Homec4science
Diffusion GYACOMO (OBSOLETE: LATEST VERSION ON gitlab.epfl.ch) (3D_parallel) Browse

gyacomo/matlab/compute/compute_fa.m
7ee703feb0b93D_parallel

Last Change
Raw File

compute_fa.m
View Options

function [SS,XX,FF] = compute_fa(data, options)
%% Compute the dispersion function from the moment hierarchi decomp.
% Normalized Hermite
Hp = @(p,s) polyval(HermitePoly(p),s)./sqrt(2.^p.*factorial(p));
% Hp = @(p,s) hermiteH(p,s)./sqrt(2.^p.*factorial(p));
% Laguerre
Lj = @(j,x) polyval(LaguerrePoly(j),x);
% Maxwellian factor
FaM = @(s,x) exp(-s.^2-x);
%meshgrid for efficient evaluation
[SS, XX] = meshgrid(options.SPAR, options.XPERP);
[~, ikx0] = min(abs(data.kx));
[~, iky0] = min(abs(data.ky));
switch options.SPECIE
case 'e'
Napj_ = data.Nepj;
parray = double(data.Pe);
jarray = double(data.Je);
case 'i'
Napj_ = data.Nipj;
parray = double(data.Pi);
jarray = double(data.Ji);
end
switch options.Z
case 'avg'
Napj_ = mean(Napj_,5);
otherwise
[~,iz] = min(abs(options.Z-data.z));
Napj_ = Napj_(:,:,:,:,iz,:);
end
Napj_ = squeeze(Napj_);
frames = options.T;
for it = 1:numel(options.T)
[~,frames(it)] = min(abs(options.T(it)-data.Ts5D));
end
Napj_ = mean(Napj_(:,:,:,:,frames),5);
Napj_ = squeeze(Napj_);
Np = numel(parray); Nj = numel(jarray);
FF = zeros(data.Nkx,data.Nky,numel(options.XPERP),numel(options.SPAR));
% FF = zeros(numel(options.XPERP),numel(options.SPAR));
FAM = FaM(SS,XX);
for ip_ = 1:Np
p_ = parray(ip_);
HH = Hp(p_,SS);
for ij_ = 1:Nj
j_ = jarray(ij_);
LL = Lj(j_,XX);
% FF = FF + Napj_(ip_,ij_,ikx0,iky0)*HH.*LL.*FAM;
HLF = HH.*LL.*FAM;
for ikx = 1:data.Nkx
for iky = 1:data.Nky
FF(ikx,iky,:,:) = squeeze(FF(ikx,iky,:,:)) + Napj_(ip_,ij_,ikx,iky)*HLF;
end
end
end
end
% FF = (FF.*conj(FF)); %|f_a|^2
FF = abs(FF); %|f_a|
FF = sqrt(squeeze(mean(mean(FF,1),2))); %sqrt(<|f_a|>kx,ky)
FF = FF./max(max(FF));
FF = FF';
% FF = FF.*conj(FF);
% FF = sqrt(FF);
% FF = FF./max(max(FF));
% FF = FF';
end
c4science · Help