Bessel beams #304

stefaniagl · 2021-11-11T07:22:18Z

Description

Added new incident beams - high order vector Bessel beams of LE, LM, CS, CS', TEM, TEL, and generalized types.

Related issues

Fixes #209, including #285.

Types of changes

Bugfix (non-breaking change which fixes an issue)
New feature (non-breaking change which adds functionality)
Breaking change (fix or feature that would cause existing functionality to not work as expected)

Checklist

I have read the contributing guidelines
Code compiles correctly
Tests pass locally with my changes
I have added tests that prove my fix is effective or that my feature works. And these tests pass
I have added/extended necessary documentation (if appropriate)

Further comments

Test version of Bessel beams implementation in ADDA

Add files via upload

Added other BB types (trial version).

Redefined a parallel polarization of BB taking into account its additional phase.

myurkin

This is a first quick review to fix some unexpected changes relative to the current master.

.gitignore

src/.gitignore

myurkin · 2021-12-05T04:40:24Z

src/GenerateB.c

@@ -31,7 +31,6 @@
 #include "interaction.h"
 #include "param.h"
 #include "vars.h"
-// system headers


this deletion should probably be reverted

src/GenerateB.c

myurkin · 2021-12-05T04:40:49Z

src/GenerateB.c

@@ -222,20 +320,23 @@ void InitBeam(void)
 }

 //======================================================================================================================
-


this deletion should probably be reverted

src/Makefile

myurkin · 2021-12-05T04:53:58Z

src/githash.h

@@ -0,0 +1 @@
+#define GITHASH "bc2963d"


This file should be removed

myurkin · 2021-12-05T04:54:52Z

src/ocl/Makefile

@@ -21,9 +21,9 @@
 # Depending on a particular OpenCL installation one may need to manually specify paths to corresponding headers and


No changes are expected in this file (with respect to the current master)

myurkin · 2021-12-05T04:57:33Z

src/const.h

-#define MAX_N_BEAM_PARMS 10   // maximum number of beam parameters
+#define MAX_N_BEAM_PARMS 13   // maximum number of beam parameters


This change should probably be reverted (due to usage of -beam_center)

myurkin · 2021-12-05T05:04:09Z

src/parbas.h

@@ -25,7 +25,7 @@
 * removed, then even MPI 2.0 will do.
 */
 #define MPI_VER_REQ 2
-#define MPI_SUBVER_REQ 1
+#define MPI_SUBVER_REQ 0


This change should probably be reverted.

- disables all code related to Bessel beams when NO_FORTRAN is used - edits a lot of help and information messages, including names of some Bessel beams types - beam center in laboratory reference frame is now specified always (even when at origin) - calculation of Cext through F(0) is now used only for default position of the plane wave - fixes processing of imaginary part of matrix M (for Bessel beams), by changing the type to doublecomplex - removes beam-center position in descriptive strings for beams (since it is printed separately in param.c) - plane wave in surface mode now works fine with -beam_center. Removes a lot of redundant code there (probably was used long ago for alternative calculation of Cext) minor/stylistic changes: - shortens define handles for Bessel beams - changes required number of subarguments for '-beam dipole' to UNDEF - moves beam_center below beam in lists of command line options (to be in alphabetical order) - renames use_beam_center and use_beam_subopt into beam_center_used and deprecated_bc_used, respectively - adds auxiliary function ScanDouble3Error() in param.c - removes redundant checks for number of arguments for most Bessel beam types - processing of deprecated specification of beam center as an argument to '-beam' is now in param.c - updates instructions for adding new beams - updates .gitignore - removes mentioning of eIncRefl[3],eIncTran[3] from crosssec.c - beam_center_0 is now defined in vars.c/h

myurkin

Please first carefully review and test the changes to the code, that I recently pushed.
Then, take care of the comments here.
We need to discuss the phase of the amplitude matrix (e.g. when -scat_grid is used)

src/GenerateB.c

myurkin · 2021-12-07T03:41:57Z

src/GenerateB.c

+		case B_BES_TML:
+			if (surface) PrintError("Currently, Bessel incident beam is not supported for '-surf'");
+			// initialize parameters
+			n0=round(beam_pars[0]);


Test the input order to be integer. See the function ConvertToInteger

myurkin · 2021-12-07T03:45:40Z

src/GenerateB.c

+			if (surface) PrintError("Currently, Bessel incident beam is not supported for '-surf'");
+			// initialize parameters
+			n0=round(beam_pars[0]);
+			alpha0=beam_pars[1];


The input angle should be in degrees (as other input angles in ADDA - this already described as such in help strings). See function Rad2Deg.
Moreover, we should probably check it to be between 0 and 90 degrees. Angle 0 should not be allowed unless the order is 0 (or some specific matrix M for orders +-1 or +-2 - I am not sure, please look carefully at the formulae). Similar issue is for angle 90.

