Core unit for mash-based photon transport using ray tracing algorithms. More...

#include <string.h>
#include <stdlib.h>
#include <math.h>
#include "mmc_raytrace.h"

Include dependency graph for mmc_raytrace.c:

Macros
#define	F32N(a) ((a) & 0x80000000)

#define	F32P(a) ((a) ^ 0x80000000)

Functions
void	interppos (float3 w, float3 p1, float3 p2, float3 p3, float3 *pout)
	function to linearly interpolate between 3 3D points (p1,p2,p3) using weight (w) More...

void	getinterp (float w1, float w2, float w3, float3 p1, float3 p2, float3 p3, float3 pout)
	function to linearly interpolate between 3 3D points (p1,p2,p3) using scalar weight (w) More...

void	fixphoton (float3 p, float3 nodes, int *ee)
	Function to deal with ray-edge/ray-vertex intersections. More...

float	plucker_raytet (ray r, raytracer tracer, mcconfig cfg, visitor visit)
	Plucker-coordinate based ray-tracer to advance photon by one step. More...

float	havel_raytet (ray r, raytracer tracer, mcconfig cfg, visitor visit)

float	badouel_raytet (ray r, raytracer tracer, mcconfig cfg, visitor visit)

float	branchless_badouel_raytet (ray r, raytracer tracer, mcconfig cfg, visitor visit)

void	onephoton (size_t id, raytracer tracer, tetmesh mesh, mcconfig cfg, RandType ran, RandType ran0, visitor visit)
	The core Monte Carlo function simulating a single photon (!!!Important!!!) More...

float	reflectrayroi (mcconfig cfg, float3 c0, float3 ph, raytracer tracer, int eid, int inroi, RandType *ran, int roitype, int roiidx, int refeid)
	Calculate the reflection/transmission of a ray at the ROI surface. More...

float	reflectray (mcconfig cfg, float3 c0, raytracer tracer, int oldeid, int eid, int faceid, RandType ran, int inroi)
	Calculate the reflection/transmission of a ray at an interface. More...

void	launchphoton (mcconfig cfg, ray r, tetmesh mesh, RandType ran, RandType *ran0)
	Launch a new photon. More...

void	albedoweight (ray r, tetmesh mesh, mcconfig cfg, visitor visit)
	Deal with absorption using MCML-like algorithm (albedo-weight MC) More...

void	visitor_init (mcconfig cfg, visitor visit)

void	visitor_clear (visitor *visit)

void	updateroi (int immctype, ray r, tetmesh mesh)

Variables
const int	e2n [6][2] = {{0, 1}, {0, 2}, {0, 3}, {1, 2}, {1, 3}, {2, 3}}

const int	fc [4][3] = {{0, 4, 2}, {3, 5, 4}, {2, 5, 1}, {1, 3, 0}}
	Tetrahedron face edge indices. More...

const int	nc [4][3] = {{3, 0, 1}, {3, 1, 2}, {2, 0, 3}, {1, 0, 2}}
	Tetrahedron faces, in clock-wise orders, represented using local node indices. More...

const int	out [4][3]
	Tetrahedron faces, in counter-clock-wise orders, represented using local node indices. More...

const int	facemap [4]
	The local index of the node with an opposite face to the i-th face defined in nc[][]. More...

const int	ifacemap [4]
	Inverse mapping between the local index of the node and the corresponding opposite face in nc[]'s order. More...

const int	faceorder [5]
	Index mapping from the i-th face-neighbors (facenb) to the face defined in nc[][]. More...

const int	ifaceorder [4]
	Index mapping from the i-th face defined in nc[][] to the face-neighbor (facenb) face orders. More...

const char	maskmap [16] = {4, 0, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3}
	A mapping from SSE4 mask output to the face index.

Detailed Description

Core unit for mash-based photon transport using ray tracing algorithms.

Macro Definition Documentation

◆ F32N

#define F32N ( a ) ((a) & 0x80000000)

< Macro to enable SSE4 based ray-tracers Macro to test if a floating point is negative

◆ F32P

#define F32P ( a ) ((a) ^ 0x80000000)

Macro to test if a floating point is positive

Function Documentation

◆ albedoweight()

void albedoweight	(	ray *	r,
		tetmesh *	mesh,
		mcconfig *	cfg,
		visitor *	visit
	)

Deal with absorption using MCML-like algorithm (albedo-weight MC)

This function performs MCML-like absorption calculations

Parameters

[in,out]	r	the current ray
[in]	mesh	the mesh data structure
[in,out]	cfg	simulation configuration structure
[out]	visit	statistics counters of this thread

◆ fixphoton()

void fixphoton	(	float3 *	p,
		float3 *	nodes,
		int *	ee
	)

Function to deal with ray-edge/ray-vertex intersections.

when a photon is crossing a vertex or edge, (slightly) pull the photon toward the center of the element and try again

Parameters

[in,out]	p	current photon position
[in]	nodes	pointer to the 4 nodes of the tet
[in]	ee	indices of the 4 nodes ee=elem[eid]

◆ getinterp()

void getinterp	(	float	w1,
		float	w2,
		float	w3,
		float3 *	p1,
		float3 *	p2,
		float3 *	p3,
		float3 *	pout
	)

inline

function to linearly interpolate between 3 3D points (p1,p2,p3) using scalar weight (w)

pout=p1*w1+p2*w2+p3*w3

Parameters

[in]	w1	weight for p1
[in]	w2	weight for p2
[in]	w3	weight for p3
[in]	p1	1st point
[in]	p2	2nd point
[in]	p3	3rd point
[out]	pout	output 3D position

◆ havel_raytet()

float havel_raytet	(	ray *	r,
		raytracer *	tracer,
		mcconfig *	cfg,
		visitor *	visit
	)

Havel, Badouel and Branch-less Badouel-based SSE4 ray-tracer require to compile with SSE4 only. Give an error if not compiled with SSE.

◆ interppos()

void interppos	(	float3 *	w,
		float3 *	p1,
		float3 *	p2,
		float3 *	p3,
		float3 *	pout
	)

inline

function to linearly interpolate between 3 3D points (p1,p2,p3) using weight (w)

pout=p1*w1+p2*w2+p3*w3

Parameters

[in]	w	weight vector
[in]	p1	1st point
[in]	p2	2nd point
[in]	p3	3rd point
[out]	pout	output 3D position

◆ launchphoton()

void launchphoton	(	mcconfig *	cfg,
		ray *	r,
		tetmesh *	mesh,
		RandType *	ran,
		RandType *	ran0
	)

Launch a new photon.

This function launch a new photon using one of the dozen supported source forms.

Parameters

[in,out]	cfg	simulation configuration structure
[in,out]	r	the current ray
[in]	mesh	the mesh data structure
[in,out]	ran	the random number generator states
[in,out]	ran0	the additional random number generator states

Here is the call graph for this function:

◆ onephoton()

void onephoton	(	size_t	id,
		raytracer *	tracer,
		tetmesh *	mesh,
		mcconfig *	cfg,
		RandType *	ran,
		RandType *	ran0,
		visitor *	visit
	)

The core Monte Carlo function simulating a single photon (!!!Important!!!)

This is the core Monte Carlo simulation function. It simulates the life-time of a single photon packet, from launching to termination.

Parameters

[in]	id	the linear index of the current photon, starting from 0.
[in]	tracer	the ray-tracer aux data structure
[in]	mesh	the mesh data structure
[in,out]	ran	the random number generator states
[in,out]	ran0	the additional random number generator states
[in,out]	cfg	simulation configuration structure
[out]	visit	statistics counters of this thread

◆ plucker_raytet()

float plucker_raytet	(	ray *	r,
		raytracer *	tracer,
		mcconfig *	cfg,
		visitor *	visit
	)

Plucker-coordinate based ray-tracer to advance photon by one step.

this function uses Plucker-coordinate based ray-triangle intersection tests to advance photon by one step, see Fang2010.

Parameters

[in,out]	r	the current ray
[in]	tracer	the ray-tracer aux data structure
[in]	cfg	simulation configuration structure
[out]	visit	statistics counters of this thread

◆ reflectray()

float reflectray	(	mcconfig *	cfg,
		float3 *	c0,
		raytracer *	tracer,
		int *	oldeid,
		int *	eid,
		int	faceid,
		RandType *	ran,
		int	inroi
	)

Calculate the reflection/transmission of a ray at an interface.

This function handles the reflection and transmission events at an interface where the refractive indices mismatch.

Parameters

[in,out]	cfg	simulation configuration structure
[in]	c0	the current direction vector of the ray
[in]	tracer	the ray-tracer aux data structure
[in]	oldeid	the index of the element the photon moves away from
[in]	eid	the index of the element the photon about to move into
[in]	faceid	index of the face through which the photon reflects/transmits
[in,out]	ran	the random number generator states

◆ reflectrayroi()

float reflectrayroi	(	mcconfig *	cfg,
		float3 *	c0,
		float3 *	ph,
		raytracer *	tracer,
		int *	eid,
		int *	inroi,
		RandType *	ran,
		int	roitype,
		int	roiidx,
		int	refeid
	)

Calculate the reflection/transmission of a ray at the ROI surface.

