Mercurial > hg > ltpda

comparison m-toolbox/classes/@ssm/cpsdForIndependentInputs.m @ 0:f0afece42f48

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Import.
author Daniele Nicolodi <nicolodi@science.unitn.it>
date Wed, 23 Nov 2011 19:22:13 +0100
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:f0afece42f48
1 % CPSDFORINDEPENDENTINPUTS computes the output theoretical CPSD shape with given inputs.
2 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3 %
4 % DESCRIPTION: cpsdForIndependentInputs computes the output theoretical
5 % CPSD or PSD shape for given input shapes.
6 % It returns summed and individual contributions and takes
7 % input vectors of objects (instead of square matrices)
8 %
9 % CALL: [mat_outSum, mat_out] = PSD(sys, pl)
10 %
11 % INPUTS:
12 % sys - (array of) ssm object
13 %
14 % OUTPUTS:
15 % mat_outSum - contains specified returned aos, noise is
16 % summed over all the specified input noises
17 % mat_out - contains specified returned aos
18 %
19 % <a href="matlab:utils.helper.displayMethodInfo('ssm', 'cpsdForIndependentInputs')">Parameters Description</a>
20 %
21 % VERSION: $Id: cpsdForIndependentInputs.m,v 1.2 2011/05/23 14:18:20 adrien Exp $
22 %
23 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24
25 function varargout = cpsdForIndependentInputs(varargin)
26
27 %% starting initial checks
28
29 % use the caller is method flag
30 callerIsMethod = utils.helper.callerIsMethod;
31
32 % Check if this is a call for parameters
33 if utils.helper.isinfocall(varargin{:})
34 varargout{1} = getInfo(varargin{3});
35 return
36 end
37
38 utils.helper.msg(utils.const.msg.MNAME, ['running ', mfilename]);
39
40 % Collect input variable names
41 in_names = cell(size(varargin));
42 for ii = 1:nargin,in_names{ii} = inputname(ii);end
43
44 % Collect all SSMs and plists
45 [sys, ssm_invars, rest] = utils.helper.collect_objects(varargin(:), 'ssm', in_names);
46 [pl, invars2, rest] = utils.helper.collect_objects(rest(:), 'plist');
47 if ~isempty(rest)
48 pl = combine(pl, plist(rest{:}));
49 end
50 pl = combine(pl, getDefaultPlist());
51
52 %%% Internal call: Only one object + don't look for a plist
53 internal = strcmp(varargin{end}, 'internal');
54
55 %% begin function body
56
57 %% retrieve system infos
58
59 if numel(sys)~=1
60 error('noisespectrum needs exactly one ssm as an input')
61 end
62 if ~sys.isnumerical
63 error(['error because system ',sys.name,' is not numerical']);
64 end
65 if ~sys.isStable
66 error('input ssm is not stable!')
67 end
68 if sys.timestep==0
69 timestep = 1;
70 else
71 timestep = sys.timestep;
72 end
73 if ~internal
74 inhist = sys.hist;
75 end
76
77 %% modifying system's ordering
78 if find(pl, 'reorganize')
79 sys = reorganize(sys, pl, 'set', 'for cpsdForIndependentInputs', 'internal', 'internal');
80 end
81
82 %% collecting functions i/o data
83 aos_in = find(pl, 'aos');
84 PZ_in = find(pl, 'PZmodels');
85 cov_in = find(pl, 'variance');
86 cpsd_in = find(pl, 'PSD');
87
88 noise_mat = [...
89 reshape(cov_in, [numel(cov_in),1]) ; ...
90 reshape(cpsd_in, [numel(cpsd_in),1]) / (timestep*2)];
91 if sum(noise_mat<0) > 0
92 error('input PSD is not positive!')
93 end
94
95 %% getting system's i/o sizes
96 inputSizes = sys.inputsizes;
97 outputSizes = sys.outputsizes; %#ok<NASGU>
98
99 Naos_in = inputSizes(1);
100 Nnoise = inputSizes(2);
101 NPZmodels = inputSizes(3);
102
103 %% retrieving frequency vector
104 if isempty(Naos_in)==0
105 f1 = find(pl,'f1');
106 f2 = find(pl,'f2');
107 NFreqs = find(pl,'nf');
108 if isempty(f1) || isempty(f2)|| isempty(NFreqs)
109 error('### Please specify frequency vector a start and stop frequency .');
110 else
111 freqs = 10 .^ linspace(log10(f1), log10(f2), NFreqs);
112 end
113 else
114 freqs = aos_in(1).x;
115 end
116
117 %% checking frequency vector
118 for i=2:numel(aos_in)
119 if ~isequal(freqs,aos_in(i).x)
120 error('there exist different frequency vectors');
121 end
122 end
123
124 %% reshape pzmodels and aos for input cross-spectra
125 if numel(PZ_in)==NPZmodels
126 PZdata = zeros(NPZmodels,NFreqs);
127 for i=1:NPZmodels
128 a = resp(PZ_in(i), freqs);
129 PZdata(i,:) = reshape(a.y,[1,NFreqs]) ;
130 end
131 else
132 error('Wrong size for field PZ_in')
133 end
134
135 if numel(aos_in)==Naos_in
136 AOdata = zeros(Naos_in,NFreqs);
137 for i=1:Naos_in
138 AOdata(i,:) = reshape(aos_in(i).y,[1,NFreqs]) ;
139 end
140 else
141 error('Wrong size for field aos_in')
142 end
143
144 %% SSM Transfer function
145 [a, b, c, d, Ts, InputName, StateName, OutputName,...