myurkin · 2021-12-07T03:47:34Z

src/GenerateB.c

+					M[0]=beam_pars[2]+I*beam_pars[6];
+					M[1]=beam_pars[3]+I*beam_pars[7];
+					M[2]=beam_pars[4]+I*beam_pars[8];
+					M[3]=beam_pars[5]+I*beam_pars[9];


Imaginary parts are not necessarily defined in the command line. The number of beam parameters should be checked before this code to avoid undefined behavior (they are not necessarily zero).

src/GenerateB.c

…sel beams - corrects accounting for beam vorticity, when computing amplitude matrix (for scat_plane and scat_grid) - renames variables for Bessel beams to avoid confusion - slightly changes alignment in the code and descriptive string for Bessel beams - merges two cases for Bessel beams in InitBeam() into one - replaces several cexp(I*...) by imExp(...) - adds several TODO comments for further improvement - quick tests show that amplitude matrix remains invariant under 90-degree rotation for symmetric shapes (comparing

- updates ignores in `comp2exec` so that all tests pass (after recent #304) - adds -beam_center to various lines in test suites * Corrections in `GenerateB.c`: - removes redundant EOL in multi-line beam descriptions - fixes incorrect calculation of incident plane wave for grazing incidence in surface mode (when prop_z is on the order of 1e-16). For that the code was completely rewritten to specifically calculate potentially small quantities (e.g., expm1 instead of exp and rcp1=rc+1 instead of rc). - removes duplicate initialization of `prIncRefl` - mentions `vorticity` in instructions for adding new beam - in several places (when accounting for `beam_center`) `LinComb()` was replaced by `vSubtr()`. * Other changes: - adds functions `imExpM1()` and `cExpM1()` to `cmplx.h` - definitions of `prIncRelf` and `prIncTran` are now semi-global (only in `GenerateB.c` and `param.c`) - updates copyright year in `param.c` Co-authored-by: Maxim Yurkin <[email protected]>

commit 443f00a Author: Maxim Yurkin <[email protected]> Date: Tue Mar 29 13:34:30 2022 +0700 Minor changes in figure spacing commit 581bec6 Author: stefaniagl <[email protected]> Date: Tue Mar 29 00:03:24 2022 +0700 Bessel 0 (adda-team#315) Tests and examples for Bessel beams: - test using equivalent command line strings `tests/equiv/bb_equiv.py` - simple examples `examples/bessel` - scripts to reproduce 4 figures in a recent paper (submitted) Fixes a bug in TML type of Bessel beams. commit d7e6169 Author: Maxim Yurkin <[email protected]> Date: Mon Feb 21 23:35:49 2022 +0700 Makes the test for inconsistencies in the input shape format more sensitive. Detects extra number present in a line for ADDA formats (both single- and multi-domain). The formats themselves are now defines instead of static strings. commit 887a63c Author: Maxim Yurkin <[email protected]> Date: Mon Feb 21 14:48:46 2022 +0700 - All math constants are now given to 35 significant digits (was 32 in most places) - This includes the LDR constants (for which a separate calculation procedure was performed) - Improves a few related comments commit a993a91 Author: Maxim Yurkin <[email protected]> Date: Tue Feb 15 10:21:53 2022 +0700 - renames find_adda.sh into find_adda (and sets executable flag) - this script now returns ADDA path to stdout, hence does not need to be sourced (was a bad idea due to potential variables collision) - updates all calling scripts commit c8ea4ed Author: Maxim Yurkin <[email protected]> Date: Mon Jan 31 14:52:02 2022 +0700 minor formatting changes commit a6be5f4 Author: Maxim Yurkin <[email protected]> Date: Mon Jan 31 14:31:10 2022 +0700 - adds a special script to locate ADDA binaries (to be used by various examples). - updates existing examples in `examples/papers/2021_WKBr/` and `misc/near_field/` accordingly - other minor changes to `examples/papers/2021_WKBr/calc.sh` commit 2896575 Author: inzhevatkin <[email protected]> Date: Sun Jan 30 11:40:56 2022 +0700 WKBr scripts (adda-team#311) * Adds scripts to reproduce Figs.15 and 16 from paper (Inzhevatkin & Yurkin, 2022) * Includes general-purpose scripts to generate internal field in a sphere in the framework of the WKBr approximation (two modifications). * Also includes interface with Mie-theory calculations. Co-authored-by: Maxim Yurkin <[email protected]> commit 6c0b401 Merge: 49e6c23 deebdbf Author: Maxim Yurkin <[email protected]> Date: Tue Jan 11 19:11:38 2022 +0700 Merge branch 'master' of github.com:adda-team/adda commit 49e6c23 Author: Maxim Yurkin <[email protected]> Date: Tue Jan 11 19:09:51 2022 +0700 Updates extracting of gcc version in Makefile to be compatible with all possible configurations of gcc. commit deebdbf Author: stefaniagl <[email protected]> Date: Fri Jan 7 17:44:39 2022 +0700 Changes in tests (adda-team#310) - updates ignores in `comp2exec` so that all tests pass (after recent adda-team#304) - adds -beam_center to various lines in test suites * Corrections in `GenerateB.c`: - removes redundant EOL in multi-line beam descriptions - fixes incorrect calculation of incident plane wave for grazing incidence in surface mode (when prop_z is on the order of 1e-16). For that the code was completely rewritten to specifically calculate potentially small quantities (e.g., expm1 instead of exp and rcp1=rc+1 instead of rc). - removes duplicate initialization of `prIncRefl` - mentions `vorticity` in instructions for adding new beam - in several places (when accounting for `beam_center`) `LinComb()` was replaced by `vSubtr()`. * Other changes: - adds functions `imExpM1()` and `cExpM1()` to `cmplx.h` - definitions of `prIncRelf` and `prIncTran` are now semi-global (only in `GenerateB.c` and `param.c`) - updates copyright year in `param.c` Co-authored-by: Maxim Yurkin <[email protected]> commit 6a9ec2e Author: Maxim Yurkin <[email protected]> Date: Mon Jan 3 00:02:33 2022 +0700 adds ignoring of errors when getting commit hash (i.e. when compiling from downloaded source code, not from git repository)

