1 | $Id: README,v 1.1 2006/11/22 14:51:26 gcosmo Exp $ |
---|
2 | ------------------------------------------------------------------- |
---|
3 | |
---|
4 | ========================================================= |
---|
5 | Geant4 - an Object-Oriented Toolkit for Simulation in HEP |
---|
6 | ========================================================= |
---|
7 | |
---|
8 | MCTRUTH using HepMC |
---|
9 | ------------------- |
---|
10 | |
---|
11 | This example demonstrates a mechanism for Monte Carlo truth handling |
---|
12 | using HepMC as the event record. The user does not interact directly |
---|
13 | with the HepMC classes but with the MCTruthManager class which takes |
---|
14 | care with storing all the necessary information about particles, |
---|
15 | vertices and relations between them. A specialized tracking action is |
---|
16 | used to test whether given particle is to be stored or not. The |
---|
17 | decision criteria for storing particle are configurable via the |
---|
18 | MCTruthConfig class. |
---|
19 | |
---|
20 | HOW TO BUILD THE EXAMPLE ? |
---|
21 | |
---|
22 | - install HepMC event record (version 1.27) |
---|
23 | |
---|
24 | - set HEPMC_DIR variable to point to the directory where HepMC is installed; |
---|
25 | add the path "$HEPMC_DIR/lib" to your LD_LIBRARY_PATH variable. |
---|
26 | |
---|
27 | - compile and link to generate the executable: |
---|
28 | % gmake |
---|
29 | |
---|
30 | - execute the application: |
---|
31 | % mctruthex |
---|
32 | |
---|
33 | DESCRIPTION OF THE MCTRUTH HANDLING MECHANISM |
---|
34 | |
---|
35 | The main element of the MC truth handling machinery is the |
---|
36 | MCTruthManager class. This class is responsible for all the |
---|
37 | interaction with the HepMC event and does not depend on Geant4. It is |
---|
38 | a singleton, therefore it is guaranteed to be instanciated only once |
---|
39 | and the static 'GetInstance' method allows to access it from anywhere |
---|
40 | in the code. It contains methods like 'NewEvent' to start a new event, |
---|
41 | 'AddParticle' to add particle to the current event, as well as |
---|
42 | 'PrintEvent' for the purpose of the debugging. The core of the |
---|
43 | algorithm which deals with building up the MC truth event tree within |
---|
44 | the HepMC event is implemented in AddParticle method. |
---|
45 | |
---|
46 | The AddParticle method is called with the following arguments: |
---|
47 | four-momentum, production position and 'end' position of the particle, |
---|
48 | PDG code of the particle, as well as the particle ID (unique identifier, |
---|
49 | as we will see later, corresponding to Geant4 TrackID) and the ID of |
---|
50 | the mother. Finally, there is a boolean flag specifying whether the |
---|
51 | direct mother of the given particle has been stored, or not. |
---|
52 | |
---|
53 | The first step, which always takes place, is to instanciate a new |
---|
54 | HepMC::GenParticle with the barcode corresponding to particle ID, as |
---|
55 | well as to instanciate a new HepMC::GenVertex which will represent the |
---|
56 | 'end' vertex of the particle. The barcode of the 'end vertex' is equal |
---|
57 | to minus the barcode of the particle. |
---|
58 | |
---|
59 | We can now distinguish several cases: |
---|
60 | |
---|
61 | 1) the particle is a primary in the Geant4 language, i.e. its |
---|
62 | mother ID is 0 |
---|
63 | |
---|
64 | This is the simplest case, we just instanciate a new 'primary' |
---|
65 | (without any incoming particles) GenVertex, we add to it the |
---|
66 | particle and we put it all in the event. Additionally we store the |
---|
67 | ID of the particle in a special vector, where all the IDs of |
---|
68 | primary particles will be stored, allowing quick access to each of |
---|
69 | the main 'branches' of the event. We return from the method. |
---|
70 | |
---|
71 | 2) the particle is not a primary |
---|
72 | |
---|
73 | We use the 'event->barcode_to_particle(motherID)' method to get the |
---|
74 | pointer to its mother. |
---|
75 | |
---|
76 | We check if the 'end vertex' of the mother corresponds to the |
---|
77 | 'production vertex' of the particle in question. |
---|
78 | |
---|
79 | 2.1) If the two vertices do match, we attach the new particle to |
---|
80 | the 'end vertex' of the mother. We return from the method. |
---|
81 | |
---|
82 | 2.2) If the two vertices do not match, i.e. the new particle is not |
---|
83 | a product of the 'end vertex' of the mother particle, we can |
---|
84 | have two cases: |
---|
85 | |
---|
86 | 2.2.1) The boolean flag says that the direct mother of the |
---|
87 | particle has _not_ been stored. This means that the |
---|
88 | particle has been 'adopted' by one of its ancestors, or |
---|
89 | in other words, the mother ID of the particle does not |
---|