146 inputvarunits, ssvarunits, outputvarunits] = double(sys); %#ok<ASGLU>
147 resps = ssm.doBode(a, b, c, d, 2*pi*freqs, Ts);
148 Noutputs = numel(OutputName);
149
150 %% power for each frequency with SVD computation
151 diagOnly = pl.find('DIAGONAL ONLY');
152 if diagOnly
153 Result = zeros(Noutputs, Nnoise+Naos_in+NPZmodels, NFreqs);
154 else
155 Result = zeros(Noutputs, Noutputs, Nnoise+Naos_in+NPZmodels, NFreqs);
156 end
157
158 for ff=1:NFreqs
159 for ii = 1:(Nnoise+Naos_in+NPZmodels)
160 powWhiteNoise = zeros(1,Nnoise);
161 powAO = zeros(1, Naos_in);
162 powPZ = zeros(1,NPZmodels);
163 if ii<Nnoise+1,
164 %% contribution from white noise
165 powWhiteNoise(ii) = noise_mat(ii) ;
166 elseif ii<Nnoise+Naos_in+1
167 %% contribution from aos
168 i_input2 = ii-Nnoise;
169 if AOdata(i_input2,ff)<0
170 error('input PSD is not positive!')
171 end
172 powAO(i_input2) = AOdata(i_input2,ff);
173 else
174 %% contribution from PZmodels
175 i_input2 = ii-Nnoise-Naos_in;
176 if PZdata(i_input2,ff)<0
177 error('input PSD is not positive!')
178 end
179 powPZ(i_input2) = real( PZdata(i_input2,ff) * conj( PZdata(i_input2,ff)) );
180 end
181 %% computing CPSD
182 pow = diag([powAO; powWhiteNoise; powPZ]);
183 RespLoc = squeeze(resps(:,:,ff));
184 noise = RespLoc * pow * RespLoc' * (2*timestep);
185 if diagOnly
186 Result(:,ii,ff) = real(diag(noise));
187 else
188 Result(:,:,ii,ff) = noise;
189 end
190
191 end
192 end
193
194 %% saving in aos
195 if diagOnly
196 ao_outSum = ao.initObjectWithSize(Noutputs, 1);
197 %% sum of all inputs
198 for oo=1:Noutputs
199 ao_outSum(oo,1).setData(fsdata(freqs, squeeze(sum(Result(oo,:,:),2)) ));
200 ao_outSum(oo,1).setName( ['PSD of ' , OutputName{oo} ' due to all contributions']);
201 ao_outSum(oo,1).setXunits('Hz');
202 ao_outSum(oo,1).setYunits(outputvarunits(oo)^2/unit('Hz'));
203 ao_outSum(oo,1).setDescription( ['PSD of ' , OutputName{oo} ' due to all contributions']);
204 end
205 if nargout ~= 1;
206 ao_out = ao.initObjectWithSize(Noutputs, Nnoise+Naos_in+NPZmodels);
207 %% individual inputs
208 for oo=1:Noutputs
209 for ii=1:(Nnoise+Naos_in+NPZmodels)
210 ao_out(oo,ii).setData(fsdata(freqs, squeeze(Result(oo,ii,:)) ));
211 ao_out(oo,ii).setName( ['PSD of ' , OutputName{oo} ' due to ' InputName{ii}]);
212 ao_out(oo,ii).setXunits('Hz');
213 ao_out(oo,ii).setYunits(outputvarunits(oo)^2/unit('Hz'));
214 ao_out(oo,ii).setDescription( ['PSD of ' , OutputName{oo} ' due to ' InputName{ii}]);
215 end
216 end
217 end
218 else
219 ao_outSum = ao.initObjectWithSize(Noutputs, Noutputs);
220 %% sum of all inputs
221 for oo=1:Noutputs
222 for pp=1:Noutputs
223 ao_outSum(oo,pp).setData(fsdata(freqs, squeeze(sum(Result(oo,pp,:,:),3)) ));
224 ao_outSum(oo,pp).setXunits('Hz');
225 ao_outSum(oo,pp).setYunits(outputvarunits(oo)^2/unit('Hz'));
226 if oo~=pp
227 ao_outSum(oo,pp).setName( ['Cross PSD of ', OutputName{oo}, ' and ', OutputName{pp} ' due to all contributions']);
228 ao_outSum(oo,pp).setDescription( ['Cross PSD of ', OutputName{oo}, ' and ', OutputName{pp} ' due to all contributions']);
229 else
230 ao_outSum(oo,pp).setName( ['PSD of ' , OutputName{oo}]);
231 ao_outSum(oo,pp).setDescription( ['PSD of ' , OutputName{oo}]);
232 end
233 end
234 end
235 if nargout ~= 1;
236 ao_out = ao.initObjectWithSize(Noutputs, Noutputs, Nnoise+Naos_in+NPZmodels);
237 %% individual inputs
238 for oo=1:Noutputs
239 for pp=1:Noutputs
240 for ii=1:(Nnoise+Naos_in+NPZmodels)
241 ao_out(oo,pp,ii).setData(fsdata(freqs, squeeze(Result(oo,pp,ii,:)) ));
242 ao_out(oo,pp,ii).setXunits('Hz');
243 ao_out(oo,pp,ii).setYunits(outputvarunits(oo)^2/unit('Hz'));
244 if oo~=pp
245 ao_out(oo,pp,ii).setName( ['Cross PSD of ', OutputName{oo}, ' and ', OutputName{pp} ' due to ' InputName{ii}]);
246 ao_out(oo,pp,ii).setDescription( ['Cross PSD of ', OutputName{oo}, ' and ', OutputName{pp} ' due to ' InputName{ii}]);
247 else
248 ao_out(oo,pp,ii).setName( ['PSD of ' , OutputName{oo} ' due to ' InputName{ii}]);
249 ao_out(oo,pp,ii).setDescription( ['PSD of ' , OutputName{oo} ' due to ' InputName{ii}]);
250 end
251 end
252 end
253 end
254 end
255 end
256
257
258
259 %% construct output matrix object
260 if nargout ~= 1;
261 out = matrix(ao_out);
262 end
263 outSum = matrix(ao_outSum);
264 if callerIsMethod
265 % do nothing
266 else
267 myinfo = getInfo('None');
268 if nargout ~= 1;
269 out.addHistory(myinfo, pl , ssm_invars(1), inhist );
270 end
271 outSum.addHistory(myinfo, pl , ssm_invars(1), inhist );
272 end
273
274 %% Set output depending on nargout
275 if nargout == 1;
276 varargout = {outSum};
277 elseif nargout == 2;
278 varargout = {outSum out };
279 elseif nargout == 0;
280 iplot(ao_outSum, ao_out);
281 else
282 error('Wrong number of outputs')
283 end
284 end
285
286
287
288
289 %--------------------------------------------------------------------------
290 % Get Info Object
291 %--------------------------------------------------------------------------
292 function ii = getInfo(varargin)
293
294 if nargin == 1 && strcmpi(varargin{1}, 'None')
295 sets = {};
296 pl = [];
297 else
298 sets = {'Default'};
299 pl = getDefaultPlist;
300 end
301 % Build info object
302 ii = minfo(mfilename, 'ssm', 'ltpda', utils.const.categories.op, '$Id: cpsdForIndependentInputs.m,v 1.2 2011/05/23 14:18:20 adrien Exp $', sets, pl);
303
304 end
305
306 %--------------------------------------------------------------------------
307 % Get Default Plist
308 %--------------------------------------------------------------------------
309 function pl = getDefaultPlist()
310 pl = ssm.getInfo('reorganize', 'for cpsdForIndependentInputs').plists;
311 pl.remove('set');
312
313 p = param({'variance', 'The variance vector of this noise between input ports for the <i>time-discrete</i> noise model. '}, []);
314 pl.append(p);
315
316 p = param({'PSD', 'The one sided psd vector of the white noise between input ports.'}, []);
317 pl.append(p);
318
319 p = param({'aos', 'A vector of input PSD AOs, The spectrum of this noise between input ports for the <i>time-continuous</i> noise model.'}, ao.initObjectWithSize(1,0));
320 pl.append(p);
321
322 p = param({'PZmodels', 'vector of noise shape filters for the different corresponding inputs.'}, paramValue.DOUBLE_VALUE(zeros(0,1)));
323 pl.append(p);
324
325 p = param({'reorganize', 'When set to 0, this means the ssm does not need be modified to match the requested i/o. Faster but dangerous!'}, paramValue.TRUE_FALSE);
326 pl.append(p);
327
328 p = param({'f2', 'The maximum frequency. Default is Nyquist or 1Hz.'}, paramValue.EMPTY_DOUBLE);
329 pl.append(p);
330
331 p = param({'f1', 'The minimum frequency. Default is f2*1e-5.'}, paramValue.EMPTY_DOUBLE);
332 pl.append(p);
333
334 p = param({'nf', 'The number of frequency bins. Frequencies are scale logarithmically'}, paramValue.DOUBLE_VALUE(200));
335 pl.append(p);
336
337 p = param({'diagonal only', 'Set to true if you want the PSD instead of the CPSD'}, paramValue.TRUE_FALSE);
338 pl.append(p);
339
340 end
341