This function handles the reflection and transmission events at the roi surface wall where the refractive indices mismatch.

Parameters

[in]	c0	the current direction vector of the ray
[in]	u	the current direction vector of the edge
[in]	ph	hitting position at the edgeroi
[in]	E0	any point on the edge
[in]	tracer	the ray-tracer aux data structure
[in]	eid	the index of the CURRENT element
[in,out]	ran	the random number generator states

Variable Documentation

◆ e2n

const int e2n[6][2] = {{0, 1}, {0, 2}, {0, 3}, {1, 2}, {1, 3}, {2, 3}}

from edge index to node index

◆ facemap

const int facemap[4]

The local index of the node with an opposite face to the i-th face defined in nc[][].

nc[i] <-> node[facemap[i]] the 1st face of this tet, i.e. nc[0]={3,0,1}, is opposite to the 3rd node the 2nd face of this tet, i.e. nc[1]={3,1,2}, is opposite to the 1st node etc.

◆ faceorder

const int faceorder[5]

Index mapping from the i-th face-neighbors (facenb) to the face defined in nc[][].

facenb[i] <-> nc[faceorder[i]] the 1st tet neighbor shares the 2nd face of this tet, i.e. nc[1]={3,1,2} the 2nd tet neighbor shares the 4th face of this tet, i.e. nc[3]={1,0,2} etc.

◆ fc

const int fc[4][3] = {{0, 4, 2}, {3, 5, 4}, {2, 5, 1}, {1, 3, 0}}

Tetrahedron face edge indices.

fc[4] points to the 4 facets of a tetrahedron, with each triangular face made of 3 directed edges. The numbers [0-5] are the local indices of the edge, defined by 0:[0->1], 1:[0->2], 2: [0->3], 3:[1->2], 4:[1->3], 5:[2->3] where the pair in [] are the local nodes forming the edge

◆ ifacemap

const int ifacemap[4]

Inverse mapping between the local index of the node and the corresponding opposite face in nc[]'s order.

nc[ifacemap[i]] <-> node[i] the 1st node of this tet is in opposite to the 2nd face, i.e. nc[1]={3,1,2} the 2nd node of this tet is in opposite to the 3rd face, i.e. nc[1]={2,0,3} etc.

◆ ifaceorder

const int ifaceorder[4]

Index mapping from the i-th face defined in nc[][] to the face-neighbor (facenb) face orders.

nc[ifaceorder[i]] <-> facenb[i] nc[0], made of nodes {3,0,1}, is the face connecting to the 4th neighbor (facenb[3]), nc[1], made of nodes {3,1,2}, is the face connecting to the 1st neighbor (facenb[0]), etc.

◆ nc

const int nc[4][3] = {{3, 0, 1}, {3, 1, 2}, {2, 0, 3}, {1, 0, 2}}

Tetrahedron faces, in clock-wise orders, represented using local node indices.

node-connectivity, i.e. nc[4] points to the 4 facets of a tetrahedron, with each triangular face made of 3 nodes. The numbers [0-4] are the local node indices (starting from 0). The order of the nodes are in clock-wise orders.

◆ out

const int out[4][3]

Tetrahedron faces, in counter-clock-wise orders, represented using local node indices.

out is like nc[] but with counter-clock-wise orientation. this makes the normal vec of each face pointing outwards.defined in mmc_mesh.c, node orders for each face, in counter-clock-wise orders

node-connectivity, i.e. nc[4] points to the 4 facets of a tetrahedron, with each triangular face made of 3 nodes. The numbers [0-4] are the local node indices (starting from 0). The order of the nodes are in counter-clock-wise orders.

Macros

Functions

Variables

Detailed Description

Macro Definition Documentation

◆ F32N

◆ F32P

Function Documentation

◆ albedoweight()

◆ fixphoton()

◆ getinterp()

◆ havel_raytet()

◆ interppos()

◆ launchphoton()

◆ onephoton()

◆ plucker_raytet()

◆ reflectray()

◆ reflectrayroi()

Variable Documentation

◆ e2n

◆ facemap

◆ faceorder

◆ fc

◆ ifacemap

◆ ifaceorder

◆ nc

◆ out