90 | correspond to its direct mother (so clearly the |
---|
91 | vertices cannot match). This for instance could happen |
---|
92 | if we decided not to store gamma coming from pi0 decay |
---|
93 | but did decide to store e+/- coming from the gamma |
---|
94 | conversion (so the gamma between pi0 and e+/- was |
---|
95 | missing). In such a case we instanciate (or use one of |
---|
96 | the existing ones, if vertices match) a 'dummy' |
---|
97 | particle (with pdg = -999999) which then acts as the |
---|
98 | link between the 'adopted' particle and the |
---|
99 | (non-direct) mother. In such a way, the navigability up |
---|
100 | in the event is still possible, but in the same time, |
---|
101 | we can clearly see that the link is not a direct |
---|
102 | one. We return from the method. |
---|
103 | |
---|
104 | 2.2.2) The boolean flag says that direct mother of the |
---|
105 | particle _has_ been stored. Taking into account that |
---|
106 | the vertices do not match, it can mean only one |
---|
107 | thing. The new particle has been produced 'on the |
---|
108 | flight', i.e. somewhere 'before' the 'end vertex' of |
---|
109 | the mother. This can be the case, for instace, for |
---|
110 | delta electrons, bremsstrahlung gammas, etc. In such a |
---|
111 | situation, we 'split' the mother particle in two |
---|
112 | particles and create a new vertex from which the |
---|
113 | secondary will be going out. The complication, however, |
---|
114 | arises when we have more than one generated 'on the |
---|
115 | flight' particle attached to the same mother. In such a |
---|
116 | case, for each secondary we need to locate the right |
---|
117 | 'segment' of the mother particle (i.e. we need to find |
---|
118 | between which two vertices we need to add a new |
---|
119 | one). To keep track of those segmentations we introduce |
---|
120 | a map where each particle ID we map into the number of |
---|
121 | existing segments (in the normal case one). Each new |
---|
122 | 'segment' gets barcode equal to the barcode of the |
---|
123 | original particle + N*10000000, where N is the segment |
---|
124 | number. In such a way, one can easily follow the |
---|
125 | 'segmentation' (if any) of each particle. We return |
---|
126 | from the method. |
---|
127 | |
---|
128 | This concludes the description of MCTruthManager. The MCTruthConfig |
---|
129 | class is a collection of criteria (minimal energy, PDG, creator |
---|
130 | process, etc) that we want to apply when deciding whether to store or |
---|
131 | not given particle. These values are used by the |
---|
132 | 'MCTruthTrackingAction' which we describe below. This class can |
---|
133 | certainly be extended with other members. |
---|
134 | |
---|
135 | The actual Geant4-dependent part of the MCTruth handling machinery |
---|
136 | consists of a few 'G4 user actions' as well as an implementation of |
---|
137 | G4VUserTrackInformation. The later one is, for the moment, used only |
---|
138 | to store one boolean flag indicating whether the direct mother of the |
---|
139 | given track has been stored or not. |
---|
140 | |
---|
141 | The first user action is MCTruthEventAction which is only reponsible |
---|
142 | for calling MCTruthManager::GetInstance()->NewEvent() at the beginning |
---|
143 | of each event. It can also be used for printing out events for the |
---|
144 | purpose of debugging. |
---|
145 | |
---|
146 | The actual 'decision making' concerning which particle to store is |
---|
147 | done in MCTruthTrackingAction. At the end of each track the method |
---|
148 | trackToBeStored(track) is called to check for various characteristics |
---|
149 | of the particle. These, for instance can be energy, particle ID, |
---|
150 | creator process, etc. |
---|
151 | |
---|
152 | If the particle satisfies the conditions the |
---|
153 | MCTruthManager::GetInstance()->AddParticle is called and all the |
---|
154 | procedure described above is performed. The important element here is |
---|
155 | that the Geant4 TrackID is used as the unique particle ID in |
---|
156 | MCTruthManager and eventually as the barcode of the |
---|
157 | HepMC::GenParticle. |
---|
158 | |
---|
159 | If the particle does not qualify to be stored, there are two actions |
---|
160 | performed. First the 'ParentID' of the _daughters_ is set to the |
---|
161 | 'ParentID' of the currenly processed particle. In other words, the |
---|
162 | 'ParentID' of the daughters is set to the ID of the last stored |
---|
163 | particle. Second, the 'directParent' flag from MCTruthTrackInformation |
---|
164 | of the __daughters__ is set to FALSE. In such a way, one is still able |
---|
165 | to navigate up in the event (to get the ancestors of the particle), |
---|
166 | but in the same time, the particle is flagged as 'not having direct |
---|
167 | parent'. |
---|