stefaniagl and others added 30 commits June 9, 2019 00:08

Add files via upload

3372bc2

Test version of Bessel beams implementation in ADDA

Merge pull request #1 from stefanyagl/stefanyagl-patch-1

8e74216

Add files via upload

Added other BB types (not tested)

5292e62

Merge pull request #2 from stefanyagl/bessel_1

4ee2f85

Added other BB types (trial version).

Log file parameters have been changed a bit for BBs

b5e2eb0

1

0a7ff48

Examined Bessel beams

f7238e1

Added Generalized BB

51bfec3

cleaning

2958bf9

fixing Makefiles

7dffe17

Merge remote-tracking branch 'upstream/master'

8b49f8d

merging from upstream

15d3f60

Removed ASD code

bb4f787

Fixed beam description

597fd7f

Merge remote-tracking branch 'upstream/master'

21b201d

Fixed besselGen initialization

3fcc7c6

Changed besselGen -> besselM

61e931e

minor fixes

97852cd

Merge branch 'master' of https://github.com/stefaniagl/adda

1016574

Added CS' type

25b9461

Improved field expressions

88f15bc

Merge branch 'master' of https://github.com/stefaniagl/adda

44cd648

Fixed some problems

d515d92

Fixed some bugs

505dd45

Command line changes

93df21c

Added -beam center

1df9ca2

Added vortex phase + minor changes

99523db

Redefined a parallel polarization of BB taking into account its additional phase.

minor changes

899eeb6

Added negative orders of BBs

1e58292

Extracted bessel ASD

bc2963d

Delete .DS_Store

0f2143d

stefaniagl marked this pull request as ready for review November 13, 2021 06:32

minor changes

84297f0

myurkin reviewed Dec 5, 2021

View reviewed changes

stefaniagl added 10 commits December 5, 2021 16:15

minor changes

836fa5c

minor changes

9b87dd4

minor changes

3401982

minor changes

bad2a81

minor changes

736010c

minor changes

e0887a4

Delete githash.h

9d2dc6e

minor changes

b2a40b0

minor changes

bede24c

minor changes

1c4323e

myurkin self-assigned this Dec 5, 2021

myurkin requested changes Dec 7, 2021

View reviewed changes

This was referenced Dec 7, 2021

Remove deprecated specification of beam center as argument to -beam #306

Open

Remove dependence on Fortran for calculation of Bessel functions #307

Open

new cmdline argument "-beam_center" #285

Open

stefaniagl and others added 7 commits December 11, 2021 15:54

minor changes

192db9b

minor changes

f59da27

minor changes

4ea4e1a

minor changes

5791530

minor changes

de78d36

minor changes

51aac9e

myurkin merged commit 7ef46af into adda-team:master Dec 19, 2021

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Bessel beams #304

Bessel beams #304

stefaniagl commented Nov 11, 2021

myurkin left a comment

myurkin Dec 5, 2021

myurkin Dec 5, 2021

myurkin Dec 5, 2021

myurkin Dec 5, 2021

myurkin Dec 5, 2021

myurkin Dec 5, 2021

myurkin left a comment

myurkin Dec 7, 2021

myurkin Dec 7, 2021

myurkin Dec 7, 2021

		@@ -222,20 +320,23 @@ void InitBeam(void)
		}

		//======================================================================================================================

		@@ -21,9 +21,9 @@
		# Depending on a particular OpenCL installation one may need to manually specify paths to corresponding headers and

		#define MAX_N_BEAM_PARMS 10 // maximum number of beam parameters
		#define MAX_N_BEAM_PARMS 13 // maximum number of beam parameters

Bessel beams #304

Bessel beams #304

Conversation

stefaniagl commented Nov 11, 2021

Description

Related issues

Types of changes

Checklist

Further comments

myurkin left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

myurkin left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment