignalAnalysisProcess()

TRestRawSignalAnalysisProcess::TRestRawSignalAnalysisProcess ( )

Default constructor.

Definition at line 259 of file TRestRawSignalAnalysisProcess.cxx.

◆ ~TRestRawSignalAnalysisProcess()

TRestRawSignalAnalysisProcess::~TRestRawSignalAnalysisProcess ( )

Default destructor.

Definition at line 264 of file TRestRawSignalAnalysisProcess.cxx.

Member Function Documentation

◆ GetInputEvent()

RESTValue TRestRawSignalAnalysisProcess::GetInputEvent ( ) const
inlineoverridevirtual

Get pointer to input event. Must be implemented in the derived class.

Implements TRestEventProcess.

Definition at line 73 of file TRestRawSignalAnalysisProcess.h.

◆ GetOutputEvent()

RESTValue TRestRawSignalAnalysisProcess::GetOutputEvent ( ) const
inlineoverridevirtual

Get pointer to output event. Must be implemented in the derived class.

Implements TRestEventProcess.

Definition at line 75 of file TRestRawSignalAnalysisProcess.h.

◆ GetProcessName()

const char * TRestRawSignalAnalysisProcess::GetProcessName ( ) const
inlineoverridevirtual

Implements TRestEventProcess.

Definition at line 99 of file TRestRawSignalAnalysisProcess.h.

◆ InitFromConfigFile()

void TRestRawSignalAnalysisProcess::InitFromConfigFile ( )
overridevirtual

To make settings from rml file. This method must be implemented in the derived class.

Reimplemented from TRestEventProcess.

Definition at line 285 of file TRestRawSignalAnalysisProcess.cxx.

◆ Initialize()

void TRestRawSignalAnalysisProcess::Initialize ( )
overrideprivatevirtual

Function to initialize input/output event members and define the section name.

Reimplemented from TRestMetadata.

Definition at line 270 of file TRestRawSignalAnalysisProcess.cxx.

◆ InitProcess()

void TRestRawSignalAnalysisProcess::InitProcess ( )
overridevirtual

Process initialization.

Reimplemented from TRestEventProcess.

Definition at line 279 of file TRestRawSignalAnalysisProcess.cxx.

◆ PrintMetadata()

void TRestRawSignalAnalysisProcess::PrintMetadata ( )
inlineoverridevirtual

Implemented it in the derived metadata class to print out specific metadata information.

Prints metadata content on screen. Usually overloaded by the derived metadata class.

Reimplemented from TRestMetadata.

Definition at line 83 of file TRestRawSignalAnalysisProcess.h.

◆ ProcessEvent()

TRestEvent * TRestRawSignalAnalysisProcess::ProcessEvent ( TRestEvent * inputEvent)
overridevirtual

The main processing event function.

We define (or re-define) the baseline range and calculation range of our raw-signals.

Important call we need to initialize the points over threshold in a TRestRawSignal

Implements TRestEventProcess.

Definition at line 296 of file TRestRawSignalAnalysisProcess.cxx.

Field Documentation

◆ fBaseLineOption

std::string TRestRawSignalAnalysisProcess::fBaseLineOption
private

Option for calculation of baseline parameters, can be set to "ROBUST".

Definition at line 46 of file TRestRawSignalAnalysisProcess.h.

◆ fBaseLineRange

TVector2 TRestRawSignalAnalysisProcess::fBaseLineRange = TVector2(5, 55)
private

The range where the baseline range will be calculated.

Definition at line 40 of file TRestRawSignalAnalysisProcess.h.

◆ fChannelType

std::string TRestRawSignalAnalysisProcess::fChannelType
private

avoid error when loading parameters (not used)

Definition at line 55 of file TRestRawSignalAnalysisProcess.h.

◆ fChannelTypes

std::set<std::string> TRestRawSignalAnalysisProcess::fChannelTypes = {}
private

Definition at line 67 of file TRestRawSignalAnalysisProcess.h.

◆ fIntegralRange

TVector2 TRestRawSignalAnalysisProcess::fIntegralRange = TVector2(10, 500)
private

The range where the observables will be calculated.

Definition at line 43 of file TRestRawSignalAnalysisProcess.h.

◆ fPointsOverThreshold

Int_t TRestRawSignalAnalysisProcess::fPointsOverThreshold = 5
private

The minimum number of points over threshold to identify a signal as such.

Definition at line 58 of file TRestRawSignalAnalysisProcess.h.

◆ fPointThreshold

Double_t TRestRawSignalAnalysisProcess::fPointThreshold = 2
private

The number of sigmas over baseline fluctuations to identify a point over threshold.

Definition at line 49 of file TRestRawSignalAnalysisProcess.h.

◆ fRangeEnabled

Bool_t TRestRawSignalAnalysisProcess::fRangeEnabled = false
private

Just a flag to quickly determine if we have to apply the range filter.

Definition at line 37 of file TRestRawSignalAnalysisProcess.h.

◆ fReadoutMetadata

TRestRawReadoutMetadata* TRestRawSignalAnalysisProcess::fReadoutMetadata = nullptr
private

Definition at line 63 of file TRestRawSignalAnalysisProcess.h.

◆ fSignalEvent

TRestRawSignalEvent* TRestRawSignalAnalysisProcess::fSignalEvent
private

A pointer to the specific TRestRawSignalEvent input.

Definition at line 34 of file TRestRawSignalAnalysisProcess.h.

◆ fSignalsRange

TVector2 TRestRawSignalAnalysisProcess::fSignalsRange = TVector2(-1, -1)
private

It defines the signals id range where analysis is applied.

Definition at line 61 of file TRestRawSignalAnalysisProcess.h.

◆ fSignalThreshold

Double_t TRestRawSignalAnalysisProcess::fSignalThreshold = 5
private

A parameter to define a minimum signal fluctuation. Measured in sigmas.

Definition at line 52 of file TRestRawSignalAnalysisProcess.h